Digital Signal Averager and Operating Software for Windows 2000 Professional and XP Professional SP 2. Hardware/Software User s Manual

Transcription

1 69;,* Digital Signal Averager and Operating Software for Windows 2000 Professional and XP Professional SP 2 Hardware/Software User s Manual Software Version 1.0 Installation begins on Page 7 Printed in U.S.A. ORTEC Part No Manual Revision P1

2 $GYDQ HG0HDVXUHPHQW7H KQRORJ\,Q a/k/a/ ORTEC, a subsidiary of AMETEK, Inc. WARRANTY ORTEC* warrants that the items will be delivered free from defects in material or workmanship. ORTEC makes no other warranties, express or implied, and specifically NO WARRANTY OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. ORTEC s exclusive liability is limited to repairing or replacing at ORTEC s option, items found by ORTEC to be defective in workmanship or materials within one year from the date of delivery. ORTEC s liability on any claim of any kind, including negligence, loss, or damages arising out of, connected with, or from the performance or breach thereof, or from the manufacture, sale, delivery, resale, repair, or use of any item or services covered by this agreement or purchase order, shall in no case exceed the price allocable to the item or service furnished or any part thereof that gives rise to the claim. In the event ORTEC fails to manufacture or deliver items called for in this agreement or purchase order, ORTEC s exclusive liability and buyer s exclusive remedy shall be release of the buyer from the obligation to pay the purchase price. In no event shall ORTEC be liable for special or consequential damages. Quality Control Before being approved for shipment, each ORTEC instrument must pass a stringent set of quality control tests designed to expose any flaws in materials or workmanship. Permanent records of these tests are maintained for use in warranty repair and as a source of statistical information for design improvements. Repair Service If it becomes necessary to return this instrument for repair, it is essential that Customer Services be contacted in advance of its return so that a Return Authorization Number can be assigned to the unit. Also, ORTEC must be informed, either in writing, by telephone [(865) ] or by facsimile transmission [(865) ], of the nature of the fault of the instrument being returned and of the model, serial, and revision ("Rev" on rear panel) numbers. Failure to do so may cause unnecessary delays in getting the unit repaired. The ORTEC standard procedure requires that instruments returned for repair pass the same quality control tests that are used for new-production instruments. Instruments that are returned should be packed so that they will withstand normal transit handling and must be shipped PREPAID via Air Parcel Post or United Parcel Service to the designated ORTEC repair center. The address label and the package should include the Return Authorization Number assigned. Instruments being returned that are damaged in transit due to inadequate packing will be repaired at the sender's expense, and it will be the sender's responsibility to make claim with the shipper. Instruments not in warranty should follow the same procedure and ORTEC will provide a quotation. Damage in Transit Shipments should be examined immediately upon receipt for evidence of external or concealed damage. The carrier making delivery should be notified immediately of any such damage, since the carrier is normally liable for damage in shipment. Packing materials, waybills, and other such documentation should be preserved in order to establish claims. After such notification to the carrier, please notify ORTEC of the circumstances so that assistance can be provided in making damage claims and in providing replacement equipment, if necessary. Copyright 2005, Advanced Measurement Technology, Inc. All rights reserved. *ORTEC is a registered trademark of Advanced Measurement Technology, Inc. All other trademarks used herein are the property of their respective owners.

3 TABLE OF CONTENTS SAFETY INSTRUCTIONS AND SYMBOLS... vii 1. INTRODUCTION User Manual Organization A Brief Description of the Application INSTALLATION, CONFIGURATION, AND INITIAL STARTUP PC Requirements Software Installation Hardware Installation Connecting the FASTFLIGHT-2 Hardware to AC Power and the PC Connecting to the Preamplifier and Detector Connecting to the TOF-MS Adjusting Pulse Heights Using the Over- and Under-Range LEDs Using the RAPID PROTOCOL Port FASTFLIGHT-2 Duet Operation for a Greater Dynamic Range FASTFLIGHT-2 Triggers the TOF-MS The TOF-MS Triggers the FASTFLIGHT-2 Units Initial Operating Suggestions TOF-Only Mode The Chromatograph/Trend Mode HOW FASTFLIGHT-2 WORKS The Basic Architecture The Acquisition Control Module and the USB 2.0 Interface The 10-MHz Reference and the 2-GHz Sampling Clock The Trigger Circuits The RAPID PROTOCOL Port Additional Information on Rapid Protocol Selection When FIFO Memories are Full Estimating the Maximum Possible Data Rate The FASTFLIGHT-2 Precision Enhancer Compressed Data Formats Lossless Data Compression Lossy (Peak-Preserving and Background-Suppressing) Compression Stick Diagram Spectra: Only Centroids and Area Noise and Ion Counting Statistics Random vs. Correlated Noise Methods to Suppress Correlated Noise Ion Counting Statistics and Detector Gain Statistics iii

4 FASTFLIGHT 2 Digital Signal Averager The Systematic Error Due to the Sampling Interval How the Centroid Error Transforms from the Time to the m/z Domain Centroid and Gross or Net Area Readout FASTFLIGHT-2 STARTUP AND SCREEN FEATURES Spectrum Display Modes Single-Instrument and Duet Modes Calibrated vs. Uncalibrated TOF Spectra Displaying Total-Ion and Specific-Ion Chromatograph Data Effect of Data Compression Options on Spectrum Display Docking and Undocking the Tool Bars Undocking THE TOOLBAR The Marker Tool Moving the Marker with the Mouse Marking a Peak with the Mouse The Zoom Tool MENU COMMANDS File New Open... and Save Changing Drive and Pathname FASTFLIGHT-2 File Types Save TOF As Print, Print Preview, and Print Setup Exit Edit View Acquisition Start and Stop Instrument Properties General Settings Tab Protocol Settings Tab Window Help THE RIGHT-MOUSE-BUTTON MENUS Calibration Enabled View Protocol iv

5 TABLE OF CONTENTS 7.3. Sum/Average TOFs Zoom Tool Active/Marker Tool Active Add Overlay Remove Overlay Export data to file Graph Properties Overlay properties Experiment Properties Show Total Ion Count and Show Specific Ion Count THE UTILITY BAR Acquisition Tab Calibration Tab Introduction Creating a Calibration File Editing a Calibration File Updating an Existing Calibration Introduction Performing the Update Opening a Different Calibration File C/Trend Info Tab TOF Info Tab APPENDIX A. ACTIVEX METHODS AND PROPERTIES A.1. The Instrument Operation ActiveX DLL (FF2Ctrl.dll) A.1.1. FF2CtrlObj Properties A.1.2. FF2CtrlObj Methods A.1.3. GSObj Properties A.1.4. ProtocolObj Properties A.1.5. Establishing Different Default Values in FF2.ini for the General and Protocol Settings A.1.6. TOFObj Properties A.2. ActiveX Graphing Control (GSX) A.2.1. Properties A.2.2. Methods A.2.3. ActiveX Events Sent to Parent APPENDIX B. FILE FORMATS B.1..FLT2 File Format (TOF Mode) B.1.1. The General Settings Format (TOF or Chromatograph/Trend Mode) B.1.2. Format for the Protocol Settings (TOF or Chromatograph/Trend Mode) v

6 FASTFLIGHT 2 Digital Signal Averager B.2..FFC2 File Format (Chromatograph/Trend Mode) B.3. The.FFT2 File Format (Chromatograph/Trend Mode) B.4. The Raw Data Format (TOF and Chromatograph/Trend Modes) B.4.1. Code Words Embedded in the Raw Data B.4.2. Spectral Data Encoding B Example: Mixed 16-Bit and 24-Bit Encoding) B.4.3. Decoding Spectral Data B.4.4. Lossy (Peak-Preserving and Background-Decimating) Data Compression B.4.5. Stick-Diagram Encoding B.4.6. Decoding the Entire Data Stream APPENDIX C. SPECIFICATIONS C.1. FASTFLIGHT-2 Hardware C.1.1. Performance C.1.2. Hardware Controls and Indicators C.1.3. Inputs and Outputs C.1.4. Electrical and Mechanical C.2. FASTFLIGHT-2 Software C.2.1. Architecture C.2.2. Standard Application Software C Hardware and Operating System Requirements APPENDIX D. CHANGING THE PREAMPLIFIER COARSE GAIN vi

7 SAFETY INSTRUCTIONS AND SYMBOLS This manual contains up to three levels of safety instructions that must be observed in order to avoid personal injury and/or damage to equipment or other property. These are: DANGER WARNING CAUTION Indicates a hazard that could result in death or serious bodily harm if the safety instruction is not observed. Indicates a hazard that could result in bodily harm if the safety instruction is not observed. Indicates a hazard that could result in property damage if the safety instruction is not observed. Please read all safety instructions carefully and make sure you understand them fully before attempting to use this product. In addition, the following symbol may appear on the product: ATTENTION Refer to Manual DANGER High Voltage Please read all safety instructions carefully and make sure you understand them fully before attempting to use this product. vii

8 NOTE We assume in this manual that you have working knowledge of 32-bit Windows function and terminology, and that you have read your Windows documentation. The convention in this manual for representing keys is to enclose the key label in angle brackets; for example, <F1>. For key combinations, the key labels are joined by a + within the angle brackets; for example, <Alt + 2>. The names of the button icons on the various toolbars and dialogs are printed in boldface; for example, OK, Cancel, Open, Save. viii

9 ix

10 x

11 1. INTRODUCTION FASTFLIGHT 2 is a second-generation digital signal averager (DSA) optimized for rapid acquisition of time-of-flight and chromatograph spectra in time-of-flight mass spectrometry (TOF-MS). Its salient features are: Signal sampling intervals down to 250 ps. Up to 1.5 million points per spectrum, spanning flight times up to 3 ms. TOF data acquisition rates up to 100 spectra per second in the Chromatograph/Trend mode (for a 50-µs spectrum length and a 500-ps sampling interval). 8-bit ADC provides real time sampling at 2 GHz, and interleaved sampling at 4 GHz. 1 Precision Enhancer provides 12-bits effective sampling on >256 records/spectrum. 2 Rapid hardware averaging up to 65,535 records/spectrum with negligible dead time (0.8-µs end-of-record and end-of-spectrum dead time). Automatic correlated noise subtraction improves detection limits by a factor of 3 to 10. Hardware data compression: Lossless (up to 2:1 compression) Peak preserving and background rejecting 3 (typical compressions from 10:1 to 40:1) Centroid-and-area-only spectra (stick-diagrams with typical compressions from 50:1 to 400:1) Total-ion and single-ion chromatographs. External trigger input or internal trigger output for synchronization with the TOF-MS. Duet operation for a 10:1 improvement in detection limits. RAPID PROTOCOL port for switching acquisition conditions within 10 µs. Interface to the supporting PC with a convenient USB 2.0 cable. Every TOF spectrum includes an acquisition time stamp and sequential spectrum number. Allows the display of one (normal mode) or two (duet mode) FASTFLIGHT-2 spectrum windows, plus additional windows for displaying spectrum files from disk. The FF2 user interface application program, plus a programmer s toolkit with Microsoft ActiveX controls. The supplied FF2 application program allows you to operate one or two FAST-FLIGHT 2 units connected directly to a host PC. However, a knowledgeable programmer can use the application s Microsoft ActiveX components as part of a customized application that communicates across a network. 1 U.S. Patent 6,094, U.S. Patent 6,028, U.S. Patent 5,995,989. 1

12 FASTFLIGHT 2 Digital Signal Averager 1.1. User Manual Organization The remainder of this chapter presents an application description. Chapter 2 provides instructions on installing the software and hardware, including system connections for single-unit and duet modes, hardware startup provisos, adjustment of the overrange and underrange indicators, and initial operating suggestions for the chromatograph/trend and time-of-flight-only modes. Chapter 3 discusses the architecture of the FASTFLIGHT-2 hardware, including the trigger circuits, RAPID PROTOCOL port, and Precision Enhancer; available data compression modes; and a discussion of noise, the FASTFLIGHT-2 s automatic correlated noise subtraction capability, and ion counting statistics. Chapters 4 through 8 describe the FF2 user interface and software functionality. The appendices provide the FASTFLIGHT-2 file formats, ActiveX methods and properties, specifications, and instructions on setting the coarse-gain setting on the Model 9326 Preamplifier input A Brief Description of the Application The application of FASTFLIGHT-2 to a TOF-MS is illustrated in Fig. 1. Simply stated, the system functions as follows: The sample to be analyzed is injected into the chromatograph. In the chromatograph, the various components of the sample are separated by differences in their rates of migration through the column. Factors affecting the rates of elution may include differences in molecular weights and differences in chemical or physical properties. The various components from the sample reach the output of the chromatograph at different times. The conventional chromatograph spectrum is recorded by plotting the current from the output detector vs. elution (or retention) time. Each eluting component typically produces a peak in the spectrum. These peaks normally have widths on the order of several seconds. The duration of a chromatograph separation usually spans a retention time of minutes. The function of the chromatograph is to separate the sample into its various components. But, further information is needed to identify the molecular species in each component. The TOF-MS performs the identification by measuring the mass-to-charge ratio (m/z) for the molecules in each chromatograph peak. 2

13 1. INTRODUCTION Fig. 1. The FASTFLIGHT-2 Digital Signal Averager applied to data acquisition with an electrospray TOF-MS receiving the output of a chromatograph. Although the details of the source are not shown in Fig. 1, the output of the chromatograph is fed to an electrospray source at the input to the TOF-MS. The task of the electrospray source is to ionize the molecules, separate them from the carrier liquid (or gas), and inject the charged molecules into the acceleration region of the TOF-MS. As the ionized molecules flow between the grounded grid and the accelerator plate in the TOF-MS, they are periodically accelerated toward the detector at the opposite end of the fieldfree drift tube. The acceleration is accomplished by applying a voltage pulse of a few microseconds duration to the plate. Molecules with a lower m/z value accelerate to a higher velocity than those with a higher m/z. Consequently, lower-m/z molecules reach the detector more quickly than higher-m/z molecules. Ignoring the minor imperfections in the ion optics, the time-of-flight through the drift tube to the detector is essentially proportional to m/z. When a packet of charged molecules (having a particular m/z ratio) arrives at the detector, it causes an 3

14 FASTFLIGHT 2 Digital Signal Averager output pulse whose amplitude is approximately proportional to the number of molecules in the packet. The centroid of this pulse defines the flight time of the packet, and designates the m/z value for the molecules in the packet. Typical flight times range from a few microseconds for the lowest m/z value to hundreds of microseconds for the highest. The ORTEC Model 9326 Fast Preamplifier amplifies the detector signal to between 0 and 0.5 V, a suitable input range for FASTFLIGHT-2. FASTFLIGHT-2 then samples the waveform from the preamplifier to record a digital representation of the analog output of the detector over the period from 0 to 100 µs after the trigger pulse. This process is repeated for a large number of trigger pulses, and the resulting records are summed to form a spectrum with an improved signal-to-noise ratio. This averaging process is necessary to reduce the fluctuations caused by ion statistics from the source and the statistical gain variations in the detector. The TOF spectrum formed by summing a large number of records is displayed via the personal computer. Calibrating the spectrum s horizontal axis in terms of the m/z of the molecules generates the mass spectrum for the molecules. From the latter spectrum the molecular species present in the chromatograph peak can be identified. FASTFLIGHT-2 can provide records and spectra with voltage samples at 0.25-ns, 0.5-ns, 1-ns, or 2-ns intervals. Normally, the sampling interval is chosen so that at least four samples are taken across the FWHM (full width at half-maximum amplitude) of the narrowest peak in the time-of-flight spectrum. To achieve sampling resolution as fine as 0.25 ns, the acceleration pulse in the TOF-MS must be synchronized with both the sampling clock and the delay steps in the FASTFLIGHT-2. Initiating the acceleration pulse in the TOF-MS with the Trigger Output from the FASTFLIGHT-2 fulfills this synchronization requirement. For mass spectrometers that cannot accept a Trigger Input, a trigger from the Acceleration Pulser can be accepted by the FASTFLIGHT-2 (see the optional connection indicated by the dashed line in Fig. 1). In this case, a hardware 0.25-ns sampling interval is not available, but a software 0.25-ns interpolation can be selected. FASTFLIGHT-2 normally produces the next trigger pulse 0.8 µs after the end of each scan through the range of flight times. For a typical 100-µs record length, this end-of-scan dead time contributes less than 1% idle time for data acquisition. If the TOF-MS is not ready to accept a trigger within 0.8 µs of completing the last scan, it can delay the next trigger pulse an arbitrary amount by holding the Trigger Enable Input disabled on the FASTFLIGHT-2. When the Trigger Enable Input is released, the Trigger Output will be generated promptly in synchronization with the internal clock. 4

15 1. INTRODUCTION FASTFLIGHT-2 can be operated either in the TOF mode or the Chromatograph/Trend mode. In the TOF mode, the DSA collects sequential TOF-MS spectra and displays them, but does not save them on disk. As each new spectrum is acquired, it replaces the previous spectrum in the display. A TOF-MS spectrum can be saved by stopping data acquisition and saving the displayed spectrum to disk. One means of stopping data acquisition after one TOF-MS spectrum has been acquired is to set the preset number of spectra to 1. In the Chromatograph/Trend mode, the TOF-MS spectra are rapidly saved to hard disk as they are acquired. By choosing the right presets (e.g., 0.5-ns sampling interval, 49-µs record length, 200 records/spectrum), FASTFLIGHT-2 can sample the chromatograph with a complete TOF-MS spectrum every 10 ms and do this continuously for at least 30 minutes. 4 That s a rate of 100 spectra per second! To correlate the TOF-MS spectra with the chromatograph peaks, FASTFLIGHT-2 generates its own chromatograph display from the total ion count in each TOF-MS spectrum. The total ion count is simply the net area above background for all the peaks in the TOF-MS spectrum. Any point in the chromatograph can be selected, and the corresponding TOF-MS spectrum for that point will pop up in the panel below the chromatograph. A chromatograph for a specific ion can also be acquired simultaneously by selecting the specific peak in the TOF-MS spectrum before acquisition commences. After the total chromatograph has been acquired, regions can be marked on selected peaks in the chromatograph, and all of the TOF-MS spectra for that region can be summed to improve the statistical definition in the corresponding m/z spectrum. The software also provides for measuring the net areas of peaks above background, and finding the peak centroids. 4 Requires a hard disk capable of sustained writing speeds >20 MB/s. 5

16 FASTFLIGHT 2 Digital Signal Averager 6

17 2. INSTALLATION, CONFIGURATION, AND INITIAL STARTUP 2.1. PC Requirements FASTFLIGHT-2 requires a PC that operates under Microsoft Windows 2000 Professional or XP Professional SP 2 or later, and meets or exceeds the following specifications. Using a PC with higher speed and capacity may improve performance. >2.0-GHz microprocessor 512MB SDRAM at 400MHz At least one USB 2.0 port that can be dedicated to the FASTFLIGHT-2 5 For the chromatograph mode, a hard drive with at least 20 GB of free space and the 20-MB/s sustained writing speed necessary to support 100-spectra/s acquisition rates. 6 CD drive VGA display NOTE The FASTFLIGHT-2 s user interface application program is designed to operate correctly only for users with full Administrator privileges. Limiting user privileges could cause unexpected results Software Installation NOTE Install the FASTFLIGHT-2 software before connecting the instrument to the PC. Insert the CD accompanying the FASTFLIGHT-2 into the CD drive. If installation does not start automatically, go to the Windows Taskbar and select Start, Run. In the Run dialog, browse to the CD drive, select Setup.exe, and click on OK; then click on OK again to start the installation wizard. Follow the wizard prompts, then restart the PC if instructed to do so Hardware Installation Connecting the FASTFLIGHT-2 Hardware to AC Power and the PC Position the FASTFLIGHT-2 hardware box at a convenient location between the detector output of the TOF-MS (time-of-flight mass spectrometer) and the supporting PC. 5 FASTFLIGHT 2 will not function with the slower USB 1.0 or 1.1 connections. 6 A 7200-rpm UDMA-150 hard disk is a minimum criterion for achieving the 20-MB/s sustained writing speed, but be sure to test the hard disk to determine its actual capability, according to the instructions in Section

18 FASTFLIGHT 2 Digital Signal Averager The FASTFLIGHT-2 power module accepts ac power at V ac at Hz and converts it to +15 V dc. Connect the ac input of this power module to the appropriate ac power outlet using the supplied cord. To avoid ground-loop noise, this source of ac power should have the same ground as is supplied to other powered items that share ground with the TOF-MS. Connect the dc output of the power module to the POWER IN +15 VDC connector in the lower left corner of the FASTFLIGHT-2 rear panel, using the cable provided with the power module. Turn the power switch above the POWER IN connector to the ON condition and verify that the POWER LED on the front panel lights up. Using the USB 2.0 cable supplied, connect the COMPUTER USB-2 connector at the right side of the FASTFLIGHT-2 rear panel to the USB 2.0 port on the supporting PC. When the FASTFLIGHT-2 is connected, Windows will indicate that new hardware has been connected and it should find the appropriate driver already available for automatic installation. If the computer fails to find the driver, insert the CD and click on OK. Once this step is complete, proceed to the instructions for connecting the preamplifier Connecting to the Preamplifier and Detector The ORTEC Model 9326 Fast Preamplifier (Fig. 2 ) is optimized for use in the TOF-MS application with the FASTFLIGHT-2. It should be located as close to the microchannel plate detector as possible. Before connecting the preamplifier to the output of the microchannel plate detector, ensure that the center conductor of the BNC providing the detector output is isolated from the high voltage that will be applied to the detector. Typically, this is achieved by ac-coupling the microchannel plate anode to the detector output connector. Fig. 2. The Model 9326 Fast Preamplifier. If the detector has dc high voltage on its output, a high-voltage isolation capacitor will need to be installed in series with the detector output to prevent damage to the preamplifier input. Turn off the high voltage to the microchannel plate detector and discharge the output to ground, using either a 50- terminator or a direct short to ground. If the connector already incorporates an internal resistor (1,000 10,000 ) to provide this protective discharge function, the external grounding will not be necessary. 8

19 2. INSTALLATION, CONFIGURATION, AND INITIAL STARTUP Connect the detector output to the Model 9326 preamplifier INPUT using a short length of RG-58A/U 50- coaxial cable with BNC connectors. Avoid the use of RG-58/U cable because the impedance of RG-58/U will cause reflections due to an impedance mismatch. It is best to keep the cable length short (<30 cm) to provide a low-resistance ground path between the detector and preamplifier grounds. This will minimize the interference from ground-loop noise. To avoid damaging the preamplifier input, always turn off the detector high voltage and ensure that the voltage on the center pin of the detector connector is zero before connecting to the preamplifier input. When turning on the detector bias voltage, always increase the voltage slowly to avoid damaging the preamplifier input. Keep the rate of increase in bias voltage much less than: û V û t 1 25 C volts / second (1) where C is the value of the isolation capacitor in farads. If a connector adapter is required, keep the cable between the detector and preamplifier as short as possible to minimize the effects of cable reflections. To avoid ground-loop noise, it may be desirable to isolate the detector ground from the TOF-MS ground with a modest resistance ( ). This ensures that the preamplifier will establish the ground reference for the detector signal, eliminating a ground loop around the circle from the TOF-MS to the detector, preamplifier, FASTFLIGHT-2, and back to the TOF-MS. The FASTFLIGHT-2 should be firmly grounded to the TOF-MS ground through the ac power cord ground and the TRIGGER connection. Connect the PREAMP POWER output on the FASTFLIGHT-2 rear panel to the +12 V/100 ma POWER input on the Model A cable with 9-pin male and female connectors is provided with the Model 9326 for this purpose. Connect the preamplifier OUTPUT to the ANALOG IN connector on the rear panel of the FASTFLIGHT-2 using the appropriate length RG-58A/U 50- coaxial cable with matching BNC connectors. (The length of this cable is not critical, but it should not be longer than necessary.) Avoid the use of RG-58/U cable, because the impedance of RG-58/U will cause reflections due to an impedance mismatch. Such reflections will distort the shape of peaks in the TOF spectrum. Later on, after acquiring some spectra, you may decide you wish to change the preamplifier gain to a more suitable value. See Appendix D or the Model 9326 hardware manual for instructions on changing the coarse-gain jumper among the 5, 10 and 20 V/V options. 9

20 FASTFLIGHT 2 Digital Signal Averager Once the preamplifier is connected, proceed to the next installation step Connecting to the TOF-MS Figure 3 shows a simplified schematic of the triggering connections between the FASTFLIGHT-2 and the time-of-flight mass spectrometer (TOF-MS). If your TOF-MS has provisions for accepting a trigger input to initiate the acceleration of the ionized molecules down the flight tube, then you should connect the TRIGGER OUT of the FASTFLIGHT-2 to the trigger input of the TOF-MS. This will deliver the minimum peak width in the time-of-flight (TOF) spectrum, and will enable you to use 250-ps sampling intervals for the best definition of the peaks. The TRIGGER OUT delivers a TTL pulse whose leading edge (low to high transition) is synchronized with the start of sampling in the FASTFLIGHT-2. The trailing edge is also synchronized with the scan and can be used to trigger the TOF-MS. See the specifications in Appendix C for details. Table 1 defines the voltage standard for TTL logic signals. Table 1. The TTL Logic Pulse Standard. Output (must deliver) Input (must respond to) Logic to +5 V +2 to +5 V Logic 0 0 to +0.4 V 0 to +0.8 V If your TOF-MS does not accept a trigger input but can supply a trigger output to the FASTFLIGHT-2, you can connect the TOF-MS trigger output to the TRIGGER IN on the rear panel of the FASTFLIGHT-2. The signal on this input should be a logic pulse that does not exceed ±5 V. Because the input impedance is 50, an RG-58A/U coaxial cable with BNC connectors should be used for this connection. There is a comparator on this input that will need to be set half way between the low and high voltages of the trigger signal supplied by the TOF- MS. The slope polarity will also have to be selected to match the slope of the leading edge of the trigger pulse. See the specifications for details. When the external TRIGGER IN is employed, the 250-ps sampling interval is not available; the shortest hardware sampling interval in this case is 500 ps. Either the TRIGGER OUT (preferable) or the TRIGGER IN can be used. Do not make both connections. Once the above installation steps are completed, you can turn on the detector high voltage. 10

21 2. INSTALLATION, CONFIGURATION, AND INITIAL STARTUP Fig. 3. A Simplified Representation of the TOF-MS and the Connections with FASTFLIGHT-2. Use the TRIGGER OUT (preferable) or the TRIGGER IN, but not both. CAUTION Do not turn on the detector bias voltage abruptly (several kv). Always increase the bias voltage slowly. See the Model 9326 hardware manual for preamplifier input protection limits. Start the FASTFLIGHT-2 s accompanying user interface application program as described at the beginning of Chapter 4. When you start the FF2 user interface application program, the FASTFLIGHT-2 unit will communicate with the software, and a spectrum window will automatically open; see Fig. 17 on page 51. The next step is to adjust the data acquisition settings using the Instrument Properties command discussed in Section The default settings might be adequate at the beginning, but eventually, you will want to adjust the operating parameters as explained in the rest of this manual. 11

