School of Engineering Science Simon Fraser University Burnaby, BC, V5A 1S6 tp-audio@sfu.ca January 24, 2004 Dr. Lakshman One School of Engineering Science Simon Fraser University Burnaby, BC, V5A 1S6 Re: ENSC 440 Project Proposal for Digital Audio Input Speakers Dear Dr. One, Attached is the document, Proposal for Digital Audio Input Amplified Speakers, which outlines Team Power Audio s plans to design and implement a speaker system that accepts audio signals in digital format and implements the signal amplification circuitry within the power supply circuitry to make for a more compact, power efficient system. The attached project proposal will provide an overview of our proposed product as well as its market potential by comparing it to other existing products being used. It will also outline a tentative budget, proposed deadline schedule, and the team mechanisms in doing this project. Team Power Audio consists of a group of three motivated engineering students of diverse abilities and a common goal in our proposed project. Members of Team Power include Dave Steele, Brandon Pun, and James Lu. Should you have and questions, comments, or concerns, please feel free to contact us at our group e-mail: 440-team-power@sfu.ca. Or please contact our CEO, Dave Steele, by phone at 604-944-2626 or 604-944-6716 after hours. We thank you for your time and effort. Sincerely, Dave Steele President and CEO Team Power Audio Solutions Enclosure: Proposal for Digital Audio Input Amplified Speakers Team Power, 2004 i
Proposal for Digital Audio Input Amplified Speakers Project Team: Contact Person: Dave Steele Brandon Pun James Lu Dave Steele 440-team-power@sfu.ca Submitted to: Revision: Dr. Lakshman One ENSC 440 Michael Henri Sjoerdsma ENSC 305 School of Engineering Science Simon Fraser University 2 Team Power, 2004 ii
Executive Summary A stereo system composed of CD player, amplifier and speakers makes up a common appliance that is found in almost every household in one form or another. Ranging from a simple all-inone CD player to a full-scale surround sound system, music and the ability to play it on demand, is an integral component of North American culture. At Team Power Audio, we re working to take a revolutionary step in the evolution of home and commercial audio equipment. The Team Power Audio Solution, put simply, is an amp and power supply built into a speaker box that decodes a low-power digital audio input signal rather than the high-power analogue signal that is common to systems today. Put differently, rather than needing a separate stereo amp to drive the speaker, the speaker drives itself. There are two important ways in which this new audio configuration benefits the customer. Firstly, by receiving a digital input rather than analogue, our system reduces the interference that naturally occurs in the wires between the speakers and audio signal source effectively to zero. This is a huge benefit to people who have the need to run long cables to their speakers or have their equipment set up in an electro-magnetically noisy environment. Secondly, by combining the signal amplification circuitry with that of the power supply circuitry, we improve speaker efficiency in terms of power as well as space. This system makes for an increasingly versatile home audio setup, but where its true value lies is in commercial audio applications such as large concert or auditorium setups. By eliminating the need for an amp rack, our system significantly reduces the labor required in set-up and takedown times for any large-scale audio system. The Team Power Audio Solution ultimately makes for a simpler, more versatile home audio setup by removing the middleman that is the stereo amplifier of the olden-days. At Team Power, we strongly believe that our system is the way of the future for home and industrial audio electronics and we re committed to make that belief a reality. Team Power, 2004 iii
Table of Contents 1) Introduction... 5 1.1) First Prototype... 6 1.2) Second Prototype... 6 2) Design Solution... 7 2.1) Chosen Solution... 7 2.2) Alternative Solutions... 7 3) Existing Products... 8 3.1) Home Audio... 8 3.2) Commercial Equipment... 8 4) Task Breakdown... 9 5) Budget... 11 6) Schedule... 12 7) The Team... 14 James Lu... 14 Brandon Pun... 14 Dave Steele... 14 8) Concluding Remarks... 15 9) References... 16 Team Power, 2004 iv
