Barry Klein October 1997

It has been close to 15 years since Electronic Music Circuits was originally published. Since that time, many IC's have become obsolete due to advances in technology or product direction changes. The purpose of this booklet is to go through my original effort and provide advice for parts replacements, sources, and design suggestions on a chapter by chapter basis. I would appreciate any additions you may wish to submit for future versions of this booklet. As I am on a "print on demand" basis I can revise this material as needed. I decided to keep the original book intact in my reprinting because this way the material suggested herein will apply directly to both the original and reprinted versions. Also, as the original effort was not electronically conceived, the effort required to scan it, perform OCR, and then recompose the whole thing just did not seem justified for the expected interest. I suggest you read the book material first and then refer back to this section for possible corrections, suggestions, and/or updates. (I think I have been successful in making the schematic changes in the reprinted version)

I have a pile of reference material for new design approaches and IC's to use in electronic music designs. Some time in the future, I hope to create a "Book Two" that would include these new designs as well as replacement discrete designs for common modules that previously relied on custom chips. In this update booklet I can only refer to and suggest most of these as I have not had the time to try them out and finalize designs.

Finally, please pass on any influence the original book may have made upon your life and also let me know if any of the circuits you have built from it have been used in any published CD efforts (I'd like to hear them). Please pass on any corrections or suggestions for new material to me at :

barry.klein@deltronix.com or barry.l.klein@wdc.com

or by snail mail...... 32041 Pleasant Glen Rd.

Trabuco Canyon, CA 92679

You may wish to share your questions and suggestions with those of us on the following Internet mail lists:

Analogue Heaven Analogue@hyperreal.com

Bode Bode@veda.is

Synth-DIY synth-diy@horus.sara.nl

Electronica electronica@andrew.cmu.edu

Bode and Synth-DIY are lists concentrating on analog modular synthesis and design. Analog Heaven is more general; discussing analog synthesizers of all varieties. Electronica is a list open to discussions on a broad variety of topics dealing with

electronic music (it's ok to talk about samplers here).

This chapter throws a lot of concepts and terms at you that may get you all confused. I did not want to spend a lot of time on this material because there were already quite a few books out on these subjects already and I did not want to sacrifice any schematic content to rehash what was already covered in detail elsewhere. Unfortunately, not many electronic music theory books remain - especially any dealing with actual use of modular analog synthesizers. Your best bet is to look for books on electronic music in your local public or university libraries. Periodically a bibliography of related books is posted on Internet mail lists. Check the rear of this booklet for suggested locations to try.

Synthesizer design has changed greatly since 1982 when the book was published. I think I did not even mention the word "MIDI" in the whole book! And now MIDI as we know it may become obsolete - as XMIDI and/or ZIPI are being tossed about. Although analog designs continue to be manufactured the majority of emphasis in design is has turned largely digital, often using proprietary custom IC's and/or DSP. Yamaha helped begin the original shift in design with the introduction of the DX7 that utilized FM synthesis techniques featuring their custom IC's. Yamaha later simplified these design concepts for the popular FM synthesis-based sound cards. Yamaha had the market tied up on FM synthesis because of their patent coverage on the technique. Unfortunately for us homebuilders, Yamaha will only sell their FM chips to companies making sound cards - they ask you to sign a form stating you will not use them for standalone synthesizer products. Now that FM synthesis is losing favor to DSP and wavetable design approaches maybe they will "change their tune?" In any case, DX7 programming is not nearly as intuitive as with analog. Its main reason for its success was the wide range of new sounds that it provided that was difficult to get with previous methods. And it was a lot lighter than the Rhodes to transport from place to place! The DX7 was a very successful synthesizer even though it began as a monophonic instrument without effects such as reverb, chorus, or delay.

Several interim synthesis techniques appeared between then and now, but ROM-based sample synthesis has been the most prevalent. The Korg M1 was one of the more popular units to exemplify the possibilities of ROM-based sample synthesis. Actual sound waveforms of intruments etc. are recorded and normalized to form ROM (read only memory) wavetables for an embedded processor to access, modify with effect algorithms, and output to a stereo digital to analog converter. This synthesizer provided the musician with instant access to a large variety of high quality sounds. New sounds are added in the form of program cards or floppy disks.

In addition to the ROM-based synthesizers there are the RAM-based samplers. These units allow the user to create his/her own samples in RAM. These can be read from or written to floppy or hard disk. Still, these units lack much realtime control of parameters.

Synthesizers like the Kurzweil K2000 have samples in ROM form and allow the user to manipulate and combine them in various ways in a RAM work area. A sampler option allows the user to input totally new sample data (the unit can also read samples and .wav files from floppy disk). A unit like this is a great companion to an analog modular system as you can use the modular system to create monophonic sounds that can be played polyphonically after being sampled into the K2000.