22 FASTFLIGHT 2 Digital Signal Averager Once you have specified the measurement settings, you are ready to begin acquiring data with the FASTFLIGHT-2. Section offers some initial startup suggestions Adjusting Pulse Heights Using the Over- and Under-Range LEDs This adjustment requires the use of the user interface application program. Detailed instructions on the program features and function begin in Chapter Start the FASTFLIGHT-2 s accompanying user interface application program as described at the beginning of Chapter 4. The software will detect the attached one or two FASTFLIGHT-2 units and will open a spectrum window for each detected instrument. 2. Select the TOF mode from the Acquisition menu. 3. Go to the Toolbar and click on the Instrument Properties button ( ). This will open the Measurement Settings dialog, which contains the data acquisition properties. Once the acquisition settings have been selected, use the Start ( ) and Stop ( ) buttons to control data collection. 4. With the ion source turned off, start acquisition and observe the UNDER RANGE LED on the front panel of the FASTFLIGHT-2. Click on the Instrument Properties button, then on the Protocol Settings tab, and adjust the vertical offset until the LED turns on. To make this adjustment, enter a value in the Vertical offset field then click on the Apply button, and repeat until the LED responds. 6. Readjust the Vertical offset until the LED just turns off. This sets the dc level of the input near the lower limit of the sampling ADC range. Note that the vertical offset describes the dc voltage added to the signal at the input to the FASTFLIGHT-2. The input pulses have a negative polarity. However, that polarity is inverted by the software so that the peaks protrude in a positive direction in the displayed spectrum. Therefore, a more negative vertical offset moves the baseline of the displayed spectrum in a more positive direction. 7. Now turn on the source of ions in the TOF-MS. While observing the front-panel OVER RANGE LED on the FASTFLIGHT-2, increase the voltage on the microchannel plate detector until the LED turns on. Reduce the voltage on the detector until the LED just turns off. This sets the gain so that the range of pulse heights spans the digitizing range of the flash ADC. To optimize the bias voltage on the microchannel plate detector, you might find it desirable to change the gain of the preamplifier; see Appendix D for instructions. 12

23 2. INSTALLATION, CONFIGURATION, AND INITIAL STARTUP Using the RAPID PROTOCOL Port Most applications will not require use of the RAPID PROTOCOL port. This feature is employed when the operating parameters of the mass spectrometer must be changed in a few tens of microseconds between the acquisitions of successive spectra. The corresponding operating parameters for the FASTFLIGHT-2 are predetermined and stored in memory under their respective protocol number. When the conditions must be switched, only the protocol number needs to be communicated to the FASTFLIGHT-2. That causes a change to the conditions specified by the protocol number in a time as short as 10 µs. Employing the RAPID PROTOCOL port requires an experienced digital electronics engineer to design the handshaking interface between FASTFLIGHT-2 and the TOF-MS, and the circuits necessary to support rapid switching of operating parameters in the TOF-MS. For details, consult the signal definitions in Appendix C. See also Section The RAPID PROTOCOL connector includes 4 bits to select the protocol number and 4 tag bits. The tag bits are captured by the first trigger pulse in each spectrum, and can be used to identify conditions unique to each spectrum, without changing the operating parameters of the FASTFLIGHT-2. The 4 bits for the protocol number are processed when the Select Protocol input is supplied by the mass spectrometer control electronics. When the protocol number has been accepted by the FASTFLIGHT-2, it generates a confirming Protocol Accepted output. The operating conditions are switched to those specified by this new protocol number when the next spectrum acquisition begins. The Acquiring TOF Spectrum output can be used to confirm that a spectrum acquisition is in progress. In addition to the handshaking signals on the RAPID PROTOCOL port connector, the TRIGGER ENABLE IN, START OUT, BUSY OUT, and ABORT IN connectors on the rear panel can be used to synchronize the mass spectrometer controls with the FASTFLIGHT FASTFLIGHT-2 Duet Operation for a Greater Dynamic Range Figure 4 shows how two units of the FASTFLIGHT-2 and two Model 9326 Fast Preamplifiers can be operated in parallel to improve the range of signal amplitudes that can be accommodated. With the single FASTFLIGHT-2 illustrated in Fig. 3, the maximum pulse height that can be processed is limited by the 0.5-V range digitized by the flash ADC in the DSA. The minimum pulse height (or peak height) that can be analyzed is limited by the correlated noise floor. These two limits determine the analyzable dynamic range of molecular ion concentrations for a single DSA. The scheme in Fig. 4 lowers the effective correlated noise floor by a factor of 10 to improve detection limits and dynamic range by that same factor. This is accomplished by amplifying the output of the normal preamplifier (A) by a factor of 10, and applying that amplified signal to a 13

24 FASTFLIGHT 2 Digital Signal Averager Fig. 4. Duet Arrangement for Increased Dynamic Range, with FASTFLIGHT 2 Triggering the TOF-MS. second FASTFLIGHT-2. This makes the signals a factor of 10 larger in the second DSA, thus rendering the correlated noise a factor of 10 less important. FASTFLIGHT A analyzes the larger signals, while FASTFLIGHT B processes the bottom 10% of the pulse amplitudes accommodated by unit A. This scheme works best for improving detection limits on isolated 14

25 2. INSTALLATION, CONFIGURATION, AND INITIAL STARTUP peaks. Overload recovery characteristics of the preamplifiers may make it difficult to analyze small peaks on the tails of large peaks. What follows is a more detailed description of the connections and functions in Fig. 4. Except for the PREAMP POWER and USB 2.0 connections, all interconnections should be made with high-quality, 50-, RG-58A/U coaxial cables with compatible connectors. The preamplifier power connections are made using the cables incorporating 9-pin D connectors, as supplied with the Model The bi-directional communications and data connections between the FASTFLIGHTs and the PC are made via standard USB 2.0 cables. Each FASTFLIGHT-2 should be connected to a separate USB 2.0 port on the computer. Follow the instructions in Sections 2.2 through 2.3 for installing a single unit of the FAST- FLIGHT 2, and then make the changes outlined next. Insert a 50- power splitter (signal splitter) between the output of the Preamplifier A and the input to FASTFLIGHT A. This device equally splits the output signal from Preamplifier A into two paths. Connect one output of the splitter to the ANALOG IN of FASTFLIGHT A. Connect the other output of the power splitter to the INPUT of Preamplifier B. The power splitter should be capable of passing signals from dc to beyond the maximum frequency components in the signal from preamplifier A. For rise times 1 ns, an ORTEC Model MT Matched Tee Signal Splitter will suffice. For faster rise times, a more expensive splitter with a higher bandpass should be selected, such as the Pasternak Model PE2063, 7 or equivalent. BNC-to-SMA connector adapters will be needed for the Pasternak power splitter (see ORTEC models BNC/SMA and SMA/BNC). Set the gain of Preamplifier B to 10V/V (see Appendix D) in order to achieve a factor of 10 improvement in detection limits via FASTFLIGHT B. Connect the output of Preamplifier B to the ANALOG IN of FASTFLIGHT B. Using the power cable supplied with the preamplifier, power Preamplifier B from FASTFLIGHT B. This completes the analog signal connections to the two DSAs. The next set of connections involves synchronizing the two DSAs to collect coordinated spectra from the same ion acceleration pulses in the TOF-MS. Connect the 10 MHz CLOCK OUT of FASTFLIGHT A to the 10 MHz CLOCK IN of FASTFLIGHT B. This causes DSA B to lock the phase and frequency of its sampling clock to the sampling clock in DSA A. That ensures the picoseconds per bin in the horizontal scale for B will exactly match and track A

26 FASTFLIGHT 2 Digital Signal Averager Connect the BUSY OUT of unit B to the TRIGGER ENABLE IN of unit A. This guaranties that FASTFLIGHT A will not accept or generate a trigger until FASTFLIGHT B has finished its previous scan and is available to accept the next trigger. To make this synchronization work, the FASTFLIGHT A must be set up to expect a low TTL level for enabling triggering on the TRIGGER ENABLE IN input. This is done within the application program by selecting the spectrum window for instrument A, clicking on Acquisition/Instrument Properties..., then setting the Trigger Enable polarity to - [Negative] on the General Settings tab (see Section ). Next, decide if you want FASTFLIGHT A to trigger the TOF-MS (go to the next section) or if you want the TOF-MS to trigger the FASTFLIGHTs (skip to The TOF-MS Triggers the FASTFLIGHT 2 Units ). FASTFLIGHT-2 Triggers the TOF-MS To synchronize the triggers of the two DSAs and trigger the TOF-MS (refer to Fig. 4), connect a BNC tee to the Acceleration Pulser input. Couple the TRIGGER OUT of FASTFLIGHT A to one side of the BNC connector. Connect the other side of the BNC connector to the TRIGGER IN on FASTFLIGHT B. Unit B terminates the coaxial cable in 50. Therefore, you will need to ensure that the input impedance to the Acceleration Pulser is large compared to 50. Using the Instrument Properties... command in the unit-b spectrum window, set Trigger Enable polarity to - [Negative], mark the Accepting Trigger-In radio button, set the Trigger-In slope to Rising edge, and enter a trigger threshold of circa +1.4 V. Next, activate the spectrum window for unit A and mark Generating Trigger-Out. Note that the start of the time spectrum in unit B will lag the spectrum in unit A by a few tens of nanoseconds. This can be compensated via the horizontal scale calibration feature in the software (refer to Section ). Also, the duet operation limits the minimum sampling interval to 500 ps, because unit B must accept an external trigger input. You are now ready to turn on power to all parts of the system, including detector bias. CAUTION Do not turn on the detector bias voltage abruptly (several kv). Always increase the bias voltage slowly. See the Model 9326 hardware manual for preamplifier input protection limits. When you start the FF2 user interface application program as described at the beginning of Chapter 4, the two hardware units will communicate with the software, and a spectrum window will open for each unit; see Fig. 17 on page

27 2. INSTALLATION, CONFIGURATION, AND INITIAL STARTUP Finally, set up the remaining Measurement Settings dialogs for each unit. Once this is accomplished, you are ready to begin data acquisition. Section offers some initial startup suggestions. The TOF-MS Triggers the FASTFLIGHT-2 Units If you want the TOF-MS to trigger the duet, connect the trigger output of the Acceleration Pulser to the TRIGGER IN on FASTFLIGHT A and on FASTFLIGHT B as shown in Fig. 5. Because the input impedance of the TRIGGER IN is 50, a matched-impedance, signal splitter may be needed to split the TOF-MS trigger output and supply it to the two FASTFLIGHTs. The ORTEC model MT Matched Tee Signal Splitter should suffice in this case. Via the software, set the threshold and slope of the TRIGGER IN for both FASTFLIGHTs to the appropriate values for the trigger pulse supplied by the TOF-MS. Connect the BUSY OUT of FASTFLIGHT A to the TRIGGER ENABLE IN of FASTFLIGHT B. This prevents unit B from triggering when unit A is still busy with a scan. The FASTFLIGHT B should be set to expect a low TTL level for enabling triggering on the TRIGGER ENABLE IN input. In the application program, select the unit-b spectrum window, select the Acquisition/Instrument Properties... command, and set Trigger Enable polarity to - [Negative], mark the Accepting Trigger-In radio button, set the Trigger-In slope to Rising edge, and enter a trigger threshold of circa +1.4 V. Repeat these settings for instrument A. As a further precaution to ensure that both FASTFLIGHT units accept the same trigger pulses to produce coordinated scans, the Record length parameter for both instruments (located on the Protocol Settings tab; see Section ) should be set shorter than the period between the trigger pulses delivered by the TOF-MS acceleration pulser. Note that the 250-ps interlaced sampling is not possible in the duet mode. The minimum available hardware sampling interval will be 500 ps. You are now ready to turn on power to all parts of the system, including detector bias.. CAUTION Do not turn on the detector bias voltage abruptly (several kv). Always increase the bias voltage slowly. See the Model 9326 hardware manual for preamplifier input protection limits. 17

28 FASTFLIGHT 2 Digital Signal Averager Fig. 5. Duet Arrangement for Increased Dynamic Range, with the TOF-MS Triggering the FASTFLIGHT 2 Units. When you start the FF2 user interface application program as described at the beginning of Chapter 4, the two hardware units will communicate with the software, and a spectrum window will open for each unit. This window is illustrated in Fig. 17 on page

29 2. INSTALLATION, CONFIGURATION, AND INITIAL STARTUP Finally, set the remaining Measurement Settings dialogs for each unit. Once this is accomplished, you are ready to begin data acquisition. The following section offers initial operating suggestions TOF-Only Mode 2.4. Initial Operating Suggestions For initial familiarization, the TOF-only mode is recommended as the simplest to interact with. 1. Start the FASTFLIGHT-2 user interface application program as described at the beginning of Chapter 4. The software will detect the attached one or two FASTFLIGHT-2 units and will open a spectrum window for each detected instrument. For duet mode, you will perform the following steps for each unit. To select the instrument to be configured, click in the spectrum window for unit A. This window will now be the active window, the color of its titlebar will reflect the active window colors for your current Windows desktop theme, and the changes you make will affect only unit A. When you have adjusted unit A, click in the spectrum window for unit B and repeat the setup. 2. Select the TOF mode from the Acquisition menu. 3. Go to the Toolbar and click on the Instrument Properties button ( ). This will open the Measurement Settings dialog, which contains the data acquisition properties. Once the acquisition settings have been selected, use the Start ( ) and Stop ( ) buttons to control data collection. The default settings provided by the software might serve as adequate initial operating conditions that will facilitate learning how to run the FASTFLIGHT-2. However, here are a few suggestions regarding parameters you might want to change. 4. From the Acquisition menu, select Instrument Properties. On the General Settings tab choose the following parameters: Maximum time 300 seconds Maximum spectra 5000 spectra Active protocol 0 Rapid protocol Checkbox unmarked Trigger Enable polarity Positive (unless installation above dictates otherwise) 19

30 FASTFLIGHT 2 Digital Signal Averager If the chosen installation accepts a trigger from the TOF-MS, mark the Accepting Trigger-In radio button. Choose Rising edge or Falling edge according to the signal provided by the TOF-MS, and set the Trigger-In threshold approximately halfway between the high and low voltages of the trigger signal. If the selected installation has the FASTFLIGHT-2 supplying the trigger to the TOF-MS, click on the Generating Trigger-Out button. Set the width of the trigger output pulse (Trigger output width) to whatever is needed by the TOF-MS Acceleration Pulser input. 5. Click on the Protocol Settings tab and set the following properties: Protocol number 0 Protocol description Enter a description or protocol name, if desired Record length 100 µs Sampling interval 0.5 ns Records per spectrum 512 Time offset 0 µs Vertical offset V (or adjust this per Section ) Precision enhancer Checkbox marked Specific ion region Start position 1000 ns Length 90,000 ns Calibration file (Leave blank for now) Compression method Lossless Noise sensitivity threshold 3 Minimum noise threshold 10 Correlated noise subtraction Checkbox marked 6. Click on the Advanced button to access the following advanced compression settings: Background Sampling Interval Adjacent Background Maximum Peak Width Minimum Peak Width Ringing Protection 400 bins 16 bins 400 bins 4 bins 2 ns Click on OK to close this dialog and OK to close the Measurement Settings dialog. You are now ready to acquire TOF spectra. 20

31 2. INSTALLATION, CONFIGURATION, AND INITIAL STARTUP 7. Turn on the source of ionized molecules in your TOF-MS, then click on the Start button on the FASTFLIGHT-2 toolbar. If the TOF-MS is operating properly, the FASTFLIGHT-2 should begin acquiring and displaying spectra. Now is a good time to adjust the microchannel plate detector voltage to optimize the gain. Refer to Section for instructions. Clicking the Stop button completely terminates the acquisition, and clears all the elapsed time clocks and counters. Subsequently clicking the Start button will start the acquisition from zero. When you have a TOF spectrum on the display, experiment with the zooming and scaling buttons on the toolbar; these are discussed in detail in Chapter 5. Clicking on a point in the spectrum will display the X and Y coordinates of the selected point on the status bar at the bottom right corner of the application program window The Chromatograph/Trend Mode Use the same instrument settings chosen in Section 2.4.1, but select Chromat/Trend from the Acquisition menu, then click on the Start button and observe that points are being added to the chromatograph, which is displayed in the upper section of the spectrum window. Click on one of these points and note that the TOF spectrum denoted by that point is displayed in the lower section of the spectrum window. The scaling and zooming tools discussed in Chapter 5 work on the currently active spectrum window. Remember that stopping acquisition terminates the chromatograph acquisition. 21

32 FASTFLIGHT 2 Digital Signal Averager 22

33 3. HOW FASTFLIGHT-2 WORKS The purpose of this section is to provide enough information on how FASTFLIGHT-2 works to facilitate selecting the optimum operating parameters. Greater detail is provided in the specifications section, Appendix C The Basic Architecture Figure 6 is a simplified representation of the data-processing architecture of FASTFLIGHT-2. The bold arrows trace the data flow from the analog signal input through to the supporting computer. Before being sampled by the 8-bit ADC, a dc offset from a 12-bit DAC is added to the analog input signal. This offset serves two purposes. Firstly, it provides adjustment of the quiescent baseline to ensure that the baseline is near the lower limit of the active range of the sampling ADC. 8 Secondly, it adds small steps to the baseline before each scan, as required by the Precision Enhancer. 9 After this adjustment of the dc offset, the 8-bit ADC samples the analog voltage at 500-ps intervals, and converts the sample to an 8-bit digital representation. Each scan through the selected time span is initiated by a trigger event. This trigger pulse can be supplied as an external input, or the FASTFLIGHT-2 can generate the trigger output at the beginning of each scan. In the latter case, the next trigger output is automatically created 500 ns after the end of each scan. In either case, the trigger event identifies the beginning of each scan so that the ADC output for successive scans can be correctly aligned and added, point-by-point, to the sum of the prior scans in a spectrum. This point-by-point summing of successive scans takes place in the Averager Memory. For sampling intervals 500 ps, the FASTFLIGHT-2 makes only one scan through the desired time span to form a complete record. For 250-ps interleaved sampling, the FASTFLIGHT-2 makes two successive scans through the selected time span to form a complete record. These two scans are made at 500-ps sampling intervals, but the second scan is offset by 250-ps relative to the first scan. Essentially, the second scan fills in the 250-ps gaps in the first scan. Interleaved 250-ps sampling is only available when the FASTFLIGHT-2 TRIGGER OUT is used to trigger the acceleration pulse in the TOF-MS, because the 250-ps interleaving is achieved by 8 Note that the vertical offset describes the dc voltage added to the signal at the input to the FASTFLIGHT-2. The input pulses have a negative polarity. However, that polarity is inverted by the software so that the peaks protrude in a positive direction in the displayed spectrum. Therefore, a more negative vertical offset moves the baseline of the displayed spectrum in a more positive direction. 9 U.S. Patent 6,028,

34 FASTFLIGHT 2 Digital Signal Averager delaying the TRIGGER OUT pulse by 250 ps for the second scan. 10 When specifying the number of records to be summed in a single spectrum, keep in mind that this will require one scan per record for sampling intervals 500 ps, and 2 scans per record for interleaved 250-ps sampling. Fig. 6. The Basic Architecture of the FASTFLIGHT 2 Hardware. When the 8-bit data for a specific point in the time span arrives from the ADC, the Averager Memory recalls the sum of all previous records for that point, adds the new value to that sum, and writes the new sum back into the same memory location. During acquisition of the last 10 U.S. Patent 6,094,

35 2. HOW FASTFLIGHT-2 WORKS record in a spectrum, the new sum is also written to the Output Buffer Memory. At the end of the last record, the Averager Memory is immediately free to start accumulating the next spectrum, while the Output Buffer Memory pushes the prior spectrum through the rest of the data processing path to the computer. This avoids the large dead times that would be suffered if the Averager Memory had to wait for the entire spectrum to be transferred to the computer before acquiring the next spectrum. From the Output Buffer Memory, the data moves to the Data Formatter and Compressor, where the 24-bit spectra are compacted as efficiently as possible into 32-bit words for transmission to the computer. This is the block that provides whichever data compression mode has been selected, i.e., a) lossless data compression, b) lossy (peak preservation with background decimation), or c) stick diagrams consisting of peak centroids and peak areas. From the Compressor, the data is dumped into an 8-million-word FIFO (first-in-first-out) memory. This buffer smoothes the data flow in spite of the choppy transmission afforded by the USB 2.0 interface to the computer. This FIFO is big enough to hold at least 7 spectra. The USB 2.0 interface transmits the data from the 8-million-word FIFO to the computer, where it is processed by software and stored on hard disk The Acquisition Control Module and the USB 2.0 Interface The USB 2.0 interface provides two-way communication between the FASTFLIGHT-2 hardware and the supporting computer. In addition to handling the flow of spectra to the computer, the USB 2.0 passes information and commands in both directions between the computer and the Acquisition Control Module in the FASTFLIGHT-2 box. The Acquisition Control Module, in turn, organizes and oversees all the functions within the hardware The 10-MHz Reference and the 2-GHz Sampling Clock The 2-GHz sampling clock is a very stable crystal-controlled clock that provides the sampling clock ticks to the 8-bit flash ADC at 500-ps intervals. These clock ticks tell the ADC when to sample the analog input voltage. To provide synchronization of a second FASTFLIGHT-2 (or other instruments) with this internal 2-GHz clock, a 10-MHz oscillator is phased-locked to the 2-GHz crystal-controlled clock. That phase-locked 10-MHz signal is delivered to the outside world on the 10 MHz OUT connector. When it is desirable to synchronize the internal 2-GHz sampling clock to the sampling clock of another FASTFLIGHT-2 (or some other instrument), a 10-MHz reference clock can be supplied on the 10 MHz IN connector. The FASTFLIGHT-2 automatically detects the presence of a clock signal on this 10-MHz input and responds by phase-locking the 2-GHz sampling clock to the 10-MHz input signal. 25

36 FASTFLIGHT 2 Digital Signal Averager For the typical use of the 10-MHz IN/OUT signals, see the duet operation described in Section The Trigger Circuits The Trigger Synchronization and Logic block in Fig. 6 coordinates the triggering functions. Via the software, one can select either the TRIGGER IN or the TRIGGER OUT for synchronizing the scans in the FASTFLIGHT-2 with the ion acceleration pulses in the TOF-MS. Using the TRIGGER OUT to initiate the acceleration pulses in the TOF-MS provides the least jitter in coordinating the start of each scan with the sampling clock in the FASTFLIGHT-2. The output is a TTL logic pulse that rises from the low to the high voltage state when the scan begins. The pulse width can be adjusted from 64 to 5120 ns. Both the leading and trailing edges of this trigger output pulse are precisely synchronized with the 2-GHz sampling clock. So, either edge can be used to trigger the acceleration pulse in the TOF-MS. For mass spectrometers that are unable to accept a triggering pulse, but can deliver a trigger output pulse that is derived from the actual acceleration pulse, the TRIGGER IN connection can be used on the FASTFLIGHT-2. This input accepts a wide range of signal amplitudes and polarities, because it incorporates a comparator, whose sensitive slope can be selected to be positive or negative, and whose threshold can be adjusted over the range of 2.5 to +2.5 V in 10-mV steps. The jitter in synchronizing the Trigger In pulse with the 2-GHz sampling clock is ±250 ps, i.e., the trigger starts the scan on the next available clock pulse. The TRIGGER IN option delivers lower correlated noise than the TRIGGER OUT function. This is only important for >10,000 records/spectrum. Interleaved sampling at an effective 4-GHz rate is only available when the TRIGGER OUT is used to initiate the acceleration pulse in the TOF-MS. For 250-ps interleaved sampling two scans are required to form a complete record. Each scan actually uses 500-ps sampling. But, on the second scan of each pair, the phase relationship between the 2-GHz sampling clock and the TRIGGER OUT pulse is shifted by 250 ps for the entire second scan. Thus the second scan fills in the data points mid way between the data points on the first scan. This provides an effective 4-GHz sampling rate for analog signal patterns that are precisely repeatable, or for fluctuating analog signals where a lot of records are averaged to reduce the fluctuations in the final spectrum. The Trigger Enable Input can be used to control whether or not the FASTFLIGHT-2 responds to a Trigger Input or generates a Trigger Output during a specific time period. As long as the Trigger Enable is held in the disabled state by an external control signal, the FASTFLIGHT-2 will not respond to an external Trigger Input nor generate a Trigger Output. In the enabled state, the FASTFLIGHT-2 accepts Trigger Input pulses or produces Trigger Output pulses (depending on the trigger mode selected through the software). The TRIGGER ENABLE IN connector has a 10-k resistor pulling the center conductor to the high TTL state. The default 26

37 2. HOW FASTFLIGHT-2 WORKS condition considers the high TTL state to designate the enabled mode. In that case, the FASTFLIGHT-2 remains enabled with nothing connected to the TRIGGER ENABLE IN connector. Software selection also permits choosing the low TTL state as the enabled state. The Trigger Enable Input can be used to prevent triggering until the TOF-MS is ready for the next scan. If the Trigger Enable Input signal rise or fall time is <100 ns, the coaxial cable supplying the signal should be terminated in its characteristic impedance to prevent reflections. See the specifications in Appendix C for more details. The Busy Out signal provides a high TTL voltage when FASTFLIGHT-2 has accepted a Trigger Enable Input and/or an external Trigger Input and has started a scan. This Busy Out signal returns to the low state at the end of each scan, when FASTFLIGHT-2 can process another trigger. Busy Out is also held high whenever the DSA is not able to respond to a Trigger Input or a Trigger Enable Input. The Busy Out signal can be used to tell the TOF-MS controller when FASTFLIGHT-2 has finished a scan and is ready to accept instructions to begin a new scan. It is also used in conjunction with the Trigger Enable Input to keep the two FASTFLIGHT-2 units synchronized in the Duet Mode (see Section ). Another signal that is useful for synchronizing the FASTFLIGHT-2 with the TOF-MS is the Acquiring TOF Spectrum output on pin 11 of the RAPID PROTOCOL port 15-pin D connector. This TTL output resides in the high state while summing records to produce a TOF mass spectrum. It transitions from the low to high state on the first trigger for the sum, and returns to the low state at the end of the last record in the sum. The ABORT IN input can be used to promptly terminate the acquisition of a spectrum. A transition from the low to high TTL state immediately terminates the acquisition of a spectrum, even during an active scan. The minimum Abort signal duration in the high state is 50 ns. At the end of the Abort signal, the FASTFLIGHT-2 is ready to accept a Trigger In to start the first scan of a new spectrum. If operating in the Trigger Out mode, FASTFLIGHT-2 will generate the next Trigger Out immediately after the end of the Abort signal to start the acquisition of a new spectrum. When the TOF-MS and FASTFLIGHT-2 are already acquiring spectra in the chromatograph mode, the Abort In can be used to synchronize the start of the next spectrum with the exact time a sample is injected into the chromatograph that is feeding the TOF-MS. At the time of injection, one of the tag bits on the 15-pin RAPID PROTOCOL port connector is set to the high state, and the Abort signal is simultaneously asserted. The tag bit is held high long enough to overlap the next trigger pulse. Thus, the new spectrum acquisition will be marked with the tag bit, and the start of that acquisition will be synchronized with the time of sample injection into the chromatograph. The precision of the synchronization is controlled primarily by the uncertainty in determining the exact time of sample injection, and secondarily by the width of the Abort signal. 27

38 FASTFLIGHT 2 Digital Signal Averager If the TOF-MS and FASTFLIGHT-2 can idle until the sample is injected into the chromatograph, the Trigger Enable Input can be held low until the sample is injected, and then released to the enabled state upon sample injection. This starts the chromatograph/tof acquisition at the same time as the sample injection The RAPID PROTOCOL Port The RAPID PROTOCOL port provides the ability to switch the operating parameters of the FASTFLIGHT-2 hardware in a time interval as short as 10 µs, by simply accepting the desired protocol number and a Select Protocol signal. This is much faster than can be achieved through software controls. This hardware input is useful in the chromatograph mode, where it is often desirable to interleave different TOF-MS acquisition conditions on sequential time-of-flight spectra in the chromatograph. For example, the electrospray nozzle conditions might be changed to enhance different charge states. For a mass spectrometer that can rapidly switch between the fragmentation mode and the precursor ion mode, fragmentation and precursor-ion spectra can be interleaved in the chromatograph via alternating TOF spectra. By predefining the operating parameters listed under two or more protocol numbers, the RAPID PROTOCOL port can be used to change the FASTFLIGHT-2 acquisition conditions within 10 µs between sequential time-of-flight spectra to match the changed TOF-MS parameters. Up to 16 different protocols can be predefined for rapid selection on the Protocol Settings tab under Instrument Properties. The acquisition settings listed below can be specified uniquely for each of the 16 protocol numbers. You can enter a unique, meaningful name for each protocol, and the software will store that name with the spectrum. 28 Protocol number Protocol name (user defined) Record length Sampling interval Records per spectrum Time offset Vertical offset Specific ion window: begin/end (for the chromatograph vertical coordinate) Calibration file path and name (for converting TOF to m/z) Peak vs. background discrimination Minimum noise threshold Auto noise threshold: % On/off % Sensitivity factor Background sampling interval Adjacent background points surrounding a peak Maximum peak width Minimum peak width Ringing protection interval after a peak Automatic correlated noise subtraction on/off

