A dedicated data acquisition system for ion velocity measurements of laser produced plasmas N Sreedhar, S Nigam, Y B S R Prasad, V K Senecha & C P Navathe Laser Plasma Division, Centre for Advanced Technology, Indore 452 013, India This paper describes the design of a multi-channel 25 MSPS ( Mega samples per second) data acquisition system. The approach is based on a flash analog to digital converter and a FIFO (first in first out) memory. This system is used for acquiring the ion velocity profiles using Langmuir probes in laser plasma interaction experiments carried out using a high power Nd: Glass laser. These profiles are displayed on the personal computer monitor and processed for calculating the ion velocities. The digitized data is stored for further analysis.
Data acquisition plays an important role in the field of laser produced plasmas. The short duration of the plasma makes the process of data capturing and recording even more difficult. The typical time duration of the laser produced plasmas is about few nano seconds and the spatial dimension is about 100μm with the plasma temperature being few hundreds of ev. Since there are many complex processes taking place in the plasma at these high temperatures, it is of vital importance to use different types of diagnostic tools simultaneously for getting the complete plasma information. As the number of diagnostic tools increases, it becomes practically difficult to record all this data manually using oscilloscopes. Thus, data acquisition systems play a crucial role in the field of laser produced plasmas. Langmuir probes 1 are routinely used for the diagnostics of laser produced plasmas. The Langmuir probes are nothing but small electrodes inserted into the plasma chamber. A biasing voltage is applied to the electrode so as to collect or repel the ions or electrons selectively. The probe is kept at a suitable distance from the plasma so as not to disturb the plasma behaviour. The typical Langmuir signal consists of a fast rising pulse due to X-rays ( also UV and other em radiation ) generated in the plasma and the main pulse due to ions. Since the velocity of the ions is several orders less than that of electromagnetic radiation, the X-ray pulse can be considered as time fiducial for all practical purposes with little error. From the shape of the ion pulse the ion velocity profile can be estimated. Since the number of Langmuir probes generally used in the laser plasma experiments are large in number, it is of importance to acquire the data on line using data acquisition systems. Data acquisition systems generally include a signal conditioning circuit, analog multiplexer, an analog-to-digital (A/D) converter and a microprocessor. The signals coming from various sensors are routed to the A/D converter through the multiplexer and then digitized. The digital value is read by the processor and stored in its memory. Such systems are generally controlled by software and they are well suited for continuous processes. However, the Nd: Glass laser is operated in single shot at a very low repetition rate, and the signals are generated for a very short duration. Thus it is necessary to build a fast data acquisition system, which can acquire data in short duration, so as to capture the large amount of data efficiently. We have presented here a fast data acquisition circuit, developed for High Power Nd:glass Laser Laboratory, Centre for Advanced Technology, Indore, India. The circuit is simple and modular. It has been used for acquiring signals from Langmuir probes for the measurement of ion velocity profiles of laser produced plasmas. The design of the electronic system is presented in the following section. The experimental setup is also briefly discussed followed by the experimental results. System Design The Nd:glass laser is operated at an interval of 30 minutes or more at full laser energy, (typically 100 J in 12 ns or 50 J in 5 ns for plasma experiments). The laser pulse is incident on a planar solid target and plasma is generated. The X-rays generated in the hot plasma reach the Langmuir probe first and generate a photoelectric signal of few
nanosecond duration, which can be used for time fiducial. The electrons and ions reach the probe later with velocities corresponding to plasma temperature and give rise to a comparatively slower signal So, it is necessary to capture the first pulse for getting timing information and acquire waveform of the actual signal with high sampling rate. A fast data acquisition system can be designed as described in the introduction, but with the help of fast logical circuits 2,3,4. However, we have adopted a simpler approach, which is based on a FIFO (first - in - first - out) memory. The principle of this method is described in Fig.1. The signal to be digitized is fed to a flash A/D converter, followed by a FIFO. A trigger signal which arrives just prior to the main signal, derived from the laser control unit is fed to a comparator, which sets up a flip flop and initiates the process of sampling by enabling the clock signals to the A/D converter as well as the FIFO memory for reading. When the FIFO becomes full with data, it generates a FIFO Full signal (FF) which is used for stopping of the process. The data stored in the FIFO can then be read by the microprocessor at a slower speed. The detailed circuit of the digitizing card is shown in Fig.2. The analog signal is fed to a flash A/D converter and the trigger signal to the comparator. The comparator sets the flip flop F1 provided that other input of the gate G1 is enabled by the microcomputer. This enables the flip flop F2 and also the clock pulses from the crystal clock of 50 MHz to be routed alternately to the A/D converter and FIFO. This process is continued till FIFO is full, which disables further clock pulses, through FF signal. The microprocessor reads this signal and then reads the FIFO through the parallel I/O port and then resets it. The comparator used is AM 686, which has a 12 ns propagation delay time, which makes it suitable for detecting fast trigger pulse. The A/D converter is Motorola MC 