1) Introduction The purpose of our project is to design and build a prototype speaker/amp system that includes DAC, amp, power supply, and speaker all in a single box. We believe that this project has an advantage over existing mainstream audio systems in that it reduces line interference by running digital signal to the speaker as well as the advantage of delocalizing the power amplification stage in a large speaker setup. This is useful in situations such as concerts where central amps can be a large and laborious piece of equipment to have to set up and take down. Figure 1 gives a high level view of how our system will operate. Figure 1: Concept Illustration This project is heavily R&D oriented, requiring several iterations of design and prototyping before a product can be produced that is close to what we hope to finally accomplish. In order to make our logic more customizable during the prototyping phases, therefore, we will be using a Texas Instruments 6211 DSP board to implement the required DACs, ADCs, and logic required to generate a timing waveform for the switching power supply. Team Power, 2004 5
1.1) First Prototype Our first prototype is a proof of concept in the truest sense of the words. For the first prototype, our goal is to: 1) Build the power circuitry for a Boost-type switchmode power converter that draws power from the 5V logic rails of the development board. 2) Connect the DSP board in such a way as to act as a controller to the power converter while receiving an input signal from the output of the power supply and an analog audio reference signal of +/- 1V. 3) Write software for the board that will drive the power converter at a switching frequency of 100 khz and vary the duty cycle with respect to the audio input signal. 4) Drive a resistive load with a +/- 5V, 1/2 watt signal. 5) Measure the frequency response characteristics of the system for input frequencies of 0-1kHz. 1.2) Second Prototype For our second prototype, we hope to make the following modifications to the above described system: 1) Isolate the power supply from the DSP board using a 9V battery as power source and optoisolators to isolate the control signal. 2) Increase the switching speed to 500 khz. 3) Modify the power circuitry as required. 4) Measure the characteristics of the new system. Subsequent prototypes will involve higher power systems and modifications in the control logic to improve the quality and efficiency of the output signal being generated. Ultimately we hope to replace the development board with cheaper, less-flexible control logic so that the system can be marketed as an affordable stand-alone product. From this point forward, anything outside of the second prototype shall be considered outside the scope of this four-month course. Team Power, 2004 6
2) Design Solution 2.1) Chosen Solution Our solution to the proposed prototyping goals defined in Section 1 is summarized in Figure 2. As each prototype version is implemented, only minor circuitry changes in the Boost Power Converter need to be made. The majority of the changes in implementation will take place as software changes to the Control Logic. We hope that this will enable us to move more quickly through an iterative design cycle and arrive at an optimal configuration before committing to a final circuit design. 2.2) Alternative Solutions Figure 2: Block Diagram The other obvious class of solutions for our design goals is to implement the circuitry without the aid of the DSP development kit. This would involve purchasing an existing boost power converter control logic IC and implementing it with custom logic circuitry. Alternatively, control circuitry could be attempted using a more standard third party FPGA such as something from the Altera chip series. In the final market version of our product, we hope to implement our control logic using the first method discussed above, as we believe it would be the most affordable solution from the consumer s point of view. For the scope of developing this product as an ENSC 400 capstone project, however, we will confine ourselves to the method discussed in Section 2.1 Team Power, 2004 7
3) Existing Products Current audio solutions typically consist of speakers being powered by an inefficient linear amplifier. The high power dissipation and large size of the linear amplifier means it has to be mounted in a chassis with a large power supply. In our design, by using a switching design and integrating the amplifier with the power supply, much higher efficiencies can be realized. This means smaller size, lower power consumption, and less heat dissipation enabling the amplifier to be mounted right into the speaker cabinet. With the amplifier mounted in the speaker cabinet, a low power digital signal can be sent to each speaker opposed to the conventional high power analog signal. The digital signal is less susceptible to line noise and interference, which ultimately improves sound quality. 