39 2. HOW FASTFLIGHT-2 WORKS Data compression type: Lossless Lossy: peaks preserved, background suppressed Stick diagrams: only centroid and area The protocol number must be presented as 4 binary bits to the rear-panel RAPID PROTOCOL connector (see Fig. 7 and Table 2). If you wish to change to a new protocol on spectrum number N, then the new protocol should be selected during acquisition of spectrum N1 and before the first trigger is generated for spectrum N. The Acquiring TOF Spectrum output (pin 11 on the RAPID PROTOCOL connector) can be monitored to determine when FASTFLIGHT-2 is busy acquiring a spectrum. To be safe, wait at least 1 µs after the first trigger for spectrum N1 has disappeared before selecting the new protocol. Fig. 7. The RAPID PROTOCOL Port Connector. See Table 2 for the pin assignments. Table 2. Signals and Pin Assignments on the RAPID PROTOCOL Connector. Pin Number FASTFLIGHT-2 TTL Signal (on male, 15-pin D connector) 1 Protocol Number bit 0 2 Protocol Number bit 1 3 Protocol Number bit 2 4 Protocol Number bit 3 5 Tag Bit 0 6 Tag Bit 1 7 Tag Bit 2 8 Tag Bit 3 9 Select Protocol (Input) 10 Protocol Accepted (Output) 11 Acquiring TOF Spectrum (Output) 12 Ground If you are using the TRIGGER OUT from the FASTFLIGHT-2, make sure you select the new protocol number at least 1 µs before the end of acquisition for spectrum N1. Alternatively, you can hold the Trigger Enable Input in the disabled state after the last trigger in spectrum N1, and maintain the disabled state until you have selected the new protocol for spectrum N. If you are supplying the Trigger In signal, do not deliver the trigger for spectrum N until the FASTFLIGHT-2 has changed protocols and is ready for the next trigger. It takes 10 µs 29

40 FASTFLIGHT 2 Digital Signal Averager following the end of acquisition of spectrum N1 to change the parameters specified by the new protocol before FASTFLIGHT-2 is ready to generate or accept the first trigger for spectrum N. To select the new protocol, first assert the 4-bit protocol number. Next, issue the Select Protocol pulse. When the hardware has stored the new protocol number it will respond by providing a Protocol Accepted output. At that point the Select Protocol signal should be removed and the 4-bit protocol number can be withdrawn. The FASTFLIGHT-2 will switch to that new protocol number after the end of spectrum number N1 and before the first trigger in the next spectrum. If you are using the Trigger Out, a gap of 10 µs will occur between the end of spectrum number N1 and the start of spectrum N. This gap can be lengthened by holding the Trigger Enable Input in the disabled state, or by waiting longer to supply the next trigger pulse when the Trigger In is being used. One reason you might wish to extend this gap is to allow the TOF-MS to settle into its new operating conditions if you have changed the parameters for the mass spectrometer. Tag bits are also included in the RAPID PROTOCOL connector, although they are not involved in selecting the protocol. The 4 tag bits can be used to mark up to 16 different conditions in a spectrum. The first trigger in each spectrum samples the state of the tag bits, and that information is saved in the header of each spectrum. NOTE To avoid potential conflicts, the RAPID PROTOCOL port is disabled when data acquisition is not in progress. Consequently, changing the properties that define the protocols (and/or the General Settings properties) via the software is only possible when data acquisition is not taking place. The software updates the data acquisition properties in the FASTFLIGHT-2 hardware when the Measurement Settings dialog is closed, and acquisition cannot begin until this process is complete Additional Information on Rapid Protocol Selection You can set the next protocol number while the Acquiring TOF Spectrum signal is in the high state. In this case, the protocol does not change until Acquiring TOF Spectrum goes low. For this option there is one caveat. You must wait for at least 10 µs after Acquiring TOF Spectrum goes high before asserting the Select Protocol signal. The Protocol Accepted signal remains in the high state until the end of the current spectrum acquisition plus 8 µs, the time required to change to the next protocol s measurement settings. Alternatively, you can wait until the Acquiring TOF Spectrum signal is low and then set the new protocol number with the Select Protocol signal, while holding the TRIGGER ENABLE IN in the disabled state. 30

41 2. HOW FASTFLIGHT-2 WORKS When FIFO Memories are Full The digital data representing the spectra flows through many processing blocks in the pipeline from the Output Buffer Memory to the hard disk in the supporting computer. At each step of the process the data flow tends to be in bursts that are not synchronized with the needs of earlier or later stages in the pipeline. Consequently, FIFO memories are used to buffer the flow so that each block can operate according to its optimal needs. There are a number of small FIFO memories in that pipeline. But, the two memories large enough to be of major significance are the Output Buffer Memory and the 8-M-word FIFO Memory. At the end of the acquisition of each spectrum, the Output Buffer Memory holds a copy of the complete spectrum. This allows the Averager Memory to begin acquiring the next spectrum within 0.8 µs, while the Output Buffer Memory pushes the previous spectrum through the pipeline to the computer. The 8-M-word FIFO is the major resource for accepting the intermittent bursts of data from the earlier stages of the pipeline and accommodating the sporadic demands of the later stages for processing that data. The 8-M-word FIFO can hold up to 7 full spectra of 750-µs length, at 0.5-ns sampling and with lossless data compression. Under normal conditions (up to 100 spectra per second), the 8-M-word FIFO never fills up. But, if the downstream processing blocks stop taking data, the 8-M-word memory will soon fill up and stop accepting new input data. Shortly thereafter, the Output Buffer Memory will no longer be able to transmit any residual data it has not yet transferred to the Data Formatter and Compressor. If the previous spectrum has not been completely drained from the Output Buffer Memory when acquisition begins for the last record in the next spectrum, the next spectrum will not be written into the Output Buffer Memory. Thus, complete spectra are thrown away at the interface between the Averager Memory and the Output Buffer Memory, if the Output Buffer Memory has not been drained in time to accept the next spectrum when it is ready for transfer. Thus the data steam includes only complete spectra, i.e., no partial spectra. If spectrum decimation is encountered, the solution is to first make sure nothing is interfering with the efficient running of the software and expeditious transfer of the data to hard disk. This requires a hard disk with adequate sustained-writing speed (see the hard drive specification and footnote 24 on page 150). Second, ensure that no other devices are stealing processing time from the USB 2.0 bus. The FASTFLIGHT-2 must be the only device served by the USB 2.0 port. Third, check that the computer meets the minimum processor speed and memory size specified for the FASTFLIGHT-2. Fourth, if all else fails to resolve the bottleneck, increase the number of records averaged in each spectrum to lower the spectra/s rate to a sustainable level Estimating the Maximum Possible Data Rate The number of spectra per second that will be saved on hard disk in the chromatograph mode is given by: 31

42 FASTFLIGHT 2 Digital Signal Averager S 1 (T d L ) Rs (2) provided that the maximum data rate has not been exceeded for the limiting bottleneck in the system. In Eq. 2, T d is the delay from the trigger to the start of each record (in seconds), L is the length of the spectrum in seconds, R is the number of records per spectrum, and s is the number of scans per record. The end-of-scan, end-of-record, and end-of-spectrum dead times are accounted for by the = 0.8 µs. For hardware sampling intervals 500 ps, the number of scans per record is s = 1. For 250-ps interlaced sampling, s = 2. The path through which the data flows from the sampling ADC to the hard disk has many functional blocks. Each of these can transfer data at a different maximum rate. In terms of sustainable data rates in the chromatograph mode, the most important bottlenecks are 1) the hard disk, 2) the USB 2.0 interface, and 3) the data transfer from the Output Buffer Memory to the Compressor. With the computer speed and memory size recommended in Section C.2.2.1, the USB 2.0 interface is capable of sustaining a maximum data rate of approximately 37 MB/s. To be able to save spectra with a 50-µs length and 500-ps sampling in the chromatograph mode at the rate of 100 spectra/s, with lossless data compression, the hard disk must be able to support sustained writing at 20 MB/s. Disk manufacturers and computer manufacturers typically specify the burst rate for the hard disk, but seldom mention the sustained writing speed. A 7200-rpm UDMA-150 hard disk is a minimum criterion for achieving the 20-MB/s sustained writing speed, but you should test your hard disk to determine its actual capability. This can be done by acquiring data in the chromatograph mode for a preset period of approximately 60 seconds. Divide the resulting file size in MB by the elapsed time to calculate the MB/s data rate. If the chromatograph spectrum exhibits gaps where spectra have been lost, you are exceeding the maximum possible data rate. Increase the number of records per spectrum until the missing spectra are filled in. The continuous data rate generated by the FASTFLIGHT-2 hardware can be predicted by Eq. 3: D 3L r c T s (T d L ) Rs (3) 32

43 2. HOW FASTFLIGHT-2 WORKS D is the data rate in bytes/s, and T s is the sampling interval in seconds. The compression ratio, r c, is measured as the ratio of the MB/s flowing into the Data Compressor divided by the MB/s flowing out of the Data Compressor. With lossless data compression, r c 3/2 for R 255 (4) because 3 bytes are reduced to 2 bytes by discarding the unpopulated top byte. If the entire spectrum consists of 3-byte data points, r c = 1. It will be rare to find a significant portion of the spectrum occupying more than 2 bytes. Consequently, r c will usually be close to, but slightly below, 1.5 for a typical spectrum with lossless compression and R > 256. Much larger values for r c will be experienced for lossy data compression (peak-preserving and background-decimating) and for stick diagrams (only peak centroid and area). Here is an example calculation using Eqs. 2 and 3. Presume the following values for the parameters: T d = 0 L = 50 µs r c = 3/2 T s = 0.5 ns R = 196 s = 1 From Eq. 2 the spectral rate is S = 100 spectra/s, and from Eq. 3 the data rate is D = 20 MB/s. There is one further limit to consider. The Output Buffer Memory must always be ready to accept the new spectrum during the last record in the spectrum. In the time required to acquire R1 records, the Output Buffer Memory must complete transferring the previous spectrum to the Data Compressor. This ensures that the dead time at the end of a spectrum is no more than 0.8 µs, and that no spectra are lost because of an unavailable Output Buffer Memory. There will be no lost spectra if the number of records per spectrum satisfies the following condition: R min 3L r c T s (T d L s) D max 1 (5) Where D max in bytes/s is the maximum data rate allowed by the most restrictive bottleneck in the data pipeline, and R min is the minimum required number of records per spectrum. Using the 33

44 FASTFLIGHT 2 Digital Signal Averager previous example, we find that R min must be 197.9, whereas we had used 196. This invalid operating condition can be rectified by reducing the record length to L = 49.7 µs, and employing R = 198. The result from Eqs. 2, 3, and 5 is: S D R min = 100 spectra/s = MB/s (which is less than D max = 20 MB/s) (which is less than the R = 198 records/spectrum we used) This set of conditions is viable. Pragmatically, one deals with this issue by simply increasing the number of records/spectrum, R, until the chromatograph exhibits no lost spectra. However, the above equations can be used for some guidance in advance of the measurement The FASTFLIGHT-2 Precision Enhancer 11 For analog input signals incorporating noise that is negligible compared to the width of 1 LSB in the sampling ADC, the 8-bit resolution of the ADC limits the size of the smallest signal that can be uncovered through averaging. Figs. 8 and 9 illustrate the principle. Figure 8 shows a signal exhibiting two small peaks on either side of one tall peak. The amplitude scale is calibrated in terms of the boundaries of the first 10 LSB of the ADC. The tall peak spans 8 LSB of the ADC, while the small peaks are contained entirely within the boundaries of 1 LSB. Note that the noise on the signal is negligible compared to 1 LSB. Figure 9 shows the digital output of the ADC after acquiring a single record of the signal in Fig. 8. Because the small peaks never cross a bit boundary, they are completely lost in the digital output. No matter how many records are averaged, the spectrum will always look like Fig. 9. Figure 10 illustrates the result of summing 256 records. Specifically, except for a multiplication of the amplitude scale, Fig. 10 looks exactly like Fig. 9. One solution to the problem of the ADC resolution limit is to add random noise to the input signal. The FWHM of the Gaussian noise is chosen to be large compared to 1 LSB. This allows the small peaks to randomly cross the LSB boundaries from record to record. Consequently, averaging over a large number of records will reveal the details of the small peaks in the final spectrum. The drawback to adding noise is that it takes a very large number of records in the average to get back to the signal-to-noise ratio inherent in the signal before the noise was added. 11 U.S. Patent 6,028,

45 2. HOW FASTFLIGHT-2 WORKS Analog Input Signal Amplitude (LSB) Time (ns) Fig. 8. An analog input signal with negligible noise. The amplitude scale is calibrated in terms of LSB of the ADC. Sampled Data from ADC Amplitude (LSB) Sample Time (ns) Fig. 9. The digital output of the ADC for the signal in Fig. 8. The amplitude scale is calibrated in terms of LSB of the ADC. The Precision Enhancer in FASTFLIGHT-2 uses a different scheme to recover low-noise signals having amplitudes <1 LSB. It utilizes the 12-bit DAC in Fig. 6. On each successive record, the dc offset added by the DAC is changed. Because the offset is known and, in principle, can be subtracted from the digital output, the DAC does not add any residual noise. The basic increment of the DAC is approximately equal to 1/16 LSB at the ADC, and the range 35

46 FASTFLIGHT 2 Digital Signal Averager Precision Enhancer OFF: Sum of 256 Records Amplitude (LSB) Sample Time (ns) Fig. 10. The spectrum of Fig. 8 acquired by summing 256 records with the Precision Enhancer turned OFF. The major peak is evident, but the smaller peaks to the left and right of the major peak are not revealed. of increments spans approximately ±8 LSB at the ADC. The selection of the DAC offset is driven by a pseudo-random-number generator. Consequently, after a number of records, the small signals have been randomly moved up and down and have been compared with a variety of ADC LSB boundaries. This allows the instrument to record small signals to a resolution of 1/16 LSB. Because the technique adds no noise to the signal, it converges to 1/16 LSB precision much faster than the method of adding random noise. Furthermore, adjusting the dc offset over a range of ±8 LSB improves the differential nonlinearity of the 8-bit ADC by a factor of 16. This is critical for sub-lsb resolution because the 8-bit ADC typically has a differential nonlinearity of ±0.9 LSB. With the Precision Enhancer this improves to about ±1/16 LSB. Figure 11 shows the result of using the Precision Enhancer with the spectrum of Fig. 8. After summing 256 records, the two small peaks are clearly defined. Although it theoretically takes 256 records to sweep through all the Precision Enhancer codes, the precision converges most rapidly in the first 16 records. Thus it is not necessary to use 256 records in order to benefit from the Precision Enhancer. Conversely, there is no penalty in using more than 256 records. On the first record in a spectrum, the Precision Enhancer contributes no additional offset. On subsequent records it randomly steps the offsets in positive and negative directions to cover the ±8 LSB range. Consequently, the average offset added is extremely close to zero, and it is not 36

47 2. HOW FASTFLIGHT-2 WORKS Precision Enhancer ON: Sum of 256 Records Amplitude (LSB) Sample Time (ns) Fig. 11. The spectrum of Fig. 8 acquired by summing 256 records with the Precision Enhancer turned ON. Note that the two smaller peaks to the left and right of the main peak are now visible. necessary to subtract the digital representation of the offset from the final spectrum. However, it is important to set the starting value of the dc offset high enough that the Precision Enhancer does not step below the lower limit of the ADC range. The starting dc offset must be greater than8 LSB (16 mv). A good default choice for the vertical offset on the Protocol Settings tab is 0.05 V. The Precision Enhancer can be turned on and off in the Protocol dialog to demonstrate the improvements in digital resolution and differential nonlinearity on small or slowly varying signals that exhibit sub-lsb noise. Without the Precision Enhancer, the digital precision and accuracy of the 8-bit sampling ADC are limited by the width of one least significant bit (1 LSB) and by the variability of that width throughout the voltage span of the ADC. The variability of the width is commonly referred to as the differential nonlinearity. Figure 12 demonstrates the digital resolution and differential nonlinearity of the 8-bit ADC. A voltage that linearly ramps from 0 to 0.5 V in 70 µs was applied to the FASTFLIGHT-2 input, and 8192 records were averaged. Figure 12 shows an expanded view around mid-scale. Each stair step corresponds to 1 LSB of the ADC. The variability in the width of the stair steps is caused by the differential nonlinearity. 37

48 FASTFLIGHT 2 Digital Signal Averager In Fig. 13, conditions are the same as in Fig. 12, however, the Precision Enhancer has been turned on. Note that the stair steps have essentially disappeared because the digital resolution has improved by a factor of 16, as has the differential nonlinearity. The Precision Enhancer transforms the 8-bit ADC into a 12-bit ADC for a factor of 16 improvement in dynamic range. Fig. 12. The digital resolution and differential nonlinearity of the 8-bit ADC with the Precision Enhancer OFF. Fig. 13. The same conditions as for Fig. 12, but with the Precision Enhancer ON. Note that the digital resolution and differential nonlinearity improve by a factor of

49 2. HOW FASTFLIGHT-2 WORKS This permits accurate digital resolution of signal amplitudes less than 2 mv. It also improves the accuracy with which large signals are digitized. The Precision Enhancer should be left on for all serious data acquisition. The Precision Enhancer can improve the accuracy in the measurement of the centroid and area of a peak whose amplitude is not large compared to 1 LSB of the 8-bit ADC. This improvement occurs because of the reduction in digital resolution to 1/16 LSB and the improvement of differential nonlinearity to ±1/16 LSB Lossless Data Compression 3.2. Compressed Data Formats Because the sampling ADC digitizes the analog signal with 8 bits, each record consists of a series of 8-bit words, one word for each sample. The samples are spaced at 500-ps intervals, if 0.5-ns sampling has been chosen. The Averager Memory permits the summing of up to 65,535 records to form a spectrum. Consequently, 24 bits must be reserved for each data point in the Averager Memory. Thus, the averager naturally produces a stream of 24-bit (i.e., 3-byte) words, with each word representing a sampled and summed point in the time spectrum. The USB 2.0 interface and the supporting personal computer are organized around 4-byte (32- bit) words. Therefore, one function of the Data Formatter and Compressor is to rearrange the 3- byte words into 4-byte words for efficient transfer to the computer. A simple algorithm for this compaction is to combine 4 adjacent 3-byte words into three 4-byte words by rearranging the lower (least significant) byte (LB), the middle byte (MB) and the high byte (HB). For example, the 3-byte words for bins 1001, 1002, 1003 and 1004 can be depicted as the following series of words. Byte 1 Byte 2 Byte 3 1st Word: LB(1001) MB(1001) HB(1001) 2nd Word: LB(1002) MB(1002) HB(1002) 3rd Word: LB(1003) MB(1003) HB(1003) 4th Word: LB(1004) MB(1004) HB(1004) The numbers in parentheses denote the bin number. These four words can be rearranged into three 4-byte words as follows. 39

50 FASTFLIGHT 2 Digital Signal Averager Byte 1 Byte 2 Byte 3 Byte 4 1st Word: HB(1001) HB(1002) HB(1003) HB(1004) 2nd Word: MB(1001) MB(1002) MB(1003) MB(1004) 3rd Word: LB(1001) LB(1002) LB(1003) LB(1004) In applications where less than 256 records are summed in each spectrum, all the data points will have zeros in the high byte (HB = 0). Only 2 bytes are needed to convey the data. Even in a spectrum that incorporates the sum of 2,048 records, data seldom extends into the highest byte. Most of the points in the spectrum contain low-level background, and require at most 2 bytes (presuming that the baseline has been set at <10% of full scale). Consequently, additional data compression can be achieved by using only 2 bytes to represent the background data. In these two cases, four 2-byte words can be combined into two 4-byte words. For example, Byte 1 Byte 2 Byte 3 1st Word: LB(1005) MB(1005) HB(1005) = 0 2nd Word: LB(1006) MB(1006) HB(1006) = 0 3rd Word: LB(1007) MB(1007) HB(1007) = 0 4th Word: LB(1008) MB(1008) HB(1008) = 0 can be combined into: Byte 1 Byte 2 Byte 3 Byte 4 1st Word: MB(1005) MB(1006) MB(1007) MB(1008) 2nd Word: LB(1005) LB(1006) LB(1007) LB(1008) The alternative to this compaction, would be to send the original 24-bit words to the computer as 32-bit words with zeros in the most significant bytes. Compared to that alternative, the reorganization described above reduces the file size by as much as a factor of 2. This is the basis of the lossless data compression option. No information is lost as the data is compressed by a factor approaching 2. How does the computer decipher the compacted data? A system of header words is used to tell the decoding software the starting bin number of the next block of data, and whether that block originated as 3-byte or 2-byte data. The header words are only needed when the structure 40

51 2. HOW FASTFLIGHT-2 WORKS changes between 3-byte and 2-byte origins. Thus, long background regions require only one decoding header for a large number of points. Also, spectra formed from less than 256 records only require one decoding header for the entire spectrum. For details on the code words in the data stream, see Section B Lossy (Peak-Preserving and Background-Suppressing) Compression The lossy compression uses the same compaction of 2- and 3-byte words into 4-byte words, but it transmits only a small fraction of the useless background while preserving all the interesting details in the peak regions. 12 An automatic noise threshold determines what parts of the spectrum are background and what features correspond to peak regions. In the background region every n th packet of 4 adjacent points is transmitted to the computer, while the rest of the background points are discarded. The value n is adjustable, with a setting between 200 and 400 being typical. With a setting of 400, the contribution of the background points to the final file size for the compressed spectrum is reduced by a factor of 100. This is the primary source of compression, because background accounts for the vast majority of points in an electrospray TOF-MS spectrum. Not only are the peak regions preserved, but adjacent background points are included on either side of the peak, so that background subtraction can be performed on the saved spectrum to determine the net area of the peak above background, and to permit an accurate peak centroid to be computed. Using this scheme for data compression, an additional factor of 10 to 20 reduction in the file size can be achieved. Including the lossless compaction, an overall compression by a factor of 10 to 40 can be obtained. The sensitivity for detecting small peaks can be adjusted with the Noise sensitivity multiplier field on the Protocol Settings tab under Instrument Properties (see Section ). Values of 2, 3, or 4 are the available options. A sensitivity of 2 will find the smallest peaks, but it will also falsely identify some background regions as peaks. A sensitivity of 4 will completely ignore background excursions, but will miss the smallest peaks. A sensitivity of 3 is recommended for general use. In addition to the adaptive automatic noise threshold, the same dialog offers a selectable minimum value for the threshold above noise. This Minimum noise threshold rides a fixed distance above the varying mean of the background. The value determined by the automatic noise threshold is added to the minimum threshold. A good default value for the Minimum noise threshold is 10 LSB (the LSB units refer to the least significant bit in the digital output of the sampling ADC) to handle regions of the spectrum where the arrival of ions is sparse. 12 U.S. Patent 5,995,

52 FASTFLIGHT 2 Digital Signal Averager Higher values of both the Noise sensitivyt multiplier and Minimum noise threshold reduce the false identification of background as peaks, but will cause the algorithm to miss some of the smallest peaks. From the Protocol Settings tab, click on the Advanced... button to examine the Advanced Compression Settings dialog. This allows you to adjust the other parameters that affect this data compression mode. The adjustable parameters include: The factor n for the background decimation. The number of adjacent background points to be preserved on either side of a peak. The maximum number of bins that can still be considered a peak and not a hump in the background. The minimum expected peak width (important for avoiding false identification of single-bin background fluctuations as peaks). The region following a peak that might be distorted by ringing in the microchannel plate detector (prevents the auto-noise sensor from sampling the baseline and the background noise during the ringing). Note that the automatic noise threshold assists the algorithm that determines the local background in the spectrum. The total net area above background for all the peaks in the spectrum forms the Total Ion Count displayed as the vertical scale in the chromatograph plot Stick Diagram Spectra: Only Centroids and Area Once the compressor has detected peak regions and suppressed background, the next step for further data compression is to reduce the peak regions to two numbers: The net area above background. The centroid of the peak region. The resulting compressed spectra are displayed as sticks. The position of the stick on the horizontal axis corresponds to the time for the centroid, while the height of the stick designates the net area of the peak region. 42

53 2. HOW FASTFLIGHT-2 WORKS Actually, the compressor in the FASTFLIGHT-2 transmits The net area of the peak. The product of the centroid and the net area. The software divides the former into the latter to display the centroid. An additional compression by a factor of 5 to 10 is available via this extra step. So, the overall data compression can be 50 to 400 in the stick diagram mode. This compression can significantly reduce the file size for chromatograph/tof-ms acquisitions Noise and Ion Counting Statistics Random vs. Correlated Noise A DSA, such as the FASTFLIGHT-2, can be a powerful tool for reducing the parasitic random noise and fluctuations that make it difficult to capture a precise record of the analog signal waveform. FASTFLIGHT-2 accomplishes this reduction of the random noise by averaging many records to form each spectrum. This improves the signal-to-noise ratio by the square root of the number of records averaged, as far as random noise is concerned. The FASTFLIGHT-2 hardware implements the averaging by rapidly summing the contributions of all the records to produce the final spectrum. If the user wishes to convert this sum to the average, the displayed data must be divided by the number of records in the sum. Because this division does not result in a more meaningful display for the TOF-MS application, FASTFLIGHT-2 simply displays the sum and omits the division by the number of records. The reason the summing improves the signal-to-noise ratio, is that the signal is systematic and sums linearly in proportion to the number of records, while the noise is random in polarity and amplitude and sums as the square root of the number of records. Thus, the signal-to-noise ratio improves in proportion to the square root of the number of records summed. Not all sources of noise succumb to reduction with averaging. Any source of interference that is always the same and is correlated with the sampling clock or the triggering time in the FASTFLIGHT-2 will not be reduced by sampling. This occurs because such correlated noise sums linearly in proportion to the number of records averaged, just like the signal of interest does. For correlated noise, the signal-to-noise ratio does not improve as the number of records summed is increased. The suppression of random noise is normally apparent as the number of records summed is initially increased. But, when the number of summed records becomes very large, the correlated noise emerges as the limiting noise contribution. Thus, the correlated noise sets the ultimate detection limit for small peaks in the spectrum. 43

54 FASTFLIGHT 2 Digital Signal Averager Typical sources of correlated noise are: 1. Electrical interference caused by, or synchronized with the TOF-MS accelerator pulse. 2. Coaxial cable reflections of the analog input signal. 3. Ringing of the detector output signal following each pulse. 4. Residual noise correlated with the sampling clock and trigger within the DSA. Typical sources of random (uncorrelated) noise are: a. Statistical fluctuations in the number of ions from the acceleration source in the TOF-MS. b. Statistical fluctuations in the gain of the microchannel plate detector. c. Thermal (random) noise from the input stages of the preamplifier and the input circuits of the DSA. d. Unrelated environmental electrical interference. e. Power supply and power line noise Methods to Suppress Correlated Noise Correlated noise sources 1 3 can be reduced by improving the design of the TOF-MS, keeping the coaxial signal cables as short as possible and terminated in the correct impedance, and by choosing a detector that exhibits minimum ringing. Most of the design choices that suppress correlated source 4, have already been implemented in the FASTFLIGHT-2. Given that the design of the entire system has minimized the correlated noise, there is one remaining technique available to reduce the internal correlated noise contribution from the FASTFLIGHT-2: correlated noise subtraction. In order to acquire a correlated-noise reference spectrum, increase the number of records in the spectrum to the point where correlated noise significantly dominates over random noise, and prevent the molecular ions from reaching the detector. Such a spectrum deletes the signal and contains only the correlated noise pattern. This correlated noise spectrum can be stored and subtracted from subsequent spectra that include the signal. As a result the correlated noise will be subtracted from the spectra of interest, thus reducing detection limits for small peaks. Pragmatically, the correlated noise can be reduced by a factor between 5 and 10 using this correlated noise subtraction technique. The method in the previous paragraph requires that acquisition conditions are identical between the correlated-noise reference spectrum and the TOF-MS spectra. This is sometimes hard to guaranty. To overcome that restriction, the FASTFLIGHT-2 incorporates an adaptive correlated noise subtraction process that happens automatically, and intrinsically uses the identical acquisition conditions. Most of the residual correlated noise within the FASTFLIGHT-2 is caused by circuits switched by the sampling clock. This correlated noise has a pattern that is repeated with a constant period from one end of the spectrum to the other. As each spectrum begins to flow through the Data 44