10319 5, which is operated at 25 MSPS. The FIFO used is IDT 7205 6, which is 8 Kbyte X 9 memory. The crystal clock generates 50 MHz signal, which is routed alternately to the A/D converter and the FIFO. With 25 MSPS as sampling rate, the time resolution is 40 ns and the total time duration for which the signal is digitized is 328 ms. The complete data acquisition system is shown in Fig.3. One digitizing card is allotted for each channel and it is possible to include any desired number of such cards in the system. The dedicated microcomputer controls all the digitizing cards and transmits the data to a personal computer through a serial port. Thus, the data acquired from each digitizing card can be analyzed and displayed in suitable format on the personal computer. This distribution makes it possible to keep the data acquisition unit near the experimental setup, whereas the personal computer can be located near the console. Software The software developed for this system can be divided in two distinct parts. The first part is a program on the microcomputer controlling all data acquisition cards and acquiring the data from the FIFOs into the memory. This is developed in assembly language with the help of cross-assembler and stored in the EPROM in the microcomputer. The second part is developed on a Personal Computer (PC), which consists of program for the user interface and communication with the microcomputer for the control and transfer of the data. This is developed in C language and stored in the PC. The microcomputer normally waits for a signal from the PC, and when it is
received, it enables the input to the gate G1 on each card, which in turn enables the external trigger signal to start the process of digitizing. The CPU then waits for the FIFOs to become full and then reads them into its memory sequentially. The program on the PC is a menu driven software, with user selecting various options. Initially, the user can set and store parameters such as memory per channel, distance of the probe from the target etc. Then the acquisition of data can be initiated, which sends a signal to the microcomputer to start the process of digitization, followed by transfer of the same to the PC. The same data is then displayed in a graphical form on the PC, and stored for further processing. The peak velocity of ions is also calculated by the software. This is done by locating the x-ray and ion peaks and by taking the time difference δt between these two peaks. The ratio of the probe distance L to the δt gives the velocity. This is displayed on the plot of each channel in addition to other relevant information. Experimental set up Laser beam with energy varying between 1-100 J in 15 ns time duration is generated in a Q-switched Nd:glass laser system ( λ = 1.054 μm ). The plasma is generated by focusing the laser beams using a plano convex lens of 500mm focal length on to thick slab targets of copper or aluminium kept in a vacuum chamber at 8 * 10-5 torr. The typical focal spot size was about 100μm, thus yielding laser intensities of 10 11-10 13 W /cm 2. The experimental setup is schematically shown in Fig. 4. A BNC female connector with a copper tip soldiered at one end was used as Langmuir probe. A negative voltage of 30 V was applied on the electrode. The central conductor was insulated using a teflon tape leaving only tip exposed to the ions. The circuit diagram used for the probe is shown in Fig. 5. Two probes were placed at a distance of 25 cm from the target at angle of 15 with respect to the laser propagation direction and another probe was placed at a distance of 50 cm making 45. The outputs of the probes were terminated with 50 Ω for proper impedance matching and fed to the data acquisition system after proper attenuation. Experimental Results Slab targets of copper were irradiated with laser beam. Fig. 6 shows the signal acquired from one probe and displayed by the data acquisition system. The laser energy in this particular shot was ~10 J in 15 ns with laser intensity of 8.5*10 12 W/cm 2. The time difference of 2.84 μs between the X-ray pulse and the ion pulse yields the peak ion velocity of 8.8*10 6 cm/sec which agrees well with the theoretical values and the information obtained from other diagnostics. Conclusion A four channel data acquisition system with sampling rate of 25 MSPS/channel was developed and successfully used to acquire ion velocity profiles in the laser produced plasma. This system is modular and simple to design. The same system can be modified to acquire more number of channels by adding similar digitizing cards. The memory per channel can be adjusted from 256 bytes to 64 kbytes by choosing a suitable FIFO without any modification in the circuit. Then the sampling rate of each channel may be adjusted to any desired value less than 25 MSPS to suit the signal bandwidth and
total duration of the signal. Acknowledgement We are grateful to the high power laser and plasma team for their support during this work. We would also like to thank Mr Rajendra Singh and Mr N.R.Biswas for the wiring of the circuits.
References 1. Chen Francis F, Electric Probes, Plasma diagnostic techniques, edited by Huddlestone Richard H and Leonard Stanley L, (Academic Press, New York), 1965, 113. 2. Correia C, Combo A, Correia M, Simoes J B, Coelho P, Carralho B B, Sousa J & Varandas C A F, Rev.Sci.Instrum., 70(1) (1999) 509. 3. Givens M & Sanile J, Electronic Design, July 23,(1987), 153. 4. Deevy K, Sheehan D & Byrne M, Build a single shot recorder, to catch fast transients, AN-296, Application Reference Manual, Analog Devices, 1993. 5. Analog/Interfaces ICs Device Data, vol.ii, Motorola, 1995. 6. Specialized memories and modules databook, Integrated Devices Technology, 1994.
List of Illustrations: Fig.1. Working principle of a fast data acquisition system based on FIFO memory. Fig. 2. Detailed circuit of digitizing card. Fig. 3. The complete data acquisition system. Fig. 4. Experimental setup for Laser Plasma experiments. Fig. 5. Biasing circuit for Langmuir probe. Fig. 6. Signal acquired from one of the probes. Fig1
Fig2 Fig3
Fig4 Fig5
Fig6