3.1) Home Audio A majority of the consumer market is geared in this direction when it comes to buying audio equipment. Currently, these components are quite bulky because audio receivers must contain an inefficient linear amplifier in addition to the signal processing circuitry. However, with products such as flat-screen LCD and plasma TVs, the trend is to smaller components. Also, despite the fact that receivers nowadays are able to handle digital signals from components such as a DVD player, it still converts the signal to an analog audio signal before sending to the speaker via long wires. Hence, it is fairly easy for the sound to become distorted due to interferences causing line-noise in a typical home. 3.2) Commercial Equipment Size and weight are an important factor as the equipment is constantly being moved around which is the case with concert audio equipment. In addition to the speakers, the most notable piece of equipment is the Media-Track, which houses many audio amplifiers, power supplies, and other audio processing equipment. Hence, this is a massive, bulky, and heavy piece of equipment to move around, and this is done on top of moving the speakers, which are already large themselves. Commercial audio equipment that isn t moved around such as the sound systems used in auditoriums could also benefit from our product. Having the amplifiers in the same enclosures as the speakers would eliminate the need for dedicating space to house the amplifiers. Team Power, 2004 8
4) Task Breakdown We have broken our project down into the following tasks, each with an accompanying time estimate and assigned primary person. Sections 3 and 4 discuss the cost and scheduling in further detail. 1. Research Given the time constraints of our project, the research time had to be deeply scrutinized. Hence, the tentative research time was broken down as below to allow for adequate research as well as time to do the rest of the project. The majority of the time emphasis is placed into the switching power supplies and interfacing the TI development kid, as we are not very familiar in these areas. a. Learn about SPDIF audio format 4 hours (James) b. Learn about switching power supplies 8 hours (Dave, James) c. Learn about TI development kit 16 hours (Brandon, Dave) d. Research what parts are on the market and where they can be bought 6 hours (Brandon) e. Misc. Research 4 hours (James) 2. Design Design time of our product is as shown below. Again, a majority of the time will be focused on the power supply circuitry and the TI development kit. The SPDIF and DAC circuits are of less importance, as they are not required as stand-alone circuits for the first two prototypes as defined above. They are still relevant, however, if the project is to be taken to completion as a marketable product. a. SPDIF decoder & DAC Theoretical Design 4 hours (James) Design a circuit to decode the SPDIF input signal into PCM then convert that to an analog signal which can be input into the amplifier. b. Power Supply Circuitry 8 hours (Dave) Design a variable, switching power supply that can provide enough power to power the amplifier. It also needs to have the appropriate voltage output to power the SPDIF decoder & DAC. c. TI Development Kit 32 hours (Brandon, Dave) Write initial control algorithm. Team Power, 2004 9
3. Parts Ordering 4 hours (James) This is somewhat concurrent with Research and Design. This involves ordering the parts determined to be needed in the Design stage from the sources found during the Research stage. It also involves tracking the parts until they arrive finding alternate sources for parts if needed. The 4 hours allocated are just the physical act of ordering. Monitoring the shipment status should be ongoing. 4. Prototype 1 - Implementation & Unit Testing Again, the implementation work hours are divided much like the design work hours, for the same reasons. a. Connectivity of Devices Building the power supply circuitry, connecting it to the controller, and writing code to simulate an input signal. 16 hours (Dave, Brandon, James) b. Results Measurement Measuring frequency response of system given input of digitally encoded sine waves of various frequencies. Assessing distortion, SNR, and interference in the system. 16 hours (Dave, Brandon, James) 5. Prototype 2 Implementation and Unit Testing Essentially the same procedures will be repeated as with the first prototype, after making the required changes to meet the specifications of prototype 2. 32 hours (Dave, Brandon, James) 6. Debugging, Fixes, and Reports Based on comparison to existing speaker/amp configurations, possible improvements will be determined and strengths and weaknesses of the system will be documented. During this stage, we hope to acquire as much data as possible from the system so as to make our efforts worthwhile intellectually, if it turns out that they are not realizable in a marketable product. This is a largerscale iterative stage of integrating fixes and improvements and testing. 24 hours (Brandon, Dave, James) Team Power, 2004 10