55 2. HOW FASTFLIGHT-2 WORKS Formatter and Compressor, the firmware automatically detects the correlated noise pattern in the background between peaks and subtracts it from the spectrum. As the processing moves through a spectrum, this correlated noise pattern is updated to reflect local conditions. Thus the correlated noise subtraction is matched to the local acquisition conditions in each spectrum, i.e., it is adaptive. To evaluate the effectiveness of this correlated noise subtraction, it can be turned on and off via the instrument properties in the software. For an additional suppression of the correlated noise floor by a factor of 10, see Section , which describes the duet technique Ion Counting Statistics and Detector Gain Statistics For virtually all practical cases in TOF-MS, ion counting statistics is the dominant source of random error in the area and centroid of a peak in the time-of-flight spectrum. Detector gain statistics is the second most important source. A specific peak in the time-of-flight spectrum is formed by detecting N molecular ions at the input to the microchannel plate detector. As developed in ORTEC Application Note AN61, 13 the percent random error in the area of the peak is given by: 1 A % 1 A A 100% 1 N 1 n e 100% (6) where A is the area of the peak, 1 A is the standard deviation (random error) in the area, and n e is the average number of electrons created at the cathode of the microchannel plate detector by each ion. The random error in the measured centroid of the peak, 1 Ct, as a percent of the jitter, 1 ft, inherent in the flight time from acceleration source to detector has been shown to be: D. A. Gedcke, ORTEC Application Note AN61, How Counting Statistics and the ADC Sampling Interval Control Mass Accuracy in Time-of-Flight Mass Spectrometry, July 23, 2001, AMETEK Advanced Measurement Technology, 14 The order of subscripts on 1 tc has been reversed to 1 Ct for clarity. 45

56 FASTFLIGHT 2 Digital Signal Averager 1 C % 1 Ct 1 ft 100% 1 N 1 n e 100% (7) Note that the right sides of Eqs. 6 and 7 are identical. These two equations demonstrate that it is most important to collect a large number N of molecular ions to reduce the random errors in peak area and peak centroid. Secondly, making n e (the average number of electrons injected into the microchannel plate detector by each molecular ion) large, an improvement of the order of 100% 1.4 can be gained. If n e» 1, then the right sides of Eqs. 6 and 7 become.table 3 provides N a useful reminder of the order of magnitude of improvement gained by making N large. Table 3. The Dependence of the Percent Random Error on the Number of Ions, N. Number of Ions, Percent Random Error, N 100% / N , ,000, The value of N can be made larger by increasing the number of molecular ions delivered by the mass spectrometer in a given measurement time, or by increasing the acquisition time for a spectrum. The latter is accomplished by summing a larger number of records in the spectrum The Systematic Error Due to the Sampling Interval ORTEC Application Note AN61 15 shows that the systematic error in the area and centroid of a peak in the time-of-flight spectrum will be less than 1%, if at least 1.3 samples span the FWHM (full width at half maximum height) of the peak. The minimum real time sampling interval for the FASTFLIGHT-2 is T s = 500 ps. Thus, in the worst case of a peak that captures only a single ion in each record, the FWHM of the peak should be greater than 500 ps 1.3 = 650 ps. As depicted in Fig. 14, the bandpass limitations in the preamplifier and FASTFLIGHT-2 typically limit the FWHM to a minimum of circa 1 ns. Consequently, the systematic sampling error will be <0.001% for both area and centroid. See Figs. 15 and 16, and AN61 for details. In this case, 15 D. A. Gedcke, ORTEC Application Note AN61, How Counting Statistics and the ADC Sampling Interval Control Mass Accuracy in Time-of-Flight Mass Spectrometry, July 23, 2001, AMETEK Advanced Measurement Technology, 46

57 2. HOW FASTFLIGHT-2 WORKS the random error from ion counting statistics will be the overwhelmingly dominant source of error. The 1- and 2-ns sampling intervals are useful for reducing the file size on proportionately wider peaks. The 1 and 2-ns intervals are accomplished by deletion of the intervening 500-ps samples. Consequently, the 2-ns sampling interval might need a lower bandpass filter to avoid aliasing. See AN61 for an explanation. Fig. 14. Impulse responses at three points in the analog signal processing chain: the output of the microchannel plate detector (µcp), the preamplifier output, and the input to the sampling ADC in the digital signal averager. The pulse shapes are the response to a single ion striking the detector. The polarities have been inverted and the amplitudes adjusted to simplify the depiction. 47

58 FASTFLIGHT 2 Digital Signal Averager Fig. 15. The maximum systematic centroid error due to an asymmetric alignment of the sampling interval relative to the true centroid of the Gaussian peak. Fig. 16. The maximum systematic error in the estimated area caused by misalignment of the sampling interval relative to the centroid of the Gaussian peak. 48

59 2. HOW FASTFLIGHT-2 WORKS How the Centroid Error Transforms from the Time to the m/z Domain Referring to Fig. 1, the flight time, t, of a molecular ion from the grounded grid at the output of the acceleration region to the detector in the TOF-MS is given by: t 2 d 2 2V m z (8) where d is the distance from the grounded grid to the detector, V is the accelerating voltage, m is the mass of the molecular ion, and z is the charge on the molecule. By taking the logarithm of both sides of Eq. 8 and differentiating, one can show that the relative change in m/z is related to the relative change in t by d(m/z) (m/z) 2 dt t (9) Thus, when the time-of-flight spectrum is calibrated in terms of m/z, any error in time, i.e., 1 Ct, will be amplified by a factor of 2 via the equation: 1 Cm/z m/z 2 1 Ct t (10) where 1 Cm/z is the resulting centroid error as expressed in units of m/z. The FASTFLIGHT-2 software provides a choice of linear, quadratic and cubic calibration curves to convert the flight time, t, to the mass-to-charge ratio, m/z. Usually, the quadratic form will be used. In that case, the coefficients for the linear term should be miniscule, and the t 2 term should dominate. Therefore, Eq. 10 can be applied. If the cubic calibration curve is employed, the coefficient for the t 3 term will be small. Therefore it will still be appropriate to use Eq. 10 to calculate the error in m/z, given that you have a good estimate of the uncertainty, 1 Ct, in the measured centroid of the peak in the time domain Centroid and Gross or Net Area Readout The FASTFLIGHT software provides for reading the gross area, the net area above background and the centroid of a selected peak. To select a peak, click and drag the mouse from one side of the peak to the other, and release. The C/Trend Info and TOF Info tabs on the utility bar will display the net area and centroid for that peak. The centroid and gross or net area can be 49

60 FASTFLIGHT 2 Digital Signal Averager calculated and displayed for any peak in the chromatograph or the TOF-MS spectrum (see Section 8.3). The net area for the peak uses the first and last data points in the swept region to make a linear interpolation of the background to be subtracted from the peak region. This method of interpolation is sufficiently accurate on peaks that exhibit an amplitude that is large compared to the statistical fluctuations in the surrounding background. When the fluctuations in the background are a significant fraction of the peak s amplitude, the statistical uncertainty in the background estimate limits the precision in determining the net peak area. This error can be minimized by using a larger number of data points on either side of the peak to estimate the background under the peak. 16 If the number of points selected for integration of the peak area is n p, then at least n p points should be averaged on each side of the peak in order to estimate the background under the peak by linear interpolation. The Gross Area function on the C/Trend and TOF Info tabs can be used to measure the areas in the background regions on either side of the peak. From these left and right background areas, the background under the peak can be estimated and subtracted manually from the gross peak area. This method will deliver lower detection limits than the Net Area function on the two utility bar tabs. 16 Ron Jenkins, R.W. Gould, and Dale Gedcke, Quantitative X-Ray Spectrometry, Marcel Dekker, New York, 1981, Section

61 4. FASTFLIGHT-2 STARTUP AND SCREEN FEATURES To start the FASTFLIGHT-2 s user interface software, double-click on the FF2 desktop icon; or click on Start on the Windows Taskbar, then Programs, and FASTFLIGHT-2. If only one FASTFLIGHT-2 unit is connected to the PC and powered on, it will communicate with the software and a blank spectrum window will automatically open. If two hardware units are connected for duet operation (see Section ), two blank spectrum windows will open (see Fig. 20). Figure 17 shows an example of the main display for FASTFLIGHT-2 in the Chromatograph/ Trend mode with a single hardware unit attached. Fig. 17. The Main FASTFLIGHT-2 Display. 51

62 FASTFLIGHT 2 Digital Signal Averager The major screen features include: 1. Title bar at the top of the screen. On the far right are the standard Windows minimize, maximize, and close buttons. Each of the spectrum windows also has a title bar. 2. Menu bar, immediately below the title bar, showing the commands that can be selected with the mouse and/or keyboard. The menu functions are discussed in detail in Section Toolbar, containing speed buttons for frequently used functions including data acquisition, zooming, and histogram-smoothing. 4. Spectrum area, which displays the spectrum windows for up to two FASTFLIGHT-2 units. You can also use the File/Open command to open additional spectrum windows for viewing spectra stored on disk. The color scheme for the active and inactive windows depend on the desktop theme and colors you have selected in Windows Control Panel. To switch windows, click once inside the window you want, use the open-window list on the Window menu, or cycle through using <Shift + Tab>. Each spectrum window can possess the features discussed in items 5, 6, and 7 below. The right-mouse-button menus for the Chromatograph/Trend section and for the TOF section are discussed in Chapter 7, including a graph properties dialog that lets you control the appearance of the histogram features. 5. Chromatograph/Trend section (not displayed in TOFonly mode), allows you to view total- and/or specific-ion chromatographs. Pausing the mouse over a point in this part of the window pops up the box shown in Fig. 18. This information is also displayed in the All Experiment Properties dialog (see Fig. 45 on page 80). 6. TOF spectrum section (displayed in both acquisition modes). Note the scroll bar beneath the spectrum; it allows you to scroll through the entire spectrum at any X-axis scaling factor. Pausing the mouse over a point in this part of the window pops up the box shown in Fig. 19. Fig. 18. The Chromatograph/Trend Pop-Up Box. Fig. 19. The TOF Pop- Up Box. 7. Each section of the spectrum window contains a vertical marker line. It is used to mark spectrum regions for analysis. 8. Utility bar, which includes tabs (subscreens) for data acquisition, TOF-MS and chromatograph data analysis information, and calibration. Click on the tab name or press 52

63 4. STARTING FASTFLIGHT-2 <Ctrl + Tab> to switch among the subscreens. This sidebar can be undocked from the side of the screen and moved to another part of the screen (see Section 4.2). 9. Status bar, at the bottom of the screen, displays the X and Y coordinates for the point nearest the marker line in the active window. Appendix B discusses the FASTFLIGHT-2 software architecture in more detail Spectrum Display Modes Single-Instrument and Duet Modes Figure 17 above shows the screen display when a single FASTFLIGHT-2 unit is attached to the host PC. For duet-mode operation, a spectrum window opens for each instrument, as shown in Fig. 20. Fig. 20. Two Spectrum Windows Open in Duet Mode. 53

64 FASTFLIGHT 2 Digital Signal Averager Note that the bottom spectrum window is the active one. Except for the Start and Stop commands on the Acquisition menu, all menu, toolbar, and utility bar interactions affect only the active window Calibrated vs. Uncalibrated TOF Spectra The Calibration Enabled command, on the TOF spectrum section s right-mouse-button menu, allows you to switch the TOF spectrum s X-axis units from the calibration units (m/z, etc.) to nanoseconds and back (Fig. 21), for instance, to check a particular time of flight. Fig. 21. Switching the X-Axis Between Calibrated and Uncalibrated Units. 54

65 4. STARTING FASTFLIGHT Displaying Total-Ion and Specific-Ion Chromatograph Data Use the Show Total Ion Count and Show Specific Ion Count commands on the right-mousebutton menu (Section 7.10) to display one or both chromatographs at once, as illustrated in Fig. 22. See Section for instructions on defining a Specific ion region, and Section 7.8 for instructions on adjusting the graph display properties for the two spectra. Fig. 22. Viewing the Total Ion Count and Specific Ion Count Simultaneously Effect of Data Compression Options on Spectrum Display The data Compression method selected on the Protocol Settings tab (see Section ) affects how a TOF spectrum is displayed. Figures 23, 24, and 25 respectively show the display for the Lossless, Lossy, and Only centroid and area modes. 55

66 FASTFLIGHT 2 Digital Signal Averager Fig. 23. Lossless Data Compression. Fig. 24. Lossy Data Compression. Fig. 25. Only Centroid and Area. 56

67 4. STARTING FASTFLIGHT Docking and Undocking the Tool Bars FASTFLIGHT-2 s utility bar, toolbar, and status bar can either be docked (locked in a fixed position) at the top, sides, or bottom of the application program window; or undocked (freefloating windows that can be moved with the mouse to different parts of the screen) Undocking The first step in undocking a tool bar is to grab it with the mouse. To do this, move the cursor just inside the border of the bar on any side (see Fig. 26), then click and hold the left mouse button. A black outline will appear around the bar when you have successfully grabbed it. You are now ready to drag it to another part of the window. NOTE The undock or grab zone is narrow (3 pixels wide), so you may have to experiment until you find it. Figures 27 and 28 show the utility bar being undocked, dragged toward the center of the screen, and dropped. Fig. 26. The Undock ( Grab ) Zone. Fig. 27. Dragging the Utility Bar s Outline Toward the Center of the Screen. Fig. 28. Undocked Utility Bar (graphs now occupy whole width of window). 57

68 FASTFLIGHT 2 Digital Signal Averager This changes the sidebar to a free-floating window that you can move by grabbing the title bar with the mouse, and the graph window expands to fill the bar s former docking spot. NOTE You must drag the bar s black outline far enough toward the center of the screen that the outline no longer overlaps the bar s docked position. If not, when you drop it, it will snap back into its dock. If a tool bar is moved entirely off-screen, or to return all bars to their default, docked positions, click on View/Dock All. 58

69 5. THE TOOLBAR The Toolbar provides convenient shortcuts to some of the most common FASTFLIGHT-2 menu functions. The first four buttons allow you to connect to additional FASTFLIGHT-2 units, open a chromatograph/tof data file, save a file, and copy a graph to the Windows Clipboard. The next three control data acquisition, and allow you to view the current acquisition properties. These are followed by six buttons that control graph scaling. The Zoom and Marker tools are used, respectively, to magnify a region of interest in a spectrum and choose a region for analysis. The last three buttons control the degree of data smoothing performed on the graph(s). The New button establishes a connection with and opens a blank spectrum window for the one or two attached FASTFLIGHT-2 units. This is the equivalent of selecting File/New from the menu bar. In single-unit mode, only one new spectrum window opens, as illustrated in Fig. 17. In duet mode, a new spectrum window opens for each hardware unit. The color scheme for the active and inactive windows depend on the PC s Windows desktop theme colors. The Open button retrieves an existing spectrum file into a spectrum window. This is the equivalent of selecting File/Open from the menu bar. Save copies the currently displayed spectrum to disk. It duplicates the File/Save command (as distinguished from File/Save TOF As...). Copy (graph data) to Clipboard copies a metafile of the graph in the active window to the Windows Clipboard. The Clipboard contents can then be pasted into graphics files, text processor documents, spreadsheets, etc. Unlike a bitmap screen capture, the metafile can be sized or stretched. This duplicates Edit/Copy Graph. The Start button begins data acquisition in the currently active spectrum window. This duplicates the Acquisition/Start command. The timestamp and spectrum number are reset to zero, and the protocol is set to its initial number. The Stop button stops data acquisition in the currently active spectrum window. This duplicates Acquisition/Stop. The Instrument Properties button allows you to view the Measurement Settings dialog for the currently active spectrum window. This is the equivalent of selecting Acquisition/ Instrument Properties from the menu bar. Autoscale reads the data set and adjusts the X and Y axes so the entire plot fills the maximum space available to it on-screen (scaling up a small graph and scaling down a too-big graph). 59

70 FASTFLIGHT 2 Digital Signal Averager Vertical Autoscale adjusts the Y axis so the currently displayed peaks fill the maximum vertical space available. Shorter increases the full-scale value of the vertical axis so the graph peaks appear shorter. Taller decreases the full-scale value of the vertical axis so the peaks appear taller. Narrower increases the horizontal full scale of the graph so the peaks appear narrower. Wider decreases the horizontal full scale of the graph so the peaks appear wider. Zoom and Marker switch the mouse pointer between zoom mode, which allows you to magnify an area of interest (see Section 5.2), and marker mode, which allows you to choose one or more points on a chromatograph or TOF spectrum for analysis (see Section 5.1). No Smooth, 3-Point Smooth, and 5-Point Smooth The raw data in TOF-MS spectra exhibit point-to-point fluctuations caused by ion and gain statistics (see Section 3.3.3). For convenience in viewing, the fluctuations in the data can be smoothed with either the 3-Point Smooth or 5-Point Smooth buttons on the toolbar. The effect of this smoothing is only in the data display; the stored data are not altered. The smoothing algorithm results in a slight increase in the width of a peak in the spectrum. 17 The five-point smooth is equivalent to performing two three-point smooths in succession The Marker Tool Use the marker tool to set a vertical marker line at any position in the chromatograph and TOF sections of a spectrum window, and to mark regions of interest for analysis. To switch to marker mode, click on the Marker toolbar button or right-click the mouse and choose Marker Tool Active from the right-mouse-button menu. The pointer will be arrowshaped. 17 Bevington, Philip R. and D. Keith Robinson, Data Reduction and Error Analysis for the Physical Sciences, McGraw-Hill, New York, 1969 and

71 5. THE TOOLBAR AND INFORMATION BAR Moving the Marker with the Mouse To position the marker, move the mouse pointer to the point of interest and click the left mouse button once. This will move the marker to the mouse position Marking a Peak with the Mouse To mark a peak for analysis, position the pointer at either the left or right side of the desired region. Press and hold the left mouse button, drag the mouse to the other end of the region, then release the button The Zoom Tool Use the Zoom tool to magnify an area of interest. In zoom mode, the mouse pointer is shaped like a magnifying glass instead of an arrow. To switch to zoom mode, click on the Zoom toolbar button, or right click the mouse and choose Zoom Tool Active from the right-mouse-button menu. To zoom in a region of interest, click and drag the mouse diagonally across the area to be magnified (see Fig. 29) to draw a rubber rectangle around it. Fig. 29. Zooming In on an Area of Interest (note that both axes change scale). 61

72 FASTFLIGHT 2 Digital Signal Averager When you release the mouse button, the graph axes will scale up to the approximate extent of the rubber rectangle, and the area of interest will enlarge accordingly. Use the Autoscale toolbar button to restore the graph to its original scaling. 62

73 6. MENU COMMANDS This chapter describes the FASTFLIGHT-2 menu functions and their associated dialogs. It also references other parts of the manual that will help you set up and run your experiments. The menu functions use the left mouse button and the keyboard. Note that the underlined letter in the menu item indicates the key you can use together with the <Alt> key for keyboard access to that function. 18 The accelerator for each menu is the first letter of the menu name. (So, for example, the Start command under Acquisition can be reached by typing: <Alt + A>, <Alt + S>.) An ellipsis (...) after a menu item means that a dialog will open for you to complete. FASTFLIGHT-2 contains six menus: File, Edit, View, Acquisition, Window, and Help. Figure 30 is a composite of their commands. As mentioned in the previous chapter, many of these commands can be performed from the toolbar. Most data acquisition and analysis commands are performed with the utility bar (see Chapter 4.1.4). Fig. 30. Composite Illustration of the Five FASTFLIGHT-2 Menus File The functions on this menu allow you to establish communication with one or two FASTFLIGHT-2 hardware units, open existing FASTFLIGHT-2 data files; save the current chromatograph and/or TOF-MS data to disk; save the current TOF-MS spectrum with a different calibration file; set up, preview, and print the spectrum in the active window; and exit the user application. 18 Depending on the Windows appearance Effect in use on your PC, the underlines will be visible at all times or will only be displayed when you press and hold the <Alt> key. 63

74 FASTFLIGHT 2 Digital Signal Averager New This command communicates with the attached FASTFLIGHT-2 instrument(s) and opens a spectrum window for each instrument found. If only one FASTFLIGHT-2 is connected to the PC and powered on, one blank spectrum window will automatically open. If two units are connected for duet operation (refer to Section ), two blank spectrum windows will open, as illustrated in Fig. 20 on page Open... and Save... FASTFLIGHT-2 uses standard, Windowstype file-recall and file-save dialogs (Fig. 31 shows a typical file-save dialog). These include a Look in or Save in box that lets you specify the drive and pathname, a list-of-files box, a File name box, and a Files of type box. To recall an existing file, double-click on its filename in the list-of-files box; or enter its filename in the File name field, Fig. 31. Standard File-Save Dialog. then press <Enter> or click on Open. In some cases, you will be saving new data for which no file exists yet. To do this, enter the new filename in the File name: field and click on press <Enter> or click on Save. The Save As dialog also allows you to reuse an existing filename by saving new data into an existing file, completely overwriting the previous data. To do this, double-click on a filename from the list-of-files box or enter one of those existing filenames into the File name: field, then press <Enter> or click on Save. The system will display a message saying, This file already exists. Replace existing file? Click on Yes to save the new data or No to cancel the Save As operation Changing Drive and Pathname There are two ways to change to another directory and/or drive: click on the Look in/save in field to open a drop-down list of all drives and subdirectories connected to your PC (see Fig. 32); or click on the Up One Level button (just right of the Look in/save in field) to move to higher-level directories. In both cases, movement through the drives and directories is similar to using Windows Explorer. Fig. 32. Changing Drive and Pathname with the Drop-Down List. 64

75 6. MENU COMMANDS FASTFLIGHT-2 File Types FASTFLIGHT-2 creates three different types of spectrum data files:.flt2.fft2 Created in the TOF-MS-Only mode, and contains a single TOF spectrum. Created in Chromatograph/Trend mode, and contains the real-time, streamed data (this file can occupy several gigabytes of disk space). A corresponding.ffc2 file (same name, different extension) is created in the same directory at the same time. Both the.fft2 and.ffc2 file must be stored in the same subdirectory..ffc2 Created in Chromatograph/Trend mode, and contains the roadmap to each of >18,000 TOF spectra in the corresponding.fft2 chromatograph file. The number of TOF spectra per chromatograph is limited only by the PC s memory resources and disk storage space. When you retrieve an.ffc2 file, FASTFLIGHT-2 simultaneously opens the corresponding.fft2 file. Both the.fft2 and.ffc2 file must be stored in the same subdirectory. For detailed information on file formats, including the raw data formats for TOF and Chromatograph/Trend experiments, see Appendix B Save TOF As... Save TOF As... allows you to save a specific TOF spectrum (as an.flt2 file) under several conditions: When running in the TOF-MS-Only mode. When viewing a TOF-MS spectrum in the Chromatograph/Trend mode and you wish to save a particular TOF-MS spectrum for later reference. When performing the Sum/Average TOFs command (Section 7.3) and you wish to save the averaged TOF spectrum for future use. To apply a different calibration to the TOF-MS spectrum and save it for future use. When you issue this command, the file-save dialog shown in Fig. 33 opens. Enter a filename. If a calibration file is already associated with this spectrum, its filename will be listed in the Calibration File field. If a calibration file is not associated, Calibration File will be blank. To associate a calibration file with this spectrum, click on the browse button ( ) to the right of Calibration File. This will open a second dialog, shown in Fig. 34. Select a calibration file and click on Open to return to the preceding dialog. Click on Save to finish. 65

76 FASTFLIGHT 2 Digital Signal Averager Fig. 33. Enter a Spectrum Filename in the First Dialog and the Calibration File in the Second. Fig. 34. Select a Calibration File in the Second Dialog Print, Print Preview, and Print Setup These are standard Windows printer functions. Consult your Microsoft documentation if necessary Exit This exits the FASTFLIGHT-2 program. 66

77 6. MENU COMMANDS 6.2. Edit The Copy Graph command copies a metafile of the spectrum in the active window to the Windows Clipboard. The Clipboard contents can then be pasted into graphics files, text processor documents, spreadsheets, etc. Unlike a bitmap screen capture, the metafile can be sized or stretched. This duplicates the Copy to Clipboard toolbar button View Toolbar, Status Bar, Utility Bar, and Info Bar are toggles for displaying or hiding the various sidebars. Click once on the menu item to hide or display it; click again to reverse the toggle. Use Split in Chromatograph/Trend mode to adjust the relative proportion of screen occupied by the Chromatograph and TOF graphs. This operates like the Split command in Windows Explorer. You can also position the mouse pointer on the split divider until it changes to a twosided arrow ( ), then drag the split divider up and down on the screen. Dock All returns all sidebars and toolbars to their default positions Acquisition Use the items on this menu to switch between the TOF and Chromatograph/Trend modes; start and stop data acquisition, and access the FASTFLIGHT-2 Measurement Settings dialog. You can also switch between TOF and Chromatograph/Trend modes from the Acquisition tab of the utility bar Start and Stop The Start command begins data acquisition in the currently active spectrum window. This duplicates the Start button on the toolbar. The timestamp and spectrum number are reset to zero, and the protocol is set to its initial number. The Stop command stops data acquisition in the currently active spectrum window. This duplicates the Stop toolbar button. 67

78 FASTFLIGHT 2 Digital Signal Averager Instrument Properties General Settings Tab Figure 35 shows the General Settings tab of the FASTFLIGHT-2 Measurement Settings dialog. The Protocol Settings tab, discussed in the next section, allows you to specify up to 16 acquisition protocols. The Max Time and Max Spectra checkboxes allow you to limit the duration of data acquisition by time or by the number of spectra per chromatograph, respectively. Max Time can range from 1 to s (18 hr); Max Spectra is limited only by the host PC s memory resources and disk storage space. To calculate the approximate time required to accumulate a specified number of spectra in the Chromatograph/Trend mode: 1. Add the Time Offset, the Record Length (in µs), and the end-of-scan deadtime (0.8 µs) to get the time per scan. (You will subsequently enter Time Offset and Record Length on the Protocol Settings tab per Section ) 2. Multiply the time per scan by the number of scans per record (see Section C.1), the Number of Records per Spectrum (which you will enter on the Protocol Settings tab), and the number of spectra per chromatograph, Max Spectra. The product of these factors will be an approximate value for the time required to acquire the number of spectra entered in the Max Spectra field. Enter this value as the Max Time. If the RAPID PROTOCOL Port is being used, circa 10 µs will be added each time the protocol is switched. In the TOF mode, the Max Time will stop data acquisition after the specified amount of time has elapsed. The Max Spectra value will also stop acquisition if that number is surpassed before the Max Time is reached. In the TOF mode, the computer intercepts and displays TOF spectra at a rate of about 2 per second, whereas the FASTFLIGHT-2 hardware can be acquiring spectra at a much higher rate. Consequently, the number of TOF spectra that have been displayed before acquisition stops will be a fraction of the number selected in the Max Spectra field. If this acquisition will use a single protocol, select the Active protocol from the droplist. If using the RAPID PROTOCOL port, mark the Rapid protocol box (and refer to Section for in-depth information on this feature and its associated software settings). This will change the Active protocol field label to Initial protocol. Choose the first protocol to be used in the acquisition. In the Triggers section, click on the appropriate Trigger Enable Polarity radio button. The recommended default value, +(Positive), implies that a high TTL level enables trigger generation, while a low TTL level disables triggering. (See Section for a detailed discussion of the trigger circuits.) 68

79 6. MENU COMMANDS Fig. 35. Instrument Properties, General Tab. If this FASTFLIGHT-2 must generate a trigger-out, mark the appropriate radio button and enter a Trigger output width between 64 to 5120 ns in 32-ns steps. The default value on power-up is 1 µs. If this unit will accept a trigger-in signal, mark the radio button and select the Trigger-In slope from the droplist, then enter a Trigger-In threshold between 2.5 V and +2.5 V. The final parameter on this tab is the Chromatograph data folder, where the acquisition data will be streamed. Click on the browse button (...) and navigate to the desired location, then click on OK to return to the General Settings dialog Protocol Settings Tab FASTFLIGHT-2 allows you to predetermine up to 16 protocols, each of which specifies all the selectable operating parameters for the DSA. These protocols can be used in two ways: 69

80 FASTFLIGHT 2 Digital Signal Averager 1. To store a portfolio of frequently used operating conditions from which the desired one can be quickly selected at any time. 2. To provide the suite of operating conditions that FASTFLIGHT-2 will cycle through during a chromatograph/tof-ms run. This requires manipulation of the RAPID PROTOCOL port. To use option (1), simply select the protocol number you wish to use from the Active protocol list on the General Settings tab. Make any adjustments you want to the parameters, click on OK, and you re ready to acquire data. Implementing option (2) requires special software and hardware to change the operating conditions of the TOF-MS in concert with the cycling protocols of FASTFLIGHT-2. For example, a chromatograph/tof-ms run could be acquired by alternating the electrospray nozzle voltage on successive TOF spectra between in protocol 1 the settings that cause fragmentation of the molecules and in protocol 2 settings that cause no fragmentation. The computer controlling the nozzle voltage would select the appropriate protocol in FASTFLIGHT-2 via the RAPID PROTOCOL port (see Sections and C.1). FASTFLIGHT-2 would store the resulting TOF-MS spectra as two separate chromatograph spectra corresponding to protocol 1 and protocol 2. Thus, both the fragmentation and precursorion chromatographs could be acquired in a single chromatograph run. Figure 36 shows the entries for protocol 6. See the specifications in Section C.1, and the discussion in Sections and for more specific details and the allowed ranges of the parameters. The software will not allow an invalid value to be entered. Except where specified otherwise, the values illustrated in Figs. 35, 36, and 37 can be used as default values. The browse (...) button beside the Protocol number field opens the All Experiment Properties dialog (Section 7.9) as a quick reference of all current measurement settings. The Record length sets the duration (in µs) of the TOF spectrum to be recorded. Following each Trigger Output pulse, a delay specified by the Time offset elapses before recording of the TOF spectrum commences. Time offset can range from 0 to µs in 16-ns steps. Records per spectrum determines the number of records that will be summed to form a spectrum that is sent to the computer for viewing and storing. The Sampling interval specifies the horizontal distance between points in the TOF spectrum. Note that this parameter affects the Scans per Record (Section C.1) and limits the Record length. The Vertical offset (in volts) specifies the dc voltage added at the input to the ADC (see Fig. 6). It establishes the baseline for the spectrum. A more negative number moves the spectrum up in 70

81 6. MENU COMMANDS Fig. 36. The Protocol Settings Tab for Protocol 6. the displayed spectrum. This value can range from 0.25 V to V in 0.03-mV steps. A value of 0.05 V is recommended (see Section 3.1.7). If the TOF spectrum is intended to be recorded with the horizontal scale calibrated (typically in units of m/z), enter the path and filename of the Calibration file to be used (see Section 8.2). If no calibration is desired, leave this field blank. The Precision enhancer box is marked ( on ) by default. However, this is the place to turn it off by unmarking the box if you wish to examine the degradation of digital resolution and differential nonlinearity with the Precision Enhancer turned off (see the discussion in Section 3.1.7). Establishing a Specific ion region will enable you to simultaneously collect data for and view the total-ion and specific-ion chromatographs (for an example of this dual display, see Fig. 22 on 71

82 FASTFLIGHT 2 Digital Signal Averager page 55). Determining the boundaries of the region will require you to collect a preliminary chromatograph or use the File/Open command to recall a suitable chromatograph already saved to disk. Locate the peak of interest in the Chromatograph/Trend pane. Click on one of the points near the center of that peak to display the corresponding TOF-MS spectrum in the lower pane. In the TOF spectrum, find a mass peak that definitely identifies the molecule you wish to map in the chromatograph. If the spectrum is calibrated, you will have to turn off the calibration to obtain the timestamp of the start position. To do this, right-click in the TOF spectrum section and click to unmark Calibration Enabled. Note the desired start position and duration. Go to the Protocol Settings tab under Instrument Properties and enter the Start position and Length fields. There are three Data compression options, Lossless, Lossy, and Only Centroid and Area; these are discussed in detail in Section 3.2. Normally, the Correlated noise subtraction feature should be turned on by marking its associated box. The Noise sensitivity multiplier field allows you to choose the sensitivity factor for finding small peaks above the fluctuations in the background. This can be set from 2 to 4. A sensitivity of 2 will find the smallest peaks, but it will also falsely identify some background regions as peaks. A sensitivity of 4 will completely ignore background excursions, but will miss the smallest peaks. A sensitivity of 3 is recommended for general use. The function of the Minimum noise threshold is elucidated on page 41. The recommended default value of 10 means that the peak must have an amplitude that is more than 10 LSB above the local average of the background in the summed spectrum in order to be detected as a peak. This Minimum noise threshold is operative whether or not the Correlated noise subtraction is turned on. To disable the Minimum noise threshold, enter zero. Clicking on the Advanced button opens the dialog in Fig. 37. With the recommended default values displayed in this figure, Background Sampling Interval = 200 bins Adjacent Background = 16 bins Maximum Peak Width = 400 bins Minimum Peak Width = 4 bins Ringing Protection = 2 ns the data compression feature works as follows. 72

83 6. MENU COMMANDS In suppressed data regions identified as background, a Background Sampling Interval of 200 means that up to 200 adjacent background points will be deleted between each group of 4 contiguous background points that are preserved. The periodic groups of 4 background points will supply a sketch of the background between peaks. An Adjacent Background setting of 16 indicates that 16 background data points to the left and right of an identified peak will be preserved in the compressed spectrum to permit an accurate background subtraction when analyzing the spectrum. Fig. 37. Advanced Compression Settings. A Maximum Peak Width value of 400 ensures that a background hump more than 400 bins wide will not be identified as a peak. Beyond the 400 points, the background averager, automatic noise suppressor, and correlated noise detection are turned on again. The Minimum Peak Width ensures that the specified number of contiguous data points must exceed the background threshold in order to be identified as a peak. The Minimum Peak Width should be set slightly less than the number of bins in the FWHM of the narrowest peak in the TOF spectrum. Some detectors produce noticeable ringing on the trailing edges of fast pulses. This shows up as damped oscillation on the right side of peaks in the spectrum. The ringing can interfere with estimating the background in the data compression scheme. Consequently, the Ringing Protection should be entered as the number of nanoseconds required for the ringing to subside to a negligible amplitude Window This menu contains standard Windows functions for displaying and switching between multiple. Consult your Microsoft documentation if necessary. 73

84 FASTFLIGHT 2 Digital Signal Averager Figure 38 shows the About box, which contains software version information that might be useful should you need customer service Help Fig. 38. About FASTFLIGHT-2. 74

85 7. THE RIGHT-MOUSE-BUTTON MENUS This chapter tells you how to use the right-mouse-button menus for the TOF window (Fig. 39) and the Chromatograph/Trend window (Fig. 40). Both menus give you control over FASTFLIGHT-2 s graph parameters, axes, colors, fonts, legends, and symbols; and let you switch between the zoom and marker tools. In the TOF window, you can also activate or inactivate the calibration, and add or remove data overlays. And in the Chromatograph/Trend window you can monitor which protocol is currently in use, and average a group of TOFs in a peak to improve the signal-to-noise ratio. To display the right-mouse-button menu, move the pointer to the window you want to work with and right-click once. When the menu opens, click on the command you want. Fig. 39. TOF Window Right-Mouse-Button Menu. Fig. 40. Chromatograph/Trend Window Right-Mouse-Button Menu (showing View Protocol submenu) Calibration Enabled This command turns the current calibration off and on (Fig. 41) so you can switch the TOF spectrum s X-axis units from the calibration units (m/z, etc.) to nanoseconds and back, for instance, to check a particular time of flight so you can set up the Specific-ion region on the Protocol Settings tab, as discussed in Section

86 FASTFLIGHT 2 Digital Signal Averager Fig. 41. Switching the X-Axis Between Calibrated and Uncalibrated Units View Protocol This command is used in conjunction with chromatographs collected using the Rapid Protocol Selection feature (Section 6.4.2). The RAPID PROTOCOL port is normally used to alternate data acquisition between those conditions that yield the precursor-ion spectrum and those that produce the fragmented-ion spectrum. Since the two conditions typically alternate at 10- to 100-ms intervals, a means for separating and viewing the interlaced spectra is required. FASTFLIGHT-2 keeps track of the spectra collected under the two conditions via two different protocol numbers. The View

87 7. THE RIGHT-MOUSE-BUTTON MENUS Protocol submenu allows you to display the chromatograph and TOF-MS spectra that belong to a particular protocol. In other words, it permits you to view the chromatograph/trend and TOF-MS spectra for the precursor ions separately from the chromatograph/trend and TOF-MS spectra for the fragmented ions. When you open a multi-protocol chromatograph file, FASTFLIGHT-2 reads which protocols were used during data collection. When you click on View Protocol, the submenu shown in Fig. 40 opens. Only the protocols that were used in the experiment are active (shown in black). A bullet () marks the protocol for the spectrum currently onscreen. To view the spectrum for one of the other active protocols, simply select it from the submenu Sum/Average TOFs As explained in Section 3.3.3, the dominant error in the TOF-MS spectrum is the random error caused by ion and gain statistics. When operating in Chromatograph/Trend mode, this error can be reduced by summing and averaging together a large number of TOF-MS spectra. To do this, choose a peak on the chromatograph. Click and drag the mouse across the selected peak to mark it. Right-click on the chromatograph to open its right-mouse-button menu, then select Sum/Average TOFs. The software will average all the TOF-MS spectra in the selected region of the chromatograph peak and display the result below in the TOF-MS spectrum. You can observe the improvement in signal-to-noise ratio in the TOF-MS spectrum display as each additional TOF-MS spectrum is averaged with the previous TOF-MS spectra in the selected region. The averaging progresses from left to right in the region selected in the chromatograph. When two peaks partially overlap in the chromatograph, sum and average the left side of the left overlapping peak to minimize interference from the peak on the right side, and sum/average the right side of the right overlapping peak to minimize interference from the left peak. The summed and averaged TOF-MS spectrum can be saved with a unique file name using File/Save As... The calibration file (if applicable) can be saved with it (or you can associate a different calibration file with it), along with the conditions for the pertinent acquisition protocol Zoom Tool Active/Marker Tool Active Click on one of these items to activate that tool (see also the discussions in Sections 5.1 and 5.2). 77

88 FASTFLIGHT 2 Digital Signal Averager 7.5. Add Overlay... This command allows you to compare two or more spectra to see the differences between them. Usually, the acquisition conditions will have changed slightly between the two spectra, and the operator will want to see the difference between the two spectra. Add Overlay opens a standard file-open dialog. Choose a comparison spectrum and click on OK. You may have one to four overlays open at one time. Each overlay s graph properties are assigned in Section Figure 42 shows a TOF spectrum with two graphs overlaid on it. Fig. 42. Two TOF Files Overlaid on the Current Spectrum. The following parameters must be the same in both spectra before they can be compared with Add Overlay...: Sampling interval Time offset Record length Calibration 78

89 7. THE RIGHT-MOUSE-BUTTON MENUS If the two files are inconsistent with respect to one or more of these parameters, a dialog will open that tells you the problem(s). If any of the first three parameters are mismatched, choose another file for the overlay. If only the calibration is inconsistent, use File/Save TOF As... to change the calibration file(s) as needed. This removes the most recently applied overlay Remove Overlay 7.7. Export data to file... This command exports the current spectrum s X- and Y-axis coordinates to an ASCII file in comma-delimited columns Graph Properties... Graph Properties... (Fig. 43) lets you modify the active graph s title, symbol shape and size, lines and grids, and axis scale and labels. In addition, you can set up the colors and symbols for up to four data-set overlays. Overlay 1 is also used to assign the graph properties for the specific ion spectrum displayed with the Show Specific Ion Count command (Section 7.10). Fig. 43. Graph Properties Dialog. 79

90 FASTFLIGHT 2 Digital Signal Averager Changes take effect as soon as you make them (without closing the dialog). To close the dialog, click on OK Overlay properties... To set the properties for Overlays 1 4, click on Overlay properties... to open the dialog shown in Fig. 44. Note that Overlay 1 is also used to assign the graph properties for the specific ion spectrum displayed with the Show Specific Ion Count command (Section 7.10). Fig. 44. Overlay Properties Dialog. Changes take effect as soon as you make them (without closing the dialog) Experiment Properties... This command opens the dialog shown in Fig. 45. It lists the general settings and all 16 protocol settings for the spectrum in the currently active window (from a disk file or connected FAST- FLIGHT 2). The bottom section of the dialog, Spectrum properties, shows the spectrum parameters for the TOF spectrum region currently selected with the Marker Tool Show Total Ion Count and Show Specific Ion Count These commands allow you to display the standard chromatograph histogram (Show Total Ion Count) and/or the histogram for a particular ion (Show Specific Ion Count), as set up on the Protocol Settings tab under Instrument Properties. See Fig. 22 on page

91 7. THE RIGHT-MOUSE-BUTTON MENUS Fig. 45. All Experiment Properties. 81

92 FASTFLIGHT 2 Digital Signal Averager 82

93 8.1. Acquisition Tab Use the Acquisition tab (Fig. 46) to select the TOF or chromatograph/ trend mode, track acquisition time and the number of spectra stored, and monitor the hardware status via the LED indicators at the bottom of the tab. The upper indicator is bright green when processing data and dull green when acquisition is stopped. The lower, flashing yellow indicator is displayed only for ADC under- and overflows. If both occur simultaneously, the indicator is labeled Overflow Calibration Tab 8. THE UTILITY BAR Although the natural horizontal axis of the DSA is time-of-flight in nanoseconds, the desired information is mass divided by the number of charges on the molecule, m/z. The Calibration tab allows you to calibrate the X axis to read in units of m/z. Linear, quadratic, or cubic calibration curves can be selected for a least-squares fit to calibration data. You can create an unlimited number of calibration files, assign a calibration to each of the eight data acquisition protocols if you wish, and save calibration data as part of the spectrum file. The calibration can be changed or updated by specifying a different calibration file to be used with the spectrum. Fig. 46. Utility Bar, Acquisition Tab Introduction From simple theory, the m/z value is expected to be proportional to the square of the flight time. Consequently, the quadratic calibration curve is normally used in TOF-MS. (m/z) a bt ct 2 (11) Second-order effects in the TOF-MS can sometimes be accommodated by stepping up to a cubic calibration curve. Over extremely small ranges of m/z, the quadratic curve can be approximated as a linear curve. Although it requires only three calibration points to uniquely define the quadratic curve, it is better to use many more than three points in a least-squares fit to the data (see footnote 17 on page 60). The overdetermination inherent in the least-squares fit averages the random errors in the 83

94 FASTFLIGHT 2 Digital Signal Averager calibration points. Furthermore, the residual differences of the data points from the calibration curve can be used to estimate the random error in the measurement (see footnote 17). Generating a calibration file requires a TOF-MS spectrum with accurately known m/z values. To minimize random errors, the TOF-MS spectrum should be summed over as many records as possible. Only the larger peaks should be selected for calibration points, because the smaller peaks will have larger random errors. Creating a calibration file can be painstaking. Therefore, FASTFLIGHT-2 includes tools that will help keep your calibration files valid. Once you have created a calibration file, FASTFLIGHT-2 allows you to delete points from it and correct errors. Also, the software includes an Update item(s) command that lets you quickly update a calibration file each time you re-run the calibration standard. The calibration procedure allows you to determine the start and end points for each peak, including how much or how little of the adjacent background is encompassed on either side of a calibration peak. This lets you decide how much or how little peak shift will be allowed between calibration runs before the calibration procedure must be repeated. Before deciding on the width of calibration peaks, you may wish to consider the effect of peak width on minimum detection limit (see footnote 16 on page 50) and the effect of peak shift on centroid (hence mass) determination. You may also find it useful to read how the Update item(s) command works (see Section 8.2.4) Creating a Calibration File 1. Open a TOF-MS spectrum for which the m/z values are accurately known. If a calibration file is already associated with this spectrum, the Calibrate tab will display the calibration filename and list the peak values. Click on the New button to clear all fields on the tab (see Fig. 47). 2. Click on the Calibration Type field to open its drop-down list, and choose whether this fit will be Linear (requiring a minimum of three points), Quadratic (four points minimum), or Cubic (five points minimum). Then enter the Units label for the calibrated X axis. 3. Using the Zoom Tool, zoom in on a peak (see Fig. 48). 4. With the Marker Tool, mark the peak (see Fig. 49). Fig. 47. Utility Bar, Calibration Tab. 84

95 8. THE UTILITY BAR Fig. 48. Zoom In on a Peak. 5. Click on the Add button to open the dialog shown in Fig. 49. Enter the m/z value for this flight time and click on OK. This value will now be displayed on the Calibrate tab. 6. Repeat steps 3 5 for the remaining peaks in the spectrum. 85

96 FASTFLIGHT 2 Digital Signal Averager Fig. 49. Mark the Peak, Click on the Add Button, and Enter the Calibration Value. To check the goodness of fit of your calibration points, click on View to display the dialog shown in Fig. 50. Note that it shows both a graph and the chi-square for the calibration points on the list. At this point, you are ready to use the completed Value/Centroid list to create a calibration file. Click on the Save button on the Calibration tab. A standard file-save dialog will open, allowing you to enter a filename and complete the save. To use this calibration file, display the TOF spectrum you wish to calibrate, click on File/Save TOF As..., enter a filename, select this new calibration file, and complete the save operation. Figure 51 shows a calibrated TOF spectrum and the corresponding Calibration tab. 86

97 8. THE UTILITY BAR Fig. 50. Click on View Button to Check Goodness of Fit of Calibration (note chi-square calculation at lower left) Editing a Calibration File Click on the Open... button to display a standard file-open dialog. Select the calibration file you wish to edit. Its values will be loaded into the Value/Centroid table. To delete a point, click once to highlight it, right-click to open the right-mouse-button menu (see Fig. 52), and select Delete item(s). To change the calibration value for a point, click once to highlight it, delete it, and add it again. The Value/Centroid table on the utility bar will update immediately. To save the changes to the calibration file, click on Save... and either save to the original filename (which will overwrite the old file) or enter a new filename. 87

98 FASTFLIGHT 2 Digital Signal Averager Fig. 51. Calibrated TOF Spectrum with Calibration Displayed on Utility Bar Updating an Existing Calibration This command allows you to quickly update the calibration for a known standard instead of having to repeat the formal calibration procedure in Section Introduction When you create a calibration file, FASTFLIGHT-2 stores the start and end points of each peak (in ns), the peak centroid, and the m/z value you assign to that centroid. The Update item(s) command reimposes the original start and end points for each selected peak onto your new calibration standard spectrum, and recalculates the centroid for each peak region. The new centroid values modify the coefficients used in the fit. As long as the position of the peaks in this new calibration run remain within the start and end points you chose for the original calibration s peaks, there is no need to repeat the formal calibration. 88

99 8. THE UTILITY BAR FASTFLIGHT-2 s calibration procedure allows you to determine the start and end points for each peak, including how much or how little of the adjacent background is encompassed on either side of a calibration peak. This, in effect, determines how much or how little peak shift you will allow between calibration runs before deciding to repeat the calibration procedure Performing the Update 1. Select all of the points on the Value/Centroid list: click on the first point in the list, scroll down to the last point, then press and hold the <Shift> key while clicking on the last point. 2. Right-click to open the right-mouse-button menu, and select Update item(s). FASTFLIGHT-2 will recalculate the centroids and display them in the Value/Centroid list. 3. Click on View to confirm that the updated calibration is satisfactory. If one or more of the new standard s peaks fall outside the range of the original calibration, recalibrate as in Section Fig. 52. Use the Calibration Tab to Update the Calibration File or Delete Points. 4. To keep this recalibration, click on Save... and either save to the original filename (which will overwrite the old file) or enter a new filename. To discard the recalibration, click on Open... and recall the starting calibration file Opening a Different Calibration File Click on the Open... button to display a standard file-open dialog. Select the calibration file you wish to load into the Value/Centroid table on the Calibration tab. To apply this calibration to a spectrum, use File/Save TOF As... (Section 6.1.3) C/Trend Info Tab Figure 53 shows the C/Trend Info tab. Note that a region has been marked on the accompanying chromatograph spectrum (see Section 3.3.6), and the corresponding Start and End points, Gross Area, Net Area, and Centroid are displayed. 89

100 FASTFLIGHT 2 Digital Signal Averager Fig. 53. C-Analysis Tab Showing Analysis of Region Marked on Chromatograph TOF Info Tab Figure 54 shows the TOF Info tab. The Start and End points, Gross Area, Net Area, and Centroid are displayed for the region marked on the accompanying TOF spectrum (see Section 3.3.6). Use the Experiment Properties... command on the TOF section s right-mouse-button to display a listing of the instrument properties for the TOF spectrum region currently selected with the Marker Tool (see Fig. 45). These properties are set with Acquisition/Instrument Properties..., Section Click on Export data to file... to export the comma-delimited list of X,Y coordinates for the current TOF spectrum to an ASCII (.TXT) file. A standard file-open dialog will allow you to assign a location and filename. 90

101 8. THE UTILITY BAR Fig. 54. TOF-Analysis Tab Showing Analysis of Region Marked on TOF Graph. 91

102 FASTFLIGHT 2 Digital Signal Averager 92

103 APPENDIX A. ACTIVEX METHODS AND PROPERTIES This appendix is provided for those who wish to write their own software for controlling FASTFLIGHT-2. The Instrument Operation ActiveX DLL and ActiveX Graphing Control offer the simplest means of interfacing custom software with the intricate and complicated details of operating the hardware and parsing the data FASTFLIGHT-2 delivers. Using the ActiveX methods and properties bypasses the need to grapple with the complexity of how the data are generated and saved. A.1. The Instrument Operation ActiveX DLL (FF2Ctrl.dll) The programmer will have full control of the FASTFLIGHT-2 hardware by using FF2Ctrl.dll. This DLL is an in-process ActiveX DLL, and is Microsoft COM compliant. It can be used in Microsoft Visual C++, Visual Basic, and.net environments. FF2Ctrl.dll is written in Unicode, and exposes four COM objects: FF2CtrlObj, the main controlling object. GSObj, the encapsulation of all general settings. ProtocolObj, the encapsulation of all protocol dependent settings. TOFObj, the encapsulation of the properties of a TOF spectrum. To use FF2Ctrl.dll in Visual Basic, you must first add the reference to your Visual Basic project. In the Visual Basic environment, click on the Project menu, then on References. When the list opens, choose FF2Ctrl 1.0 Type Library. You can now use the COM interface exposed by FF2Ctrl.dll. Normally, you will declare your object variable at the module (or form) level. For example, declare the following variable in the general section of your form, Dim m_obj As New FF2CTRLLib.FF2CtrlObj To use FF2Ctrl.dll in the Visual C++ environment, try the following code sample: //In the header file, may be MainFrm.h, include the reference #import " C:\Program Files\ORTEC\FF2 \ff2ctrl.dlll" named_guids using namespace FF2CTRLLib; Then, declare the variable in the CMainFrame class as below: 93

104 FASTFLIGHT 2 Digital Signal Averager IFF2CtrlObjPtr FF2Obj; HRESULT hr = FF2Obj.CreateInstance(CLSID_FF2CtrlObj); if(failed(hr)) { AfxMessageBox("Could not find the FF2 COM server"); return -1; // fail } Now you can access methods and properties of the object using syntax. For example: try{ long idevices = FF2Obj->Open(); if(idevices == 0){ AfxMessageBox("No FF2 found"); return -2; // FF2 not connected } } catch(_com_error& e) { AfxMessageBox(e.Description()); return -3; } Error Handling: All function calls to the method and properties could generate trappable errors if invalid values are provided or abnormal operations are encountered. The programmer can catch these errors by using try and catch syntax such as above in VC and using On Error GoTo syntax in VB. Duet Configuration: If more than one FASTFLIGHT-2 hardware unit is connected to the same computer, a separate instance of FF2CtrlObj properties will need to be created for each unit. Use DeviceCount to find out how many units are connected, and SerialNumber to document the serial number associated with each instance. The complete sample code can be found in the FASTFLIGHT-2 install folder \Sample Code\VC, \VB, and \.NET. A.1.1. FF2CtrlObj Properties Active Get current data acquisition status. Returns 1 if data acquisition is currently in progress. Returns 0 if data acquisition is stopped. Data type: short integer 94

105 APPENDIX A. ACTIVEX METHODS AND PROPERTIES DeviceCount Get the number of FASTFLIGHT-2 units attached through the USB port. Data type: integer FFVersion Get the firmware version of the FASTFLIGHT-2 unit. Data type: string IsInstrumentPresent Get the USB connection status of the instrument. 1 = USB connection exists between the FASTFLIGHT-2 and computer, and communication is working properly. 0 = USB connection is not working. Data type: integer Records Get number of records collected for current spectrum. Data type: integer SerialNumber Get the serial number of the FASTFLIGHT-2 unit. Data type: string Spectrums Get the number of spectra collected since the start of the acquisition. Data type: integer TimeElapsed Get the number of seconds since the start of the acquisition. Data type: double A.1.2. FF2CtrlObj Methods Close Close the connection to the FASTFLIGHT-2 hardware. No further communication is possible. Parameters None. Return Values None. 95

106 FASTFLIGHT 2 Digital Signal Averager Remarks The user will usually call this method before the termination of the application. CTOFGetCData(VARIANT* vaxdata, VARIANT* vaydata1, VARIANT* vaydata2, short ProtoNum, long* plvalidprotonums) Retrieves chromatograph spectrum from current C/TOF acquisition or a chromatograph file previously opened by calling OpenCTOFFile. Parameters vaxdata (output). Upon successful return of the function, the variant will contain a SAFEARRAY of doubles indicating time in units of seconds. Otherwise, vaxdata is empty. vaydata1 (output). Upon successful return of the function, the variant will contain a SAFEARRAY of long doubles indicating the amplitude of the spectrum (Y axis). Otherwise, vaydata is empty. ProtoNum (input) Protocol number. plvalidprotonums (output) designates which protocol numbers are valid (bit set to 1) and which are invalid (bit set to 0). The 16 protocol numbers (0 15) are assigned to bit positions in the most significant 16 bits of a 32-bit word. Protocol number 0 is assigned to the 31 st bit, protocol number 1 is associated with the 30 th bit, etc., down to protocol number 15 being denoted by the 16 th bit. Return Values 1 = Success. 0 = Failure. Remarks For example: Dim vax As Variant Dim vay1 As Variant Dim vay2 As Variant Dim ivalidprotnums As Long Dim MyGs As New FF2CTRLLib.GSObj m_obj.getgeneralsettings MyGs Dim igood As Long igood = m_obj.ctofgetcdata(vax, vay1, vay2, 0, MyGs.ActiveProtoNumber, ivalidprotnums) 96