5) Budget All costs are in US dollars and do not include shipping. Shipping costs are estimated at an additional $20. At this time, we have not included the price of speaker enclosures or drivers. We will use speakers we already own for testing purposes. Item Cost Estimated Arrival Time Manufacturer SPDIF to PCM converter [1]* $3.80 2 weeks Digikey DAC [1]* $3.67 2 weeks Digikey Board Fabrication [2]* $50 4 weeks Dyco Breadboards (for prototyping) $20 1 week Radio Shack Power Supply Parts [3] $40 2 weeks Digikey Cables $10 1 week Radio Shack Misc. $23 3 weeks Various Places * These parts would be required for final market product, but are not needed in the initial prototyping stages. Subtotal: $200 Shipping: $20 Total: $220 Team Power, 2004 11
6) Schedule Refer to Section 2 for the description of these hours. Research : Task Dates SPDIF decoder & DAC Jan. 16 Feb. 1 Switching Power Supplies Jan. 20 Feb. 1 TI development kit Jan. 20 Jan 25 Market Availability Jan. 20 Jan. 25 Misc. Jan. 16 Feb. 1 Design : Team Power, 2004 12
Task Dates SPDIF decoder & DAC Feb. 2 Feb. 6 Power Supply Circuitry Feb. 2 Feb. 9 TI Development Kit Algorithm Feb. 2 Feb. 9 Parts Ordering : Jan. 20 Jan. 30 Prototype 1 - Implementation & Unit Testing Task Dates Connection of Devices Feb. 9 Feb. 18 Results Measurement Feb. 9 Feb. 22 Prototype 2 - Implementation & Unit Testing Task Dates Connection of Devices Feb. 23 Mar. 3 Results Measurement Mar. 4 Mar. 14 Debugging, Fixes, and Reports: Mar. 15 Apr. 1 Team Power, 2004 13
7) The Team James Lu I am a fourth year Electronics Engineering student dedicated not only to my studies, but to maintaining an active, entertained and relaxed lifestyle as well. Hence the reason for my interest in tinkering with computers and various audio equipment parts. I am familiar with the basic principles of power supply operation and op-amp operation. My experiences also include PCB board layout design. Most notably, I am competent with bookkeeping, and have access to a small variety of people for information regarding the field I will be working in. What I hope to bring to this project is a strong commitment to keeping our focus on the immediate task at hand without losing our vision of what we hope to accomplish. Brandon Pun I am a Systems Engineering student at Simon Fraser University who s interests include computers, music, and audio equipment. My previous work and coop experience includes network administration and software testing. I have experience with C/C++ and Java as well as HC12 assembly. I am familiar with lab equipment such as function generators, oscilloscopes, power supplies, and DMMs. I have knowledge of circuit board design with Eagle as well as fabricating single layer boards. Dave Steele I am in my fifth year of Systems Engineering, and interested primarily in low-level design solutions. I believe that engineering is an art form that only a small percentage of our population is granted the privilege of working with. What I hope to bring to the table for this project is a sense of style and functionality with a higher purpose. There are many solutions to any given design problem, but it is the creative ingenuity forged by a strong group dynamic that turns a mere solution into a thing of beauty. Team Power, 2004 14
8) Concluding Remarks By the third week of February, we hope to complete our first working prototype as defined in (Section 1). The goal is to spend the end of February and beginning of March analyzing our system and implementing the second prototype. We hope that by the end of the semester we are able to present a detailed report of our work procedure as well as our findings in terms of system performance and suggested practical applications (aside from those discussed above). It is our hope that even if the system does not perform well enough to function as a full-range audio amplifier, it will still have other potentially marketable applications, such as functioning only as a low-range (subwoofer) audio amp. Team Power, 2004 15
9) References [1] Digi-Key.com, January 2004, www.digikey.com [2] Dyco s PCB Prototype Service, January 2004, http://www.dycocircuits.com/html/dycoexpress.html [3] Design Notes.com, January 2004, http://www.designnotes.com/ CIRCUITS/ps2_02.htm Additional Resources: Parts Express, January 2004, http://www.partsexpress.com/pe/ pshowdetl.cfm?&did=7&partnumber=302-720&ctab=2 The SPDIF Audio DAC, January 2004, http://www.aca-vogel.de/audio/ DHA/spdifda_en.html Zetex Semiconductors, January 2004, http://www.zetex.com/frame.asp?page=http:// www.zetex.com/3.0/a2-16.asp Team Power, 2004 16