107 APPENDIX A. ACTIVEX METHODS AND PROPERTIES CTOFGetOnePoint (iindex As Long, bisopenfile As Long, TOFObj As TOFObj) Retrieves a specific TOF spectrum properties from current C/TOF acquisition or a chromatograph file previously opened by calling OpenCTOFFile. Parameters Index (input) specifies the position (0-based index number) of the desired TOF spectrum in the chromatograph spectrum. bisopenfile (input) Specify the value of 0 if you wish to retrieve a specific TOF spectrum from current C/TOF acquisition. Specify 1 if you wish to retrieve a specific TOF spectrum from chromatograph file previously opened by calling OpenCTOFFile. TOFObj (output) contains the properties of the retrieved TOF spectrum. Return Values None. Remarks For example: Dim Tof As New TOFObj m_obj.ctofgetonepoint 12, FALSE, Tof CTOFGetTOFData(VARIANT* vaxdata, VARIANT* vaydata, long Index, short bisopenfile, short ProtoNum, ITOFObj* TOFObj) Retrieves a specific TOF spectrum from current C/TOF acquisition or a chromatograph file previously opened by calling OpenCTOFFile. Parameters vaxdata (output). Upon successful return of the function, the variant will contain a SAFEARRAY of doubles indicating the time stamps (in units of nanoseconds) of the spectrum (X axis). If the return is not successful, vaxdata retains old, irrelevant data. vaydata (output). Upon successful return of the function, the variant will contain a SAFEARRAY of doubles indicating the amplitude of the spectrum (Y axis). If the return is not successful, vaydata retains old, irrelevant data. Index (input) specifies the position (0-based index number) of the desired TOF spectrum in the chromatograph spectrum. bisopenfile (input) Specify the value of 0 if you wish to retrieve a specific TOF spectrum from current C/TOF acquisition. Specify 1 if you wish to retrieve a specific TOF spectrum from chromatograph file previously opened by calling OpenCTOFFile. ProtoNum (input) Protocol number. TOFObj (output) contains the properties of the retrieved TOF spectrum. 97

108 FASTFLIGHT 2 Digital Signal Averager Return Values 1 = Success. 0 = failure. Remarks If the spectrum is taken under lossless compression, the vaydata will contain all valid data. If it is taken under lossy or centroid and area only compression, the vaydata will contain some null data which is designated by the value 1.0E308. For example: Dim vax As Variant Dim vay As Variant Dim MyTof As New TOFObj Dim lgood As Long lgood = m_obj.ctofgettofdata(vax, vay, 0, 20, 0, MyTof) The format of X and Y data for the TOF spectrum is the same as is described for the X and Y data in the.flt2 spectrum as described in Section B.1. GetGeneralSettings(IGSObj* GSObj ) Obtain the current FASTFLIGHT-2 general settings. Parameters GSObj (output) is the instance of a GSObj object which will contain the settings upon successful return of the function. Return Values None. Remarks For example: Dim MyGs As New FF2CTRLLib.GSObj m_obj.getgeneralsettings MyGs GetProtocol(short ProtNum, IProtocol* ProtObj) Obtain the protocol settings for a specified protocol number. The protocol settings are encapsulated in the Protocol object. 98

109 APPENDIX A. ACTIVEX METHODS AND PROPERTIES Parameters ProtNum (input).protocol number. Must be in the range of ProtObj (output). Instance of a protocol object that documents the settings. Return Values None. Remarks For example: Dim MyProt As New FF2CTRLLib.Protocol m_obj.getprotocol(1, MyProt) GetTOFData(VARIANT* vaxdata, VARIANT* vaydata, ITOFObj* TOFObj) Retrieves current TOF spectrum from current TOF acquisition in progress. Parameters vaxdata (output) Upon successful return of the function, the variant will contain a SAFEARRAY of doubles indicating the time stamp of the spectrum (X axis) in units of nanoseconds (ns). If the return is not successful, vaxdata retains old, irrelevant data. vaydata (output) Upon successful return of the function, the variant will contain a safearray of doubles indicating the amplitude of the spectrum (Y axis). If the return is not successful, vaydata retains old, irrelevant data. TOFObj (output) contains the properties of the retrieved spectrum. Return Values 1 = Success. 0 = Failure. Remarks If the spectrum is taken under lossless compression, the vaydata will contain all valid data. If the spectrum is taken under lossy or centroid-and-area-only compression, the vaydata will some contain null data which is designated by the value of 1.0E308. For example: Dim vax As Variant Dim vay As Variant Dim igood As Long Dim MyTof As New TOFObj igood = m_obj.gettofdata(vax, vay, MyTof) 99

110 FASTFLIGHT 2 Digital Signal Averager The format of X and Y data for the TOF spectrum is the same as is described for the X and Y data in the.flt2 spectrum as described in Section B.1. Open Opens the connection with the FASTFLIGHT-2 hardware for further communication with the hardware. Parameters None. Return Values Long integer. Returns 1 for success, 0 for failure. Remarks User will first call this function in order to establish communication with FASTFLIGHT-2 hardware, ideally at the start of an application. OpenCTOFFile(BSTR FileName) Opens chromatograph spectrum stored in an.ffc2 file. The file must have an.ffc2 extension. Parameters FileName (input). Specifies the file name and full path of the file. Return Values None. Remarks For example: m_obj.openctoffile "C:\test.FFC2" A trappable error will be generated by the wrong format or an invalid file name. ResetTimeStamp Reset time stamp clock in the FF2 hardware. Parameters None. 100

111 APPENDIX A. ACTIVEX METHODS AND PROPERTIES Return Values None. Remarks None. SaveCTOFDataToFile(BSTR newfilename) Save the current CTOF data to disk. During the CTOF mode data acquisition, the data are streamed to a temporary file, this function simply renames the temporary file. Parameters newfilename (input) Specifies the new file name, not including the path. The path can be obtained or set by the property ChromatFileFolder in the GSObj. Return Values 1 = Success. 0 = Failure. Remarks None. SetGeneralSettings (IGSObj* GSObj) Set the general settings used by all protocol numbers. Parameters GSObj (input). Instances of a GSObj object that contains the general settings. Return Values None. Remarks For example: Dim MyGs As New FF2CTRLLib.GSObj Get the settings first m_obj.getgeneralsettings MyGs MyGs.TimePreset = MyGs.ExtTriggerInputEnable = 0 MyGs.PresetFlags = 1 Now send to instrument m_obj.setgeneralsettings MyGs 101

112 FASTFLIGHT 2 Digital Signal Averager SetProtocol(short ProtNum, IProtocol* ProtObj) Set the protocol settings for a specified protocol number. Parameters ProtNum (input). Protocol number. Must be in the range of ProtObj (input). Instances of a protocol object that contains the settings. Return Values None. Remarks For example: Dim MyProt As New FF2CTRLLib.Protocol Get the protocol settings first m_obj.getprotocol 1, MyProt MyProt.Name = My protocol One MyProt.RecordLength = MyProt.RecordsPerSpectrum = 30 m_obj.setprotocol 1, MyProt Start (short Mode, Long bresetts) Start a new acquisition process. Parameters Mode (input) specifies the mode of the data acquisition. 0 = TOF mode, 1 = Chromatograph mode. bresetts (input) gives you choice in resetting the time stamp before the start of an acquisition; 1 = reset time stamp, 0 = do not reset time stamp. Return Values None. Remarks None. Stop Stop an acquisition in progress. Parameters None. 102

113 APPENDIX A. ACTIVEX METHODS AND PROPERTIES Return Values None. Remarks None. A.1.3. GSObj Properties ActiveProtoNumber Set/Get the current protocol number. Data type: short ChromatFileFolder Set/Get the path for the temp file location in the chromatograph mode acquisition. Data type: string ExtTriggerInputEdge Set/Get the external trigger input edge. 0 = Rising edge; 1 = Falling edge. Data type: integer ExtTriggerInputEnable Set/Get the external trigger input enable flag. 0 = Disable. In this case the FASTFLIGHT-2 hardware will use internal trigger and output the Trigger Out signal; 1 = Enable. In this case the FASTFLIGHT-2 hardware will use the Trigger In signal. Data type: integer ExtTriggerInputThreshold Set/Get the external trigger input threshold in units of volts. Must be in the range of 2.5V to 2.5V, and in increments of 0.01 V. Data type: double PresetFlags Set/Get the data acquisition preset flag. The 0 th bit indicates time preset is on/off. 1 st bit indicates spectra preset is on/off. 1 = On, 0 = Off. Data type: integer Rps Set/Get the Rapid Protocol Selection flag. 0 = Disable; 1 = Enable. Data type: integer 103

114 FASTFLIGHT 2 Digital Signal Averager SpectrumPreset Set/Get the preset of number of spectra for a data acquisition. The PresetFlags property must also be enabled in order for this to take effect. Data type: integer TimePreset Set/Get the preset of data acquisition time in units of seconds. The PresetFlags property must also be enabled in order for this to take effect. Data type: double TriggerEnablePolarity Set/Get the TRIGGER ENABLE IN polarity assignment. A value of 1 implies a high TTL level is required to enable triggering; A value of 0 implies a low TTL level enables triggering. Data type: integer TriggerOutputWidth Set/Get the width of the output pulse in units of microseconds (µs). Must be within the range of 0.064µs to µs. Data type: double A.1.4. ProtocolObj Properties ProtocolObj contains all the protocol-dependent settings. AdjacentBkg Set/Get Adjacent Background (used in compression algorithm). Must be within the range of 1 to 96 bins; default is 32 bins. Data type: short BkgInterval Set/Get Background Sampling Interval (used in compression algorithm). Must be in range of 64 to 1024 bins, default 200 bins. BkgInterval is the number of background points that are deleted between groups of 4 sequential background points that are preserved. Data type: short CalibrationFile Set/Get calibration file name and full path. Data type: string 104

115 APPENDIX A. ACTIVEX METHODS AND PROPERTIES CompType Set/Get Compression type. 0 = Lossy compression: Peaks are preserved and background is decimated. 1 = Lossless compression (default): all data are transmitted. 2 = Stick Diagram: Centroid and net area are transmitted. Data type: short EnableCNS Set/Get Enable/Disable Correlated Noise Subtraction. 1 = Enable, 0 = Disable. Data type: integer MaximumPeak Set/Get Maximum Peak Width (used in compression algorithm). Must be within the range of 64 to 1024 bins, default is 400 bins. Data type: short MinimumPeak Set/Get Minimum Peak Width (used in compression algorithm). Must be within the range of 2 to 32 bins, default is 4 bins. Data type: short MinimumThreshold Set/Get Minimum noise threshold (used in compression algorithm). Must be within the range of 0 to Default is 10. Data type: long Name Set/Get Name of the protocol. Maximum length is 256 Unicode characters. Data type: string PrecEnhEnable Set/Get Precision Enhancer on/off. 1 = Enable (on); 0 = Disable (off). Default is Enable. Data type: short RecordLength Set/Get value of record length in units of microseconds (µs). Default is 100 µs. Data type: double RecordsPerSpectrum Set/Get number of records to collect for each spectrum. Must be within the range of 1 to 65,535. The default is 256. Data type: long 105

116 FASTFLIGHT 2 Digital Signal Averager RingingProtection Set/Get Ringing protection in units of nanoseconds (used in compression algorithm). One bin is one sampling interval. 0 to 32 ns, default 2 ns Data type: short SamplingInterval Set/Get sampling interval. 0 = 250 ps interlaced in hardware, 1 = 250 ps post-acquisition interpolated by software, 2 = 500 ps, 3 = 1 ns, 4 = 2 ns. Data type: short Sensitivity Set/Get a constant that multiplies the auto noise threshold before the threshold is used. (Used in the lossy compression algorithm, the automatic correlated noise subtraction, and for peak integration in the total-ion count). Code(Multiplier) correspondence is 0(2), 1(3), and 2(4), respectively. Data type: short SingleIonLength Set/Get the width of a region set across a specific mass peak for the purpose of integrating the net peak area above background for the specific ion chromatograph. Measured in units of nanoseconds (ns). Data type: double SingleIonStart Set/Get the starting bin index for a specific ion peak region defined in the SingleIonLength. Measured in units of nanoseconds (ns). Data type: double TimeOffset Set/Get the acquisition delay after trigger in units of microseconds (µs). Data type: double VerticalOffset Set/Get analog input dc offset in volts. The range is 0.25 V to V in steps of V (i.e., 30 µs). Data type: double 106

117 APPENDIX A. ACTIVEX METHODS AND PROPERTIES A.1.5. Establishing Different Default Values in FF2.ini for the General and Protocol Settings All general and protocol settings are stored in the FF2.ini file, which is located in the \Windows directory. The FF2.ini file is the means for saving the settings on the hard disk when exiting the FASTFLIGHT-2 software program. It contains two categories of settings: The default settings for the General Settings tab and at least one set of default Protocol Settings (up to the maximum of 16 protocols). General and protocol settings for each FASTFLIGHT-2 unit that is subsequently connected to the host PC. If you wish to use the factory default valuesfor the General and Protocol Settings tabs, the first time the application program is started, the Instrument Operation ActiveX DLL, FF2Ctrl.dll, will create the FF2.ini file and populate it with the ORTEC default general and protocol settings as illustrated in the excerpt below. Note that these defaults are demarcated by headers enclosed in square brackets, e.g., [General], [Protocol0], [Protocol1], and so on through [Protocol15]. These sections can be present in any order within FF2.ini. If you wish to present your end users with one or more different default values on the General Settings tab and/or one or more of the Protocol Settings screens, your installation program must create an FF2.ini file containing the desired defaults and install it in the \Windows folder. These defaults must use the syntax shown in the sample below. However, your file need only specify the defaults you wish to change. When the application program is started, FF2Ctrl.dll will locate your FF2.ini file, preserve the entries you have modified, and populate the file with any remaining entries, set to the factory default values. Following is an excerpt from the ORTEC factory default FF2.ini file, showing the details for the General section and Protocol 0. Protocols 1 15 repeat the structure illustrated for Protocol 0. IMPORTANT NOTE TO DEVELOPERS Note the FPGAFilePath entry below. The field-programmable gate array (FPGA) files contain the FPGA code that is downloaded into the hardware as firmware. If, in creating your own installation, you modify the path to the FPGA files, be sure that the new path is correct, and that all FPGA files, located by default in c:\program Files\FF2\FPGA Files, are installed in the new location. 107

118 FASTFLIGHT 2 Digital Signal Averager 108 [General] PresetFlags=0 MaxAcqTime= MaxSpectra=0 ActiveProtocol=0 RapidProtocol=0 TriggerPolarity=1 TriggerWidth= ExtTrig=0 ExtTrigSlope=0 ExtTrigThreshold= FPGAFilePath=C:\Program Files\ORTEC\FF2\FPGA Files ChromatFileFolder=C:\Program Files\ORTEC\FF2\ChromatDataFolder [Protocol0] ProtoName=Proto0 PrecEnhEnable=1 SamplingInterval=2 ScanLength= ScansPerSpectrum=256 TimeOffset= VerticalOffset= AdjacentBkg=32 AutoNoiseSensitivity=0 BkgInterval=200 CompressionType=1 EnableNoiseSubtraction=1 MaximumPeak=100 MinimumPeak=4 MinimumThreshold=10 RingingProtection=2 SingleIonStart= SingleIonLength= CalFile= When a FASTFLIGHT-2 instrument is connected to the PC, the set of default values are cloned and added to the end of the FF2.ini file, and each section name in the set is appended with an underscore followed by the hardware unit s serial number. For example, if the FASTFLIGHT-2 has the serial number 1, the headers in the new section of FF2.ini read [General_1], [Protocol0_1], and so on. Each time the instrument operator changes the settings for this unit, the corresponding entries in FF2.ini for this particular hardware unit are overwritten with the changes, which are retained until they are changed. Hence, the most recent user settings are recalled into the General Settings and Protocol Settings tabs when the FASTFLIGHT-2 software program is exited and re-entered, or if the computer is shut down and restarted. Each time a different FASTFLIGHT-2 unit is connected to the host PC, an additional set of default values is copied and appended to the file. For example, if FASTFLIGHT-2 number 100

119 APPENDIX A. ACTIVEX METHODS AND PROPERTIES is subsequently connected, the file (currently containing the default set of entries plus the entries for FASTFLIGHT-2 number 1) is updated to include a new set of default values with headers [General_100], [Protocol0_100], etc. For more information on the general settings, see Section B.1.1 and the discussion of GSObj properties in Section A.1.3. For more information on the protocol settings, see Section B.1.2 and the discussion of ProtocolObj properties in Section A.1.4. A.1.6. TOFObj Properties TOFObj contains all the properties for the time-of-flight spectrum. ErrFlags Get the error status of the TOF spectrum. If 0 th bit is 1, then an ADC underflow error occurred. If 1 st bit is 1, then an ADC overflow error occurred. Data type: long ProtoNum Get the protocol number under which the spectrum was taken. Data type: long SpecificIonCount Get the specific ion count for this spectrum. Data type: double SpecNum Get the spectrum number. FASTFLIGHT-2 keeps track of the number of spectra that have been collected since the most recent start of the data acquisition. Data type: long TagNum Get the tag number for this spectrum. Data type: long TimeStamp Get the timestamp for the spectrum. Measured in units of seconds. Data type: double TotalIonCount Get the total ion count for this spectrum. Data type: double 109

120 FASTFLIGHT 2 Digital Signal Averager A.2. ActiveX Graphing Control (GSX) The ActiveX Graphing Control (GSX.OCX) is a separate, in-process ActiveX (.OCX) control that incorporates all the graphing methods. It provides a standard interface for the application software to facilitate displaying the data in graphical form. The menus and spectra presented by the FASTFLIGHT-2 user interface application program are implemented via the GSX.OCX program. The custom programmer can use the GSX.OCX graphing program, or substitute another graphing program as desired. The properties and methods are documented in the following sections. A.2.1. Properties BatchMode (boolean) Enable/disable Batching Mode (allows user to perform multiple operations without updating the control). LegendState (boolean) Enable/disable the legend. AxisXText (BSTR) Set/get the text for the X axis. AxisYText (BSTR) Set/get the text for the Y axis. AxisColor (long) Set/get the axes color (0 15 base colors). GridState (boolean) Enable/disable the grid. GridColor (long) Set/get the grid color (0 15 base colors). PlotAreaColor (long) Set/get the color of the plot area (0 15 base colors). 110

121 APPENDIX A. ACTIVEX METHODS AND PROPERTIES AxisXLog (boolean) Enable/disable logging of the X-axis scale. AxisYLog (boolean) Enable/disable logging of the Y-axis scale. TitleText (BSTR) Set/get the text for the title. TitleTextColor (long) Set/get the title text color (0 15 base colors). TitleBackgroundColor (long) Set/get the title background color (0 15 base colors). TitleBorder (long) Set/get the title border type as follows: 0 = None 1 = 3D Out 2 = 3D In 3 = Etched Out 4 = Etched In 5 = Shadow 6 = Plain 7 = Frame In 8 = Frame Out 9 = Bevel LegendTextColor (long) Set/get the legend text color (0 15 base colors). LegendBackgroundColor (long) Set/get the legend bkg color (0 15 base colors). LegendBorder (long) Same as titleborder. InternalZoom (boolean) Set/get the internal (RubberRect) zoom. 111

122 FASTFLIGHT 2 Digital Signal Averager NumberOfSeries (long) Set/get the number of sets of data curves to display. ScrollBar (boolean) Set/get the internal scrollbar (scrolls data). AutoScale (boolean) Set/get the internal AutoScale (show all data points). ShowActiveGraph (boolean) Set/get the internal Active bar across top of graph. ActiveGraph (boolean) Set/get the graphs Active state. InternalMarker (boolean) Set/get the internal marker state. StartMarkerPosition (double) Set/get the start marker position. EndMarkerPosition (double) Set/get the end marker position. ActiveTool (short) Set/get the Active tool. 0 = Marker 1 = Zoom Rubber Rectangle. Smooth (short) Set/get the current data smoothing. 0 = None 1 = 3-point smooth 2 = 5-point smooth GetAxisXMin (double) Set/get the minimum X-axis value. GetAxisXMax (double) Set/get the maximum X-axis value. 112

123 APPENDIX A. ACTIVEX METHODS AND PROPERTIES GetAxisYMin (double) Set/get the minimum Y-axis value. GetAxisYMax (double) Set/get the maximum Y-axis value. A.2.2. Methods void Size(long Left, long Top, long Right, long Bottom) Size the graph. float GetVersion() Returns version of the GSX. boolean SetData(VARIANT* pvaxdata, VARIANT* pvaydata, long Series) Data into GSX. void CopyToClipboard() Copy graph to the Clipboard. void ReDraw() Force the graph to redraw itself. void SetDefaultAxis() Show all data. void Print(long Left, long Top, long Width, long Height) Prints the current graph. boolean DrawToDC(long Hdc, long Left, long Top, long Width, long Height) void AddLegendText(BSTR Text) Add legend text. void RemoveLegendText(boolean bremoveall) long GetDataLine(long Series) void SetDataLine(long Series, long nnewvalue) Turns the data lines off and on. 113

124 FASTFLIGHT 2 Digital Signal Averager long GetDataLineColor(long Series) Get data line color. void SetDataLineColor(long Series, long nnewvalue) Set data line color. long GetDataLineWidth(long Series) void SetDataLineWidth(long Series, long nnewvalue) long GetDataSymbol(long Series) void SetDataSymbol(long Series, long nnewvalue) Sets the data symbol for a specific series of data: 1 = None 2 = Dot 3 = Box 4 = Triangle 5 = Diamond 6 = Star 7 = Vertical Line 8 = Horizontal Line 9 = Cross 10 = Circle 11 = Square long GetDataSymbolSize(long Series) void SetDataSymbolSize(long Series, long nnewvalue) Set size of symbol per series. long GetDataSymbolColor(long Series) void SetDataSymbolColor(long Series, long nnewvalue) Set color of symbols per series. void ScalePixelToData(long XPixelIn, long YPixelIn, double* pxdataout, double* pydataout) void ScaleDataToPixel(double XDataIn, double YDataIn, long* PixelOut, long* pypixelout) These two functions scale data values to pixel and vice versa. 114

125 APPENDIX A. ACTIVEX METHODS AND PROPERTIES void ScalePixelToDataIndex(long XPixelIn, long YPixelIn,long Series, long* ppointout, BOOL bscreenptsonly) When using the above method bscreenptsonly should be FALSE. void ZoomXAxisIn() void ZoomXAxisOut() void ZoomYAxisIn() void ZoomYAxisOut() As stated. void ZoomYAxisAuto() Scales the y axis using only the visible data. long GetStartMarkerIndex(long Series) Gets the array index of from the current data. long GetEndMarkerIndex(long Series) Same as above except for the end marker location. long GetAnalysisFromSelection(long Series, double* pdnetarea, double* pdgrossarea, double* pdcentroid) Provide information about the current selection area. double GetYValueFromIndex(long Series, long Index, BOOL bscreenptsonly) Gets the data y value from a specified index (bscreenptsonly should be FALSE). double GetXValueFromIndex(long Series, long Index, BOOL bscreenptsonly) Same as above except for the X value. double GetYValueFromXValue(long Series, double Xvalue) Gets the y data value from an X data value. void Initialize() Reset graph double DisplaySeries(long Series) Get/sets visual state of a series of data. void ClearData(long Series) Remove a specific series of data. 115

126 FASTFLIGHT 2 Digital Signal Averager BOOL GetInternalGain() Gets the state of the internal gain. void SetInternalGain (BOOL benable) Enables/disables the internal gain operation. double GetGainValue() Gets the multiple value that is applied to the data if Internal Gain property is TRUE. void SetGainValue(double newvalue) The value that every visible point is multiplied by when Internal Gain property is TRUE. void AboutBox() Shows the GSX About box. A.2.3. ActiveX Events Sent to Parent void GSXMouseDown(short Button, short Shift, long x, long y) void GSXMouseUp(short Button, short Shift, long x, long y) void GSXMouseMove(short Button, short Shift, long x, long y) void GSXRepaint(long Hdc, long Left, long Top, long Width, long Height) void GSXClick() GSXResize(long Width, long Height) 116

127 APPENDIX B. FILE FORMATS This appendix is provided for those who wish to write their own software for controlling FASTFLIGHT-2. It provides a description of the file formats that will be encountered in the various modes of data acquisition. Section B.4 describes the format in which the FASTFLIGHT-2 hardware generates the data and transfers it to the computer. Sections B.1, B.2, and B.3 define how the data is saved on hard disk by the standard FASTFLIGHT-2 software. In the TOF mode, single spectra that are saved on the hard disk will have the.flt2 extension. The format of the.flt2 file is described in Section B.1. In the Chromatograph/Trend mode, two files define each chromatograph as it is saved on the hard disk. The file with the.fft2 extension contains the stream of TOF spectra, essentially in the format delivered by the FASTFLIGHT-2 hardware, but with the addition of a header that documents the data acquisition conditions. The second file has an.fcc2 extension. It provides a roadmap for finding the individual TOF spectra in the lengthy stream of chromatograph/tof data comprising the.fft2 file. The Chromatograph formats are described in Sections B.2 and B.3. When operating in the duet mode (see Section ), each FASTFLIGHT-2 hardware unit has a unique pair of.ffc2 and.fft2 files assigned to it. When using the standard FASTFLIGHT-2 software or the Instrument Control ActiveX DLL and ActiveX Graphing Control, it is not necessary to deal with all the details of the file formats. The ActiveX DLL and Control take care of the file format interpretation; see Appendix A for more information. B.1..FLT2 File Format (TOF Mode) The.FLT2 file format is used to save individual TOF spectra on hard disk in the TOF mode. Its format sequence, as illustrated in Fig. 55, is: 1) The general acquisition properties 2) The 16 protocol settings 3) The TOF Index record 4) The a 4-byte integer specifying the number of (X,Y) data points 5) The sequential list of X coordinates (time in ns) and the corresponding list of Y coordinates (signal amplitude) For the X and Y data, the data type is double. The TOF Index Record format is detailed by Table 6 in Section B

128 FASTFLIGHT 2 Digital Signal Averager The details of the X and Y data arrays differs slightly depending on whether the spectrum was acquired in the Lossless, Lossy, or Peak Area and Centroid Only data compression mode as set on the Protocol Settings tab under Instrument Properties (see Section ). Lossless Mode: For spectra acquired in the lossless data compression mode, the X data are spaced at the equal intervals determined by the sampling interval in the FASTFLIGHT-2 hardware. The Y data are the corresponding sampled and summed signal amplitudes expressed in units of LSB. 19 Lossy Mode: In this peak-preserving and background-decimating data compression mode, there are chunks of the background that have been deleted by the compressor in the FASTFLIGHT-2 hardware. These large gaps in the background are punctuated with periodic samples of 4 adjacent background points. When this lossy data is saved to hard disk in the.flt2 file format, the missing background points are saved as the null value, 1.0E308. The null value directs the ActiveX Graphing Control to omit those points when plotting the display of the spectrum. Thus the Y data array is filled with a mixture of valid amplitudes (in units of LSB) and null values all at the equal X spacings dictated by the sampling interval in the hardware. Fig. 55. The Data Format Sequence for the.fft2 File Format (TOF Mode). Peak Area and Centroid Only Mode: In this case, all the background information is suppressed, and the peak regions are represented by the centroid in the X data array and the area at the corresponding index in the Y data array. The rest of the X data are populated with the equally spaced sampling intervals and a corresponding null value (1.0E308) at the matching index in the Y data array. The values for the centroids can occur at unequal spacing in the X data array, because the centroid is not rounded off to the nearest sampling point. The Y data value associated with the centroid is still expressed in units of LSB, but it represents the area summed over all the sampled points in the specific peak region. Note that if you compare the centroid reported by the Only centroid and area mode to the centroid evaluated from the Lossless mode, the two numbers could differ if the limits of the peak region differ for the two calculations. 19 LSB = Least Significant Bit, referring to the LSB unit of quantization from the sampling ADC. 118

129 APPENDIX B. FILE FORMATS B.1.1. The General Settings Format (TOF or Chromatograph/Trend Mode) The details of the general settings or properties are cataloged in Table 4. Following are definitions of the General Settings file variables: The FileVersion tracks the version number of any update to the file format. This allows updated software to read all legacy versions by adapting to the actual file format. PresetFlags controls the enabling/disabling of the preset conditions that automatically stop data acquisition. Bit 0 is assigned to enabling/disabling the maximum acquisition time preset (MaxAcqTime), while bit 1 enables/disables automatic acquisition termination when the preset number of spectra has been acquired (MaxSpectra). For each bit, 1 = enabled, and 0 = disabled. ProtocolNum is the protocol number with which acquisition of the chromatograph commenced. This corresponds to the number in the Initial Protocol 20 field on the General Settings tab under Instrument Properties (see Section ). The protocol number can be changed via the RAPID PROTOCOL port on succeeding spectra after data acquisition has commenced (the changes being initiated by a signal from the mass spectrometer; see Section ). In this case, the actual protocol number used for a specific spectrum is retrieved by parsing that information from the spectrum header in the raw data stored in the.fft2 file and referenced in the.ffc2 file for the Chromatograph/Trend mode. In the TOF mode, the protocol number that was used to acquire the spectrum is embedded in the TOF Index Record (Table 6). Normally, the RAPID PROTOCOL port is not used in the TOF mode. In that case, the initial and actual protocol numbers will be the same in the TOF mode. RapidProtocol is the flag that determines whether the RAPID PROTOCOL port is enabled or disabled. TriggerEnablePolarity controls whether a high TTL level (1) is required to enable triggering, or a low TTL level (0) at the TRIGGER ENABLE IN input enables triggering. TriggerWidth sets the width of the Trigger Out pulse. ExtTrigEnable selects either the external Trigger In (1) or the Trigger Out (0) function. ExtTrigSlope determines whether the triggering time will be derived from the rising edge (1) or the falling edge (0) of the Trigger Input pulse. 20 As noted in Section , this field is labeled Active Protocol until you mark the Rapid Protocol checkbox located to the right of the field, at which point it changes to Initial Protocol. 119

130 FASTFLIGHT 2 Digital Signal Averager Table 4. General Settings (Properties) File Format (TOF or Chromatograph/Trend Mode). Variable Data Type (Bytes) Notes FileVersion LONG(4) Version number tracks any format updates. PresetFlags LONG(4) Two bits are used; 0th bit controls max. time preset; 1st bit controls max. spectra number. Bit = 1 enables preset; bit = 0 disables preset. MaxAcqTime DOUBLE(8) Max. acquisition time in seconds (s) MaxSpectra LONG(4) Max. number of spectra ProtocolNum LONG(4) Initial protocol number RapidProtocol LONG(4) 1 = Enabled, 0 = Disabled TriggerEnablePolarity LONG(4) 1 = Positive, 0 = Negative TriggerWidth DOUBLE(8) Trigger width in microseconds (µs) ExtTrigEnable LONG(4) 1 = accepting external trigger, 0 = generating trigger-out ExtTrigSlope LONG(4) 1 = Rising edge, 0 = Falling edge ExtTrigThreshold DOUBLE(8) External trigger threshold in volts SerialNum STRING(128) 64 Unicode characters ExtTrigThreshold sets the Trigger In threshold from 2.5 to +2.5 V in 10-mV steps. The time at which the Trigger In pulse crosses this threshold determines the triggering time, rounded to the next available clock pulse. SerialNum conveys the serial number, as read from the FASTFLIGHT-2 hardware by the computer. B.1.2. Format for the Protocol Settings (TOF or Chromatograph/Trend Mode) The details of the format for each of the 16 protocol settings is listed in Table 5. This format is repeated 16 times to describe the 16 possible protocols. For determining which of the 16 protocol numbers was actually used in a specific TOF spectrum, see the discussion for ProtocolNum in Section B.1.1. Following are definitions of the Protocol Settings file variables: 120

131 APPENDIX B. FILE FORMATS Table 5. Protocol File Format (TOF or Chromatograph/Trend Mode) (duplicated for each of 16 protocols). Variable Data Type (Bytes) Note RecordLength DOUBLE(8) In microsecond (µs) RecordsPerSpectrum LONG(4) SamplingInterval SHORT(2) 0 = 250 ps interlaced, 1 = 250 ps interpolated, 2 = 500 ps, 3 = 1 ns, 4 = 2 ns TimeOffset DOUBLE(8) In microsecond (µs) VerticalOffset DOUBLE(8) In volts (V) PrecEnhEnable SHORT(2) 1 = On, 0 = Off ProtocolName STRING(512) 256 Unicode characters CompressionType SHORT(2) 0 = Lossy compression; 1 = Lossless compression; 2 = Stick Diagram AutoNoiseSensitivity SHORT(2) Sensitivity multipliers 2, 3, 4 via codes 0, 1, 2 respectively RingingProtection SHORT(2) In nanoseconds (0, 1, 2, 3, or 4). MinimumThreshold LONG(8) LSB BkgInterval SHORT(2) Bins AdjacentBkg SHORT(2) Bins EnableNoiseSubtraction LONG(4) 1 = Enable correlated noise subtraction; 0 = Disable MaximumPeak SHORT(2) Bins MinimumPeak SHORT(2) Bins SpecificIonStart DOUBLE(8) Specific ion region, starting position in nanoseconds (ns) SpecificIonLength DOUBLE(8) Specific ion region length in nanoseconds (ns) CalValid LONG(4) Calibration file: 1 = valid, 0 = invalid CalFileName STRING(512) Calibration file name (256 Unicode characters) CalUnits STRING(12) Calibration units (6 Unicode characters) CalType SHORT(2) Calibration type. 0 = Linear; 1 = Quadratic; 2 = Cubic CalCoef1 DOUBLE(8) Calibration coefficient 1 CalCoef2 DOUBLE(8) Calibration coefficient 2 CalCoef3 DOUBLE(8) Calibration coefficient 3 CalCoef4 DOUBLE(8) Calibration coefficient 4 121

132 FASTFLIGHT 2 Digital Signal Averager The RecordLength is the length of the spectrum in microseconds. RecordsPerSpectrum corresponds to the number of records that are summed to form a spectrum For SamplingInterval, all the sampling intervals except.25ns interpolated are implemented in the FASTFLIGHT-2 hardware. The 250-ps interpolated sampling uses a software interpolation between the 500-ps hardware data points. Table 5 documents the correspondence between the integer value for SamplingInterval and the selected sampling interval. TimeOffset is the delay after the trigger occurs before the ADC begins saving the sampled data in each record. The delay is specified in microseconds in steps of µs. VerticalOffset is the dc offset of the ADC input in volts, referred to the rear-panel analog input on the FASTFLIGHT-2. The range is 250 mv to +250 mv in 0.03-mV steps. PrecEnhEnable documents whether the Precision Enhancer was turned on (1) or off (0). ProtocolName is the alphanumeric name assigned to the protocol by the user. CompressionType records the type of data compression that was used for the acquired data, i.e., lossless (1), lossy/peak-preservation with background decimation (0), and stick diagrams, which present only peak centroids and areas (2). AutoNoiseSensitivity adjusts the sensitivity of the automatic noise threshold that separates peaks from background. Sensitivity multipliers 2, 3 and 4 are represented by the binary numbers 0, 1, and 2, respectively. The noise threshold is computed by multiplying the automatically measured noise by the sensitivity multiplier. RingingProtection specifies the number of nanoseconds following a peak that the automatic noise sensor will ignore to avoid the distorting influence of ringing from the microchannel plate detector. The MinimumThreshold value is added to the threshold determined by the automatic noise sensor after multiplying the measured noise by the sensitivity multiplier. The units are LSB (least significant bits), referred to the spectrum. BkgInterval is the maximum number of bins that will be deleted from the background between the groups of 4 background points that are saved in the peak-preserving and backgrounddecimating data compression mode. Only numbers that are divisible by 4 are allowed. The 122

133 APPENDIX B. FILE FORMATS minimum and maximum accepted values are 64 and 1024, respectively. The recommended value is 400. AdjacentBkg is the minimum number of background bins on each side of a peak for which data points are transmitted (along with the points in the peak), when the lossy data compression mode is selected. These background bins permit estimating and subtracting the background under the peak to calculate the net peak area above background. The acceptable range of values is 1 to 96 in steps of 1. EnableNoiseSuppression: A value of 1 enables correlated noise subtraction. Zero disables the subtraction. MaximumPeak is the maximum number of bins that can be considered a peak. If this number is surpassed, the process for finding the average baseline and noise threshold turns on after being turned off during the peak. The number must be divisible by 4. Minimum and maximum values are 64 and 1024, with a default of 400. MinimumPeak is the minimum number of adjacent bins that must be above the noise threshold to be considered a peak. Minimum and maximum values are 2 and 34, with a default of 4. SpecificIonStart is the starting position of the region summed for the calculation of the Specific Ion count in the chromatograph display. The last available background value before this region is subtracted from each bin in the Specific Ion region and the net numbers for each bin are summed. The start of the region is specified in nanoseconds. SpecificIonLength is the length of the Specific Ion region following the SpecificIonStart position. The length is specified in nanoseconds. But, the number of bins is limited to the range from 1 to The CalValid variable specifies whether the calibration curve specified by the next seven variables for the horizontal axis in the TOF spectrum is applicable (1) or not (0). CalFileName is the file name supplied by the operator for identifying the calibration curve. CalUnits specifies the new units for the horizontal axis in the TOF spectrum, as entered by the operator. CalType determines whether the shape of the calibration curve is linear (0), quadratic (1) or cubic (2). 123

134 FASTFLIGHT 2 Digital Signal Averager CalCoef1, CalCoef2, CalCoef3, and CalCoef4 are the coefficients in the calibration curve represented by: X CalCoef1 (CalCoef2 t) (CalCoef3 t 2 ) (CalCoef4 t 3 ) (12) where t is the original time coordinate on the X axis in the TOF spectrum, and X is the new coordinate in the units specified by the user. The typical units for X in mass spectrometry are m/z in units of daltons/z. The parameter z is the number of positive or negative charge quanta (1, 2, 3, 4,...), referenced to the number of electrons added to or subtracted from the neutral molecule in the ionization process. B.2..FFC2 File Format (Chromatograph/Trend Mode) The.FFC2 file and the.fft2 file are generated during Chromatograph/Trend mode acquisition. These two files always occur as a pair, and always should be saved in the same directory. The.FFC2 file is the index file for finding a specific TOF spectrum in the.fft2 file. When operating in the duet mode (see Section ), each FASTFLIGHT-2 hardware unit has a unique pair of.ffc2 and.fft2 files assigned to it. The.FFC2 file format is structure as illustrated in Fig. 56, i.e., a 4-byte integer specifying the file version (1 or higher), then a 4-byte integer specifying the number of data records, followed by a series of TOF Index Records. There is one TOF Index Record for each TOF spectrum incorporated in the chromatograph. Details of the TOF Index Record format are defined in Table 6. The ProtocolNum, TotalIonCount, SpecificIonCount, and TimeStamp values from the TOF Index Records are used to plot the points in the various chromatograph spectra. Following are definitions of the TOF Index Record variables: ProtocolNum is the protocol with which the TOF spectrum was actually acquired. Because the RAPID PROTOCOL port can change the protocol during a chromatograph acquisition, ProtocolNum in the.ffc2 file could differ from the Initial Protocol number. See more information on this variable in Section B.1.1. TagNum is the tag number captured from the RAPID PROTOCOL port connector by the first trigger in the spectrum. The tag is independent of the protocol number and the RAPID PROTOCOL port handshake signals. Fig. 56. The Structure of the.ffc2 File (Chromatograph/ Trend Mode). 124

135 APPENDIX B. FILE FORMATS Table 6. The TOF Index Record Format (Chromatograph/Trend Mode). Variable Data type(bytes) Notes ProtocolNum Long (4) Protocol number the spectrum was taken with. TagNum Long(4) Tag number the spectrum was taken with. SpecNum Long (4) The sequential spectrum number TotalIonCount Double (8) Total ion count SpecificIonCount Double (8) Specific ion count (defined by user) TimeStamp Double (8) Time stamp in milliseconds, marking the beginning of the first record in the spectrum with a digital precision of 1 ms. ErrorFlags Long(4) 0th bit is Underflow, 1st bit is Overflow Length Long(4) Length of raw data in file (bytes) StartOffset Int64(8) File offset (in bytes) SpecNum: The TOF spectra are sequentially numbered to permit tracking and to document which spectra were lost if another task unreasonably diverts the computer s attention. SpecNum is the spectrum number. This number is employed in the duet mode to align the chromatographs from the two FASTFLIGHT-2 units. TotalIonCount is the net area of the peaks above background for all the peaks in the spectrum. It becomes the Y coordinate in the Total-Ion Chromatograph for the specific TOF spectrum. SpecificIonCount is the net area above background for all the peaks included in the Specific- Ion interval set by the user. Normally, the selected interval will encompass only the peak(s) from a molecule with a specific mass. See SpecificIonStart and SpecificIonLength in Section B.1.2. TimeStamp: Each spectrum captures a time stamp from the elapsed-time clock in the FASTFLIGHT-2 on the first trigger for the spectrum. This time stamp is used for the X-axis in the chromatograph. The range of the time stamp is zero to 1 ms (2 42 1) in steps of 1 ms. The maximum is greater than 139 years. The elapsed time clock is automatically reset at the start of each data acquisition. ErrorFlags: The error flags can be used to warn the user when the acquired spectrum is dangerously close to full scale or zero, such that the data might be distorted by occasionally falling outside the range of the ADC. The 0th bit is the Underflow flag, and the 1st bit is the Overflow flag. The absence of overflows and underflows is documented by 0 in both bits. An overflow or underflow is signaled by a 1 in the respective bit. The overflow and underflow in this file are extracted from the ion-count trailer in the raw data for the specific TOF spectrum. An overflow flag is set when the Y coordinate of any point in the spectrum exceeds 248 times 125

136 FASTFLIGHT 2 Digital Signal Averager the number of records in the spectrum. This amounts to about 97% of full scale. An underflow flag is set when the Y coordinate of any point in the spectrum falls below 8 times the number of records. This amounts to approximately 3% of full scale. Although the correspondence between ErrorFlags and the front-panel OVER RANGE and UNDER RANGE lights will appear to be close, they are derived from different sources. The OVER RANGE light turns on any time the output of the 8-bit sampling ADC reaches 255. The UNDER RANGE light turns on whenever the output of the ADC is 0. Length is the number of bytes dedicated to the raw data in the file for this specific TOF spectrum. StartOffset is the number of bytes from the start of the.ffc2 file to the beginning of the header for this TOF spectrum. B.3. The.FFT2 File Format (Chromatograph/Trend Mode) The beginning of the.fft2 file format is the same as the first section of the.flt2 file format, i.e., general settings followed by 16 protocols settings. But they differ in that Protocol 15 is followed by the raw data stream as delivered by the FASTFLIGHT-2 hardware in the Chromatograph/Trend mode (Section B.4). Figure 57 outlines the format of the.fft2 file. For details concerning the general settings format, see Section B.1.1 and Table 4. For the protocol formats, see Table 5 in Section B.1.2. Deciphering the raw data stream is very complicated. This task can be drastically simplified by using the methods for the Chromatograph/Trend mode as documented in Section A.1. The TOF Index Record information in the.ffc2 file enables the software to quickly locate and extract the complete data for any desired TOF spectrum in the chromatograph. For that reason, the.fft2 file is always accompanied by the.ffc2 file. When operating Fig. 57. The.FFT2 File Format (Chromatograph/ Trend Mode). in the duet mode (see Section )., each FASTFLIGHT-2 hardware unit has a unique pair of.ffc2 and.fft2 files assigned to it. 126

137 APPENDIX B. FILE FORMATS B.4. The Raw Data Format (TOF and Chromatograph/Trend Modes) This section describes the format and coding for the raw data streamed from the FAST- FLIGHT 2 hardware to the computer. This is the format in which the raw data is saved on the hard disk in the.fft2 file (see Fig. 57). It is also the format that the software must accept from the hardware in the TOF mode. In the TOF mode, the software saves the TOF spectrum on disk in the format described in Section B.1. B.4.1. Code Words Embedded in the Raw Data The hardware averager memory, used to sum successive records to form a spectrum, is limited to 24 bits per word. The largest number that can be represented by a 24-bit word is = 16,777,215. In hexadecimal representation, that number is FFFFFFh, where h indicates the number is in hexadecimal notation. However, the number of records that can be summed in the FASTFLIGHT-2 averaging memory is limited to 65,535. For an 8-bit sampling ADC this limits the maximum summed number in each bin to ,535 = 16,711,425, which corresponds to FEFF01h. Note that the summed data cannot generate a number large enough to exhibit FFh in the most significant byte, and cannot attain a number as large as FFFFFFh. This fact leaves some room for numbers that can be assigned as code words to manage the organization of the raw data, and enables the concatenation of auxiliary information needed to define the acquisition conditions for each spectrum. Before passing the spectral data to the computer, the 24-bit words are rearranged into 32-bit words for more efficient use of the 4-byte word structure in the computer. In this compacted raw data from the FASTFLIGHT-2, the 32 bits for code words are assigned the functions listed in Table 7. The address in bits 20 0 can be used to identify the bin number of the next data point to be sent, or it can be employed to transmit auxiliary information, such as the sequential spectrum number, the time stamp, or the number of 32-bit words in the compressed spectrum. The code type identified by bits 23 21, and the meaning of that code is summarized in Table 8. Table 7. Bit Assignments in Code Words. Bits Assignment Set to FFh for code words Specifies the code type 20 0 Address (or information) When bits 31 0 are added together for the SYNC code on the last line in Table 8, they form the 32-bit word, FFFFFFFFh. This word is inserted twice at the beginning of each spectrum. In a 127

138 FASTFLIGHT 2 Digital Signal Averager normal data stream, it is impossible to find two words in a row with a value of FFFFFFFFh followed by a Beginning of Spectrum code word except at the beginning of a TOF spectrum. If synchronization is lost in parsing the data stream, it can be reestablished on the next set of SYNC words. The use of code types 0, 1, and 2 is described in Sections B.4.2 through B.4.6. Several different codes have the same code type identifier. These code words can be distinguished by their position in the data stream. As the data stream flows through the data compressor in the FASTFLIGHT-2, the final length of the compressed spectrum is not known until the end of the spectrum. Thus, a type-3 code word occurs at the end of the spectrum to document the length and mark the end of the spectrum. The software that processes the data needs to know the spectrum length at the beginning of the spectrum. Consequently, the spectrum is temporarily held in a FIFO memory so that the spectrum length determined at the end of the spectrum can be written into a reserved position in the header at the beginning of the spectrum. Hence, there are three type-3 code words in each spectrum. The third type-3 code word marks the beginning of the spectrum and includes the spectrum number in its address. Table 8. Code Assignments. Code Type Code Meaning Address/Information (Bits 20 0) (Bits 23 21) 0 16-bit data is about to be sent Bin number for first data point 1 24-bit data is about to be sent Bin number for first data point 2 Stick Diagram data follows Last bin number in peak region 3 Beginning of spectrum Spectrum number (1 4,194,303) 3 Length of spectrum Number of 32-bit words in spectrum (after compression) 3 End of spectrum Number of 32-bit words in spectrum (after compression) 4 Time Stamp Low Lower 21 bits of the time stamp 5 Time Stamp High Upper 21 bits of the time stamp 6 Protocol Tag (bits 7 4) & Protocol (bits 3 0) actually used 7 Ion Count Flag Address = 0 7 SYNC Address = 1FFFFFh Because both the software and the RAPID PROTOCOL port can change the Protocol Number, the type-6 code word is the place to find out which Protocol Number was actually used for the acquisition. The lowest 4 bits of the address encode the protocol number, while the next more significant 4 bits capture the Tag number. The code-7 Ion Count Flag can be distinguished from 128

139 APPENDIX B. FILE FORMATS the SYNC code word by the fact that the address is zero for the Ion Count Flag. The Ion Count Flag code word is always followed by three 4-byte words containing the Specific Ion Count, the Total Ion Count plus the overflow and underflow flags. Note that the code types occupy the upper 3 bits of a nibble. The hexadecimal representation of the code word is written in terms of eight 4-bit nibbles. Therefore the number for the code type in Table 8 must be multiplied by 2 and added to the value of the 20 th bit in order to arrive at the representation for that nibble (bits 20 23) in the hexadecimal representation of the code word. Table 9 illustrates compartmentalizing a 32-bit word into 8 nibbles or 4 bytes. Table 9. Sectioning a 32-Bit Word into Nibbles or Bytes. Bits Nibbles Bytes B.4.2. Spectral Data Encoding To compact 24-bit data into 32-bit words, spectral data is processed in blocks of 4 bins (4 data points). The 24-bit data is broken up into bytes. For the purposes of this document, the low byte will be referred to as LB, the middle byte will be MB, and the high byte will be HB. Further, the bytes of the n th channel of the spectrum will be indicated as HB n, MB n, and LB n. Assume it is determined that bin 1000 is to be transmitted to the host computer. By default, bins are sent as well. The values of all 4 bins are inspected. If any of the 4 has a value greater than FEFFh, all three bytes are sent as follows along with a code word. First, a type-1 code word is inserted. FF2003E8h This is the code word signaling 24-bit data with a starting bin number of 1000 (decimal), i.e., 3E8h. A zero in bit 20 combines with a code-type 1 in bit 21 to yield a value of 2 for the nibble. Hence the top 3 nibbles are FF2. This code word is followed by the 3-byte data for the 4 successive bins, compacted into three 4-byte words, as depicted in Table 10. Each column in Table 10 is one byte wide. The table represents three 32-bit data words in succession from the top row to the bottom row. The most significant bytes in the 32-bit words are on the extreme left. Remember that the maximum value from the averager memory in any bin is ,535 = 16,71,425 (decimal) = FEFF01h. Therefore, the maximum value that can occur in the most significant byte in the top row of Table 10 is FEh. This cannot be confused 129

140 FASTFLIGHT 2 Digital Signal Averager Table 10. Four 3-Byte Words Rearranged into Three 4-Byte Words. HB 1000 HB 1001 HB 1002 HB 1003 MB 1000 MB 1001 MB 1002 MB 1003 LB 1000 LB 1001 LB 1002 LB 1003 with a code word, which has FFh in the most significant byte. The decoding software knows that 24-bit data must be read in sets of three 32-bit words. Consequently, FF can occur in the most significant byte of the second or third word in Table 10 without causing code-word confusion. Assuming the next bin (1004) should be sent as well, and assuming any one of the channels from has a value greater than FEFFh, no code word is necessary. The data can be sent in the same format as channels Each succeeding block of 4 bins is examined to see if any bin in the block of 4 is populated by more than 2 bytes. If the answer is YES, then that block of four 3-byte bins can be rearranged as three 4-byte words, and inserted immediately after the previous three 4-byte words. No new code word is needed, as long as all 3 bytes must be accommodated in the succeeding block of 4 bins. If the entire spectrum incorporated at least one 3-byte bin in each block of 4 bins, then the entire spectrum would be processed as 24-bit bins, and only one 24-bit data code word would be needed at the beginning of the spectrum. However, additional data compression can be achieved by noting that many blocks of 4 bins in the low-background regions of the spectrum do not utilize the most significant byte; 2 bytes is enough to document the contents of bins in the background regions. For example, assume that bin 1008 must be transmitted after bin 1007, and assume that bins all have values less than FEFFh. To compact these bins, first the 16-bit code word must be sent to signal that the format is about to switch to 16-bit data, i.e., FF0003F0h In this code word, the hexadecimal address, 03F0h is equivalent to the decimal address of the next bin, 1008, and the most significant nibbles, FF0, denote a code word signaling 16-bit data follows. Next, the 16-bit data for bins 1008, 1009, 1010, and 1011 are rearranged as two 4-byte words and aredelivered as illustrated in Table 11, with the two rows transmitted in sequence, starting with the top row. As before, the most significant bytes of the 32-bit words are on the extreme left. 130

141 APPENDIX B. FILE FORMATS Table 11. Four 2-Byte Words Rearranged into Two 4-Byte Words. MB 1008 MB 1009 MB 1010 MB 1011 LB 1008 LB 1009 LB 1010 LB 1011 If the next 4 channels only require 16 bits (<FEFFh), they can be sent immediately without a codeword. This pattern of appending additional 16-bit data in blocks of 4 bins, but rearranged as two 4-byte words, continues until the first block of 4 bins is encountered that incorporates at least one bin with 24-bit data. When 24-bit data is encountered, a 24-bit code word must be sent before transmitting the 24-bit data as described in Table 10. At this point it can be explained why the bin content FEFFh is used for the decision boundary between 16-bit and 24-bit data, instead of the expected FFFFh boundary. If the boundary were FFFFh, then the most significant byte at the extreme left of the top row of Table 11 could be as large as FF. That would make the decoding software think the first row is a code word, when that is not the intention. Limiting the 16-bit data at FEFFh allows a value no larger than FE in that most significant byte of the first row, and that cannot be confused with a code word. An FF in the most significant byte of the second row does not cause confusion because the software knows it has to read the 16-bit data in pairs of words. Choosing FEFFh instead of FFFF for the boundary between 16-bit and 24-bit data lowers the boundary by only 0.4%, i.e., a negligible change. Note that choosing the number of records per spectrum to be 255 guarantees that the maximum amplitude in the spectrum is FE01h. In that case, the entire spectrum can be compacted in the 16-bit format, and only one 16-bit code word must be inserted at the beginning of the spectrum for the entire spectrum. Less than 255 records per spectrum is typical for chromatographs. As a result, lossless data compressions very close to 2:1 can be achieved in such situations. To elaborate, if the 24-bit data points had been passed to the computer without compacting, only the lower two bytes of the 4-byte words in the computer would be occupied by non-zero digits. By squeezing 4 successive data points into two 32-bit words, instead of taking the easy route that occupies four 32-bit words, a 2:1 compression is achieved for memory utilization in the computer. B Example: Mixed 16-Bit and 24-Bit Encoding) Assume we are encoding bins 0 15 and they have the following hexadecimal values: Bin Value 0 10Ah 1 20Bh 2 30Ch 131

142 FASTFLIGHT 2 Digital Signal Averager Bin Value 3 40Dh 4 50Eh 5 60Fh 6 710h 7 811h 8 912h 9 A13h 10 B14h 11 02FF15h h 13 E16h 14 F17h h That data is encoded as: FF bit code word, bin MBs of bins 0 3 0A0B0C0D LBs of bins MBs of bins 7 4 0E0F1011 LBs of bins 7 4 FF bit code word, bin HBs of bins A0BFF MBs of bins LBs of bins HBs of bins E0F10 MBs of bins LBs of bins B.4.3. Decoding Spectral Data As an example, the data in the previous section will be decoded as follows: The first word is 16-bit code word. So, read the next two 32-bit words, parse them into 4 bins of data, and store in an array. Read the next word ( h) and determine if it is a code word or not. Since the high byte is not FF, it is not a code word; therefore, it is more data. Use this word and the following word to create bins

143 APPENDIX B. FILE FORMATS Read the next word (FF200008h) and determine if it is a code word or not. It is a code word (high byte=ff); therefore, get the address, and note that it is a 24-bit code word. Read the next 3 words and create bins Read the next word ( h) and determine if it is a code word or not. It is not a code word (high byte not FF). So read the next two words to form bins The key to the decoding spectral data is that the high byte of 24- or 16-bit data is never FF. So, if FF is found in the high byte of the next word after reading a block of 4 bins, it is a code word. B.4.4. Lossy (Peak-Preserving and Background-Decimating) Data Compression When the lossy data compression is selected, an automatic noise discriminator is used to separate peaks from background. All the data in the bins incorporating peaks are preserved and transmitted to the computer, while only a small fraction of the background points is retained. See Section for a detailed explanation of this mode of data compression. The process for compacting the 24-bit and 16-bit bin data into 32-bit words is the same for this lossy mode as described above for the lossless data compression. However, this compaction algorithm does force the number of bins in the peak regions to be an integer multiple of 4 bins. This means one to three additional background points will occasionally be added to the end of a peak region to meet the blocks of 4 bins requirement. To preserve some sketch of the background between peak regions, this data compression algorithm preserves samples of background points separated by a suppression interval specified by the user. Typically, background bins are discarded between each block of 4 background bins that are preserved. The preserved background points must incorporate 4 contiguous bins to satisfy the algorithm for compacting into 32-bit words. Otherwise, the data formatting for the lossy data compression is identical to that described above for the lossless compression. B.4.5. Stick-Diagram Encoding In the stick-diagram data compression mode, the preserved peak regions described in Section B.4.4 are further processed and reduced to two values, the peak centroid and the net peak area above local background. All the background data in Section B.4.4 is discarded. Actually, the compressor in the FASTFLIGHT-2 calculates A and AûX, where A is the net area of the peak above local background, and AûX is defined by: 133

144 FASTFLIGHT 2 Digital Signal Averager AûX A(EOR C) C is the centroid of the peak region and EOR is the bin number of the end of the [peak] region. AûX represents A times the X-interval between the centroid and the end of the region. The software must extract A and AûX from the raw data stream, and calculate the centroid from: C EOR AûX A Thus, the software must be aware of which protocol numbers call for stick-diagram compression so that the raw data can be correctly processed. Typically, the stick for each peak is displayed as a vertical bar whose height represents the area, A, and whose X coordinate is the centroid, C, of the peak. Stick spectra have exactly the same header and trailer code words as the other two types of data compression. But the spectral data are formatted differently. Each stick (which represents one peak region) is transmitted with a Stick Diagram Data code word, followed by three 32-bit words. The format for each stick, or each peak region, is defined in Table 12. Table 12. The Stick-Diagram Format for Each Peak Region. Code Word F F End of Peak Region Bin Number [20:0] Data Word WIDTH[9:0] A[35:32] A X[43:32] Data Word 2 A[31:16] A X[31:16] Data Word 3 A[15:0] A X[15:0] Bit Number A scale has been added to the bottom of Table 12 to show the positions of the bit boundaries in the 32-bit words. The top row in Table 12 is the code word for Stick Diagram Data. FF in the most significant byte denotes a code word, and 010 in bits is the code type for stick diagram data follows. Bits 20 0 contain the bin number for the end of the peak region (EOR). In the first data word, WIDTH[9:0] is bits 9 0 of the number specifying the total number of bins in the peak region. A[35:32] is bits of the number specifying the net area of the peak. AûX[43:32] is bits of the number for AûX. 134

145 APPENDIX B. FILE FORMATS In the second data word A[31:16] contains bits of the net area, and AûX[31:16] incorporates bits 31 to 16 of AûX. The remaining lower 16 bits of A and AûX are conveyed in the third data word. B.4.6. Decoding the Entire Data Stream The following gives a general procedure for synchronizing and decoding the data. It has been broken into tasks, each including decisions that lead to the appropriate next task. The initial task is Synchronize. Synchronize Search the data stream for two words in a row with a value of FFFFFFFFh. Check the next word in the stream. If it is a code word of type 3 (Beginning of Spectrum), go to Beginning of Spectrum. If any of the subsequent tasks fail, return to Synchronize. Beginning of Spectrum Extract the spectrum number from the address field in the code word, and save it as the spectrum number. Read the next word. It should be a length of spectrum code word (type 3). Extract the spectrum length from the Address field. Read the next word. Then go to Evaluate Code Word. Evaluate Code Word If the high byte is not FFh, go to Synchronize. Otherwise, extract the type from the code word (bits 23 21), and use the following table to determine what to do next. Type Go to 0 16-Bit Data 1 24-Bit Data 2 Stick-Diagram Data 3 End of Spectrum 4 Time Stamp Low 5 Time Stamp High 6 Protocol 7 Ion Count unless address is 1FFFFFFh, then Synchronize 135

146 FASTFLIGHT 2 Digital Signal Averager Protocol Extract the lower 8 bits of the code word, and assign the value for the top 4 bits to the Tag variable, and the value in the lower 4 bits to the Protocol Number variable. Read the next word, and go to Evaluate Code Word. Time Stamp Low Extract the lower 21 bits of the code word, and assign that value to the lowest 21 bits of the time-stamp variable. Read the next word, and go to Evaluate Code Word. Time Stamp High Extract the lower 21 bits of the code word, and assign that value to the highest 21 bits of the time stamp variable. Read the next word, and go to Evaluate Code Word. 16-Bit Data Extract the 21-bit bin number from the low 21 bits of the code word. Go to 16-Bit Data Actual. 16-Bit Data Actual Read the next 2 words. These two words report the contents of the bin in the code word along with the next 3 bins. Assuming the bin number in the code word is n, the two words are decoded as follows: MB n MB n+1 MB n+2 MB n+3 LB n LB n+1 LB n+2 LB n+3 Read the next word. If it is a code word (high byte=ffh) go to Evaluate Code Word, otherwise n=n+4 and go to 16-Bit Data Actual. 24-Bit Data Extract the 21-bit bin number from the low 21-bits of the code word. Go to 24-Bit Data Actual. 24-Bit Data Actual Read the next 3 words. These three words report the contents of the bin in the code word along with the next 3 bins. Assuming the bin number in the code word is n, the three words are decoded as follows: HB n HB n+1 HB n+2 HB n+3 MB n MB n+1 MB n+2 MB n+3 LB n LB n+1 LB n+2 LB n+3 136

147 APPENDIX B. FILE FORMATS Read the next word. If it is a code word (high byte=ffh) go to Evaluate Code Word, otherwise n=n+4 and go to 24-Bit Data Actual. Stick-Diagram Data Read the end of peak region bin number (EOR) from the lower 20 bits of the Stick Diagram code word. Read the next three 32-bit words, and parse all the bits for the net peak area, A, and the AûX parameter according to the map in Table 12 (repeated here for quick reference). Code Word F F End of Peak Region Bin Number [20:0] Data Word WIDTH[9:0] A[35:32] A X[43:32] Data Word 2 A[31:16] A X[31:16] Data Word 3 A[15:0] A X[15:0] Bit Number Save the value for A as the net peak area of the peak, and calculate the centroid, C, of that peak from: C EOR AûX A Save that value for the centroid along with the net area for the peak. The centroid is expressed in units of bin numbers. Read the next word, and go to Evaluate Code Word. Ion Count Read the next 3 words. These three words report the total ion (TI) and single ion (SI) counts. Each value is 6 bytes in length (48 bits). The three words are formatted as: SI Byte5 SI Byte4 TI Byte5 TI Byte4 SI Byte3 SI Byte2 TI Byte3 TI Byte2 SI Byte1 SI Byte0 TI Byte1 TI Byte0 137

148 FASTFLIGHT 2 Digital Signal Averager Extract Bit 47 of SI as the Underflow flag. Extract Bit 47 of TI as the Overflow flag. Read the next word and go to Evaluate Code Word. End of Spectrum Extract the lower 21 bits of the code word and verify that this value matches the length of the spectrum from the Beginning of Spectrum step, and from the I/O call. Go to Synchronize to look for the next spectrum. 138

149 APPENDIX C. SPECIFICATIONS C.1. FASTFLIGHT-2 Hardware C.1.1. Performance Amplitude Digitizing Resolution 8-bit ADC nominally spans 500 mv at the ANALOG IN input. Precision Enhancer Extends the limiting ADC resolution to 12 bits (for input noise <2 mv) when circa 256 or more records are averaged. 21 Can be turned on or off. Differential and Integral Nonlinearity (DNL and INL) Measured from 5% to 95% of full scale using a 500-mV, 70-µs ramp, with the Precision Enhancer on. DNL Within ±0.15 LSB referred to the 8-bit ADC. INL Within ±0.4% of full scale. Analog Input Bandwidth DC to 500 MHz; rise and fall times <1 ns. Equivalent Input Noise Uncorrelated with the Trigger <2 mv rms. Correlated with the Trigger <0.04 mv rms with Automatic Correlated Noise Subtraction turned on. Automatic Correlated Noise Subtraction Automatically assesses the correlated noise in each spectrum and subtracts it without compromising data throughput rates. Can be turned on or off. Analog DC Offset (Vertical Offset) Zero offset of the ADC is computer adjustable from 250 mv to +250 mv with 0.03-mV resolution, referred to the ANALOG IN input. Hardware Sampling Intervals 500-ps, 1-ns, or 2-ns real-time sampling with one scan per record; 250-ps interleaved sampling with two scans per record. 22 Record Size (Hardware) 8 bits per sampled point and up to 1.5 M points per record (at 250, 500, 1000, or 2000-ps/point). Record length selectable from a minimum of 10 µs to a maximum of 1.5 M points in steps of 512 points. 21 U.S. Patent 6,028, U.S. Patent 6,094,

150 FASTFLIGHT 2 Digital Signal Averager Spectrum Size (Hardware) Identical to Record Size, except 24 bits per sampled point, providing rapid hardware summing of up to 65,535 records in a spectrum. 140 ADC Sampling Interval (ps) Min. Spectrum Length (µs) Max. Spectrum Length (µs) Data Acquisition Delay (Time Offset) Computer selectable digital delay after trigger from 0 to µs in 16-ns steps. The record starts after the selected delay. End-of-Scan Dead Time 0.8 µs. Sampling Clock Internal 2 GHz with temperature sensitivity <2 ppm/ C. 10-MHz Clock Output Phase-locked to the internal 2-GHz sampling clock for supplying the sampling clock reference to another FASTFLIGHT-2 in the duet mode. Trigger-to-Sampling Clock Synchronization The first sampled point in the record is synchronized within ±250 ps relative to the leading edge of the external Trigger Input for realtime sampling. The Trigger Output is synchronized to the first sampled point in the scan with a jitter <50 ps FWHM. The Trigger Output is alternately delayed by 0 and 250 ps relative to the sampling clock in the 250-ps interleaved sampling mode. Operating Temperature Range 0 50 C. Averaging Method Linear summation of sequential records. Maximum Acquisition Time The number of TOF-MS spectra acquired can be limited by presetting the maximum time. Selectable in 1-s increments from 1 s to 65,535 s (18 hr), or disabled. Spectra per Chromatograph >18,000. Limited only by available memory in the supporting PC and the data storage disk. Data Compression Implemented in the hardware with no compromise in data throughput. Lossless The default mode. Compression down to 2/3 the normal 24-bit file size in spectra dominated by background, and with no loss of original data.

151 APPENDIX F. SPECIFICATIONS Lossy (Peak-Preserving and Background-Rejecting) 23 Automatically separates peaks from background. Transmits peaks and adjacent background points. Selectable decimation of the background points between peak regions. Data compression by a factor of 10 30, depending on peak density. Peak Centroid and Net Area Transmits only the centroid and net area of automatically detected peaks. Provides an additional factor of 9 data compression relative to lossy compression. Maximum Data (Spectra) Transfer Rate Up to 100 spectra/s transferred to PC RAM and hard disk for a 50-µs spectrum length and 500-ps sampling (for a hard disk with a sustained writing speed >20 MB/s). Total-Ion/Specific-Ion Chromatographs Automatic generation of real-time Chromatograph displays with each point in the chromatograph linked to the supporting time-of-flight spectrum. Provides exact time synchronization of the chromatograph with the TOF-MS when analyzing the output of an LC or GC. Total-Ion Chromatograph The hardware computes the sum of the areas above background for all peaks in each spectrum, and passes that number to the computer via the spectrum header. The sum is used for the vertical scale in the chromatograph. Specific-Ion Chromatograph The operator selects the boundaries of a specific peak in the spectrum to generate the Specific-Ion Chromatograph from the net area above background in that peak. The net area is included in the header for each spectrum. The Specific-Ion Chromatograph can be generated during data acquisition or post-acquisition. Time Stamp 42 bits in the spectrum header are allocated to recording the starting time of each spectrum with 10-µs precision. Maximum limit is circa 1.4 years. Within the FF2 application program, resets to zero at the start of each acquisition. ActiveX methods provide for an acquisition start without resetting the time, and a separate time-reset method. Spectrum Number 21 bits in the spectrum header are allocated to recording the sequential spectrum number. Maximum limit is circa 2 million. Resets to zero on each acquisition start. Disabled during a hardware data acquisition disable. Otherwise, this counter runs continuously, even when software stops reading the 32-MB FIFO output. Output Buffering During the last record in each spectrum the sum of all records in the spectrum is written to a 1-spectrum-deep output buffer, and the summing memory is released to acquire the next spectrum. This limits end-of-spectrum dead time to 0.8 µs. After data processing in the compressor, the spectrum is loaded into a 32-MB output FIFO to accommodate 23 U.S. Patent 5,995,

152 FASTFLIGHT 2 Digital Signal Averager intermittent data transfer over the USB 2.0 bus to the PC without loss of spectra. The output FIFO has a capacity of at least 7 spectra. Rapid Protocol Selection Provides a hardware interface to change acquisition parameters in real time within 10 µs. Includes 4 bits to select the hardware parameters defined in one of 16 protocols, and an additional 4 bits to insert one of 16 tags in the spectrum header. The tags can be used to identify unique acquisition conditions from other parts of the mass spectrometer. Duet Operation A second FASTFLIGHT-2 can be synchronized to the sampling clock and trigger of the first FASTFLIGHT-2 to extend the dynamic range. The analog signal is amplified by an additional factor of 10 for the second unit, and the two units are operated as a duet. This lowers the detection limit set by correlated noise by a factor of 10 in the unit fed from the higher gain, resulting in an overall increase of 10 in dynamic range. C.1.2. Hardware Controls and Indicators TRIGGER Front-panel LED flashed by each Trigger Output. ACQUIRE Front-panel LED is on when data acquisition has been enabled. READOUT Front-panel LED flashed by each data transfer to the PC. UNDER RANGE Front-panel LED flashed on whenever the sampling ADC reads zero for the analog input signal. Provides guidance for adjusting the dc-offset relative to the lower limit of the ADC. OVER RANGE Front-panel LED flashed on when the analog signal meets or exceeds the maximum code (255) of the ADC. Provides guidance for adjusting the maximum amplitude of the analog input signal. POWER Front-panel LED is on when the rear-panel power switch is on and power is supplied to the unit. POWER ON/OFF Rear-panel on/off switch connects/disconnects the FASTFLIGHT-2 to/from the external dc power supply. 142

153 APPENDIX F. SPECIFICATIONS C.1.3. Inputs and Outputs All inputs and outputs are on the rear panel. ANALOG IN BNC connector accepts the analog signal for time-sequenced sampling. Input impedance is 50, dc-coupled. The sampling ADC code from 0 to 255 spans an input range from 0 to 500 mv, with code zero adjustable over a ±250-mV range with a 0.03-mV resolution. TRIGGER ENABLE IN BNC connector accepts a TTL input to enable or disable both the Trigger Output and the Trigger Input. Automatically pulled to the high level if no input is supplied. Computer-selectable assignment of the enable condition to either the high or low TTL state. Holding the Trigger Enable Input in the disabled state prevents triggering. Used for synchronizing data acquisition when the RAPID PROTOCOL port is used, or in duet operation. Input impedance is a 10 k pull-up to +3.3 V. Minimum recognizable enable duration is 40 ns. Relative to the Trigger Input leading edge, minimum set-up and hold times are both 20 ns. TRIGGER IN BNC connector accepts either analog or logic signals to trigger the start of each scan from an external source for real-time sampling. Synchronizes the first sampled point in the record within ±250 ps. The triggering threshold is software adjustable from 2.5 to +2.5 V in 10-mV steps, with selectable positive or negative slope. Maximum linear input is ±5 V. Protected against overloads to ±5 Vdc, and ±15 V for pulse widths 25 ns. Minimum pulse width at threshold: 25 ns. Selected for active use via software. TRIGGER OUT BNC connector provides a TTL output for triggering an external instrument from the FASTFLIGHT-2. The leading edge of the rising pulse is synchronized with the first sample point in the scan. Synchronization jitter is <50 ps. The width of the output pulse is computer selectable from 64 to 5,120 ns. Either the leading edge or the trailing edge can be used to trigger the TOF-MS with the same jitter. The Trigger Output is generated immediately after the end-of-scan dead time, if the Trigger Input has not been activated, the BUSY signal is low, and the Trigger Enable Input is in the enabled state. The Trigger Output is alternately delayed by 0 and 250 ps relative to the sampling clock in the 250-ps interleaved sampling mode. 10 MHz CLOCK IN Accepts a 10-MHz signal and automatically phase locks the sampling clock to that external clock input. Minimum and maximum recommended peak-to-peak amplitudes for detection and automatic phase locking are 500 mv and 2.5 V, respectively. Input impedance is 50 (150 to dc ground in parallel with 75 to ac ground). Used for synchronizing the sampling clocks in the duet mode. 10 MHz CLOCK OUT Provides a 10-MHz clock output that is phase-locked to the sampling clock. Output is standard 0 to +3.3-V logic, dc-coupled, with 50 output impedance. Used to synchronize the sampling clock in another FASTFLIGHT-2 in the duet mode. 143

154 FASTFLIGHT 2 Digital Signal Averager START OUT BNC connector provides a high TTL voltage when the Start software button is active, and a low TTL voltage when the Stop software button is active. ABORT IN BNC connector accepts a high TTL voltage to terminate spectrum acquisition under the current protocol number. Minimum duration of the Abort signal in the high state is 50 ns. Input impedance is 1 k to ground. Data acquisition is terminated by the Abort signal during the current scan. BUSY OUT BNC connector provides a high TTL voltage when FASTFLIGHT-2 has accepted a Trigger Enable Input and/or an external Trigger Input and has started a scan. BUSY returns to the low state at the end of each scan, when FASTFLIGHT-2 can process another trigger. BUSY is also held high whenever the DSA is not able to respond to a Trigger Input or a Trigger Enable Input. PREAMP POWER Female, 9-pin, D connector (Fig. 58) provides the dc power for a preamplifier. Pin assignments are +12 V on pin 4, 12 V on pin 9, and ground on pins 1 and 2. RAPID PROTOCOL Male, 15-pin, D connector permits the TOF-MS hardware to select the FASTFLIGHT-2 operating protocol in <10 µs, a speed that is not possible with software through the normal PC interface. This feature is used during a chromatograph/tof-ms acquisition to change the TOF-MS operating parameters as TOF spectra are acquired across a peak Fig. 58. The PREAMP POWER Connector. in the chromatograph. One typical use is to alternate the TOF-MS between (a) the fragmentation mode and (b) the precursor-ion mode. This permits the user to collect all varieties of molecular analysis during a single chromatographic run. In the FASTFLIGHT-2, the protocol is defined by the settings for the following items: Protocol Parameters Protocol number (0 15) Sampling interval (250 ps interlaced or interpolated, 0.5, 1, or 2 ns) Record length Time offset (data acquisition delay after trigger) Vertical offset (analog input dc offset in volts) Number of records per spectrum Precision Enhancer on/off Automatic correlated noise subtraction on/off Data compression mode Lossless Lossy (Peak-preserving and background-rejecting) Peak centroid and net area 144

155 APPENDIX F. SPECIFICATIONS Data compression parameters Minimum peak detection threshold Auto-noise threshold on/off Peak-detection sensitivity factor Adjacent background Background sampling interval Minimum peak width Maximum peak width Ringing protection Chromatograph vertical Scale Total-ion area Specific-ion area a) Selected peak lower mass limit b) Selected peak upper mass limit Calibration filename and path (converts TOF to m/z) In addition there are 4 tag bits on this connector that are strobed by the first Start pulse in each spectrum. The protocol number and the tag bits are stored in the header of each TOF spectrum. The Chromatograph Spectrum Display decodes the protocol number to display the chromatogram for the selected protocol number. The operator can choose any protocol number for live display at any time during acquisition of the chromatogram. To avoid potential conflicts, the RAPID PROTOCOL port is disabled when data acquisition is not in progress. Consequently, changing the properties that define the protocols (and/or the General properties) via the graphical user interface is only possible when data acquisition is not in progress. When the Instrument Properties dialog is closed, the software updates the properties in the FASTFLIGHT-2 hardware, and it is not possible to start acquisition until this process is complete. Figure 59 shows the RAPID PROTOCOL connector. The pin assignments are presented in Table 14. COMPUTER USB-2 Standard USB 2.0 Type B connector for communication with the supporting computer over the USB 2.0 bus. FASTFLIGHT-2 includes a 3-meter (10-ft.) USB type A/B cable. Fig. 59. The RAPID PROTOCOL Port Connector. POWER IN 5.5 mm OD 2.5 mm ID dc power jack accepts 3-A, +15 V dc, power from an external power supply. 145

156 FASTFLIGHT 2 Digital Signal Averager C.1.4. Electrical and Mechanical Power Requirements External ac-to-dc power supply provides the <3-A, +15 V dc power input to the FASTFLIGHT-2 enclosure. The external power supply accepts V ac at Hz and can deliver up to 70 W at +15 V to the FASTFLIGHT-2 chassis. AC inrush current is < V ac; continuous ac current is <1.5 A. Provides an IEC320/C14 connector compatible with international ac power cords. DC plug: 5.5 mm OD 2.5 mm ID 12 mm long, compatible with jack on FASTFLIGHT-2 rear panel. External power supply operating temperature range: 0 40 C. Table 14. RAPID PROTOCOL Port Pin Assignments. Pin Number FASTFLIGHT-2 TTL Signals on male, 15-pin D connector 1 Protocol Number bit 0 2 Protocol Number bit 1 3 Protocol Number bit 2 4 Protocol Number bit 3 5 Tag Bit 0 6 Tag Bit 1 7 Tag Bit 2 8 Tag Bit 3 9 Select Protocol (Input) 10 Protocol Accepted (Output) 11 Acquiring TOF Spectrum (Output) 12 Ground 13 Ground 14 Ground 15 Ground Weight FASTFLIGHT-2 chassis net weight 4.9 kg (10.9 lb.) External Power Supply net weight 0.45 kg (1.0 lb.) Total shipping weight 6.3 kg (13.9 lb.) Package and Dimensions FASTFLIGHT-2 chassis 33 cm W 34 cm D 7.4 cm H (12.9 in in. 2.9 in.) External Power Supply 1.32 cm W 0.58 cm D 0.30 H cm (5.20 in in in.) 146

157 APPENDIX F. SPECIFICATIONS CAUTION Cooling fans near the front of the left side panel and the rear of the right side panel require unrestricted air flow. CE Complies with CE low-voltage directives and CE regulations for susceptibility and emissions (level B). C.2.1. Architecture C.2. FASTFLIGHT-2 Software The software provided with the FASTFLIGHT-2 comprises three main components: 1. The Instrument Operation ActiveX DLL 2. The ActiveX Graphing Control 3. The User Interface Application Program Figure 60 diagrams these major components. Fig. 60. FASTFLIGHT-2 Software Block Diagram. 147

158 FASTFLIGHT 2 Digital Signal Averager The FASTFLIGHT 2 operating software includes everything needed to control the hardware, acquire the data, display the chromatograph and time-of-flight spectra, and perform data manipulation, all without any further programming. The application program runs under Windows 2000 Professional and XP Professional. For those who wish to write their own user interface to integrate the FASTFLIGHT-2 software with the programs that operate the mass spectrometer, the Instrument Operation ActiveX DLL and ActiveX Graphing Control and the supporting documentation in Appendix A provide a programmer s toolkit. The Instrument Operation ActiveX DLL encapsulates the intelligence needed to control the hardware, acquire the data, and save the spectra at the highest possible rate on the PC hard disk. It also provides the means of retrieving saved spectra without impeding data acquisition. ActiveX controls standardize and simplify the interface between the User Interface Application Program and the DLL that controls the hardware and data flow. This facilitates integrating the FASTFLIGHT-2 into application-specific software. The Instrument Operation ActiveX DLL takes care of all the complications inherent in communicating with the hardware, interpreting the data format, and maximizing the data flow rate. Requested data is returned to the application program in a safe array (in a raw data format). Instrument control is implemented using the standard Get and Set property format. The ActiveX Graphing Control (GSX.OCX) is a separate, in-process ActiveX (.OCX) control that incorporates all the graphing methods. It provides a standard interface for any user-interface application software to use for the purpose of displaying the data in graphical form. The menus and spectra presented by the standard ORTEC user interface are implemented via the GSX.OCX program. The User Interface Application Program supplied with the FASTFLIGHT-2 provides all the operating and display features needed to use the FASTFLIGHT-2, with no additional programming required. This application program utilizes the Instrument Operation ActiveX DLL and the ActiveX Graphing Control to implement its tasks. The standard software runs on a PC under Windows 2000 Professional and XP Professional. Programmer s Toolkit Those who wish to integrate the FASTFLIGHT-2 into their own application program can easily do so by utilizing the Instrument Operation ActiveX DLL and the ActiveX Graphing Control in conjunction with the documentation in Appendix A. These two controls take care of the complicated details involved in data acquisition, control, data format interpretation, and display; and provide the applications programmer with a simple implementation. The ActiveX Controls are supported by Microsoft Visual C++, Visual Basic, Visual Studio.NET, and National Instruments LabVIEW. 148

159 APPENDIX F. SPECIFICATIONS C.2.2. Standard Application Software Software Controls Includes the control dialogs necessary to implement the hardware features listed above. Chromatograph Spectrum Display Provides a live display of the spectrum from the column source versus retention time as the data is collected. The total-ion chromatograph is derived from the total peak area above background for all masses in the TOF-MS spectrum. The specific-ion chromatograph is generated by the net area above background for a specific mass peak that has been selected in the TOF-MS spectrum by the operator. Both chromatographs are continuously available for live display. In addition, the RAPID PROTOCOL port permits interleaving up to 16 different data acquisition conditions for the TOF-MS spectra in each chromatograph. The chromatographs can be displayed separately for each protocol number. Four tag bits allow a further division of identification by up to 16 tag numbers. TOF-MS Spectrum Display Permits display of a TOF-MS spectrum that has already been acquired in the chromatograph, without interruption of further data acquisition. Left-clicking the mouse on a point in the chromatograph causes the TOF-MS spectrum comprising that point to be displayed. It is possible to switch back and forth between the live chromatograph display and the TOF-MS spectrum display. A mode is also available for collecting a single TOF-MS spectrum without an associated chromatograph. Time vs. Calibrated Units For the TOF-MS spectrum, the software provides a quick toggle selection of either the basic time units, or the calibrated units (typically m/z or daltons) for the horizontal scale. Data Storage The standard software streams the chromatograph/tof-ms data for storage onto a hard disk in the supporting computer. The programmer s toolkit enables custom solutions to be implemented for other storage media. C Hardware and Operating System Requirements FASTFLIGHT-2 requires a supporting personal computer operating under Microsoft Windows 2000 Professional or Windows XP Professional SP 2 or later, and meeting or exceeding the following specifications. Using a PC with higher speed and capacity could improve performance. >2.0-GHz microprocessor 512MB SDRAM at 400MHz At least one USB 2.0 port that can be dedicated to the FASTFLIGHT-2 149

160 FASTFLIGHT 2 Digital Signal Averager For the chromatograph mode, a hard drive with at least 20 GB of free space space and the 20-MB/s sustained writing speed necessary to support 100-spectra/s acquisition rates. 24 CD drive VGA display 24 A 7200-rpm UDMA-150 hard disk is a minimum criterion for achieving the 20-MB/s sustained writing speed, but be sure to test the hard disk to determine its actual capability, according to the instructions in Section

161 APPENDIX D. CHANGING THE PREAMPLIFIER COARSE GAIN The Model 9326 Preamplifier Input has three coarse-gain settings (5 V/V, 10 V/V, and 20 V/V). This appendix tells you how to change between settings. CAUTION Change the Model 9326 coarse-gain jumper before connecting the unit to any other equipment. 1. Remove the two outboard screws on the rear panel of the preamplifier. Do not remove the screw between the Output and Power connectors. CAUTION To avoid damaging the preamplifier, disconnect it from all voltage sources before removing the cover. 2. Slide the printed wiring board (PWB) out of the case by pulling on the rear panel output connector. 3. Figure 61 shows the location of the gain jumper on the board. (Note that the values of the three jumper settings are printed on the board beside their respective settings.) Change the board jumper to the desired gain setting. 4. Carefully slide the board back into the same suspension slots on the preamplifier case. 5. Insert the two outboard screws to reattach the rear panel to the case. Fig. 61. Setting the Model 9326 Gain Jumper. 151