Soundcards for the pc such as the Turtle Beach Pinnacle, the Creative Labs AWE32, and the Gravis Ultrasound MAX provide sampler and wavetable synthesis in a plug-in card. In some cases, files can be exchanged between the soundcard and the sampler synthesizer. I have an AWE32 that can record, play, and modify .wav files.

I do not really want to participate in an analog vrs. digital - which is best - battle. I think great things can be done with both. But I do think until recently, synthesizer manufacturers have forgotten that some of us that want to create sound rather than just "play" it. This is where I feel emphasis in new designs should be placed - on the intuitive user interface, so a user can experiment and create totally new sounds at whim.

To this end, we are now entering a new era in synthesis. Digital signal processing is being applied to synthesizer design, allowing design and/or modification of sound through software. Physical modeling is a technique where algorithms that represent the effect of resonant chambers in an acoustic instrument, the pluck or vibration of a string, etc. are applied to an input signal to modify its timbre. Morphing is a similar technique where the modeling can shift from one set of characteristics to another as the note progresses. DSP is also being used to model traditional and experimental synthesis techniques in an object-based manner on a computer. With programs like Turbosynth by Digidesign for the Mac you can graphically organize how you want modules arranged and control them in realtime. With the increasing success and ease of programming with programs such as Visual Basic for Windows, you will be able to modify or create sound in software or use the computer interface as a human interface for control of external hardware. This last concept may be where we can still experiment. It really isn't too hard to use a computer to read knob settings, memorize them, and apply them by D/A converter to one or more analog circuits. This is the basic concept behind the Prophet 5 but now we can use it on an expandable modular system. The recently announced Nord Modular is an example of this approach.

I want to throw out some suggestions for you to investigate if you are interested in using the computer as a sound creation and modifiction tool (also I am by no means an authority on these programs - I have only just dabbled with them).

Well, I haven't "dabbled" with this one (just haven't had the time). But from what I have read so far on the Csound mail list, it is a C based sound programming environment that is compiled for the desired platform - mac, pc, NEXT, Silicon Graphics, whatever. If you are into programming in C and want to experiment with mathmatical concepts in sound generation/modification this is your tool. Look for it at ftp.bath.ac.uk (/pub/jpff). Analog Devices announced a Sharc processor based development system for Csound that sounded

great at AES and NAMM. So far its vaporware though.

This program is a neat shareware .wav recorder/editor with some really cool features you will not find in other packages. The author, David Johnston, shares my interest in binuaral

beats and meditative sounds and has a very interesting support module that will impose binaural beat techniques upon your target .wav file. It also will allow you to create a very large file (based upon a shorter .wav file) that has a binaural beat frequency that ramps over time. There is also a minimal synthesis module, a CD player module, and several effects options. The program has evolved to the point that it is now a serious digital recording platform - included with some recent sound card offerings.

This is a program to read audio data off a CD (with a CDROM drive) in digital form and put it into a .WAV file. It's intent is to prevent any aliasing effects you may experience if you were to record the audio data with your soundcard's A/D converters. Unfortunately I get popping sounds with my Creative Labs AWE32. Even so, this program gives you a cool interface to narrow in on CD audio data and logical block addressing so you can use an alternate program to pull the data off - CDDA. Look for it in the CDROM.COM ftp site.

This program will also pull CD audio data off in digital form but it has a command line interface. Using DIDO will allow you to determine the exact location in sector counts/time of the data you want. Then you can type these data values into the CDDA

command line. It can be found in the CDROM.COM ftp site.

This program is a very full featured sampler/editor program with effects features, loop editing features, FFT displays, etc. etc. It is not cheap but is probably worth it if you do a lot of this kind of work. An operational demo version (without SAVE and a few other

things) is available from the manufacturer (Sonic Foundry) and can also be found in some net locations. Sonic Foundry, 100 S. Baldwin St., Suite 204, Madison, WI 53703

This program is a hard disk recoding program with some waveform editing features. Recorded audio is put into .WAV files. Files can be arranged in sequence like storyboard fashion. An operational demo version is available from the manufacturer or its representatives. Soundspiration Systems, 9704 White Ash Rd., Dallas TX 75249 214-298-DISK (US Representative).

This again is a hard disk recording program for Windows, allowing 8 tracks of audio on your system hard disk. It too has several nice editing features. An operational demo version is available. Note that some soundcards, such as all Creative Labs cards, will not allow simultaneous playback and record with this and similar programs.

Page 20 and Fig. 1-13 describe the top octave generator and its use in organs and string synthesizers. You may find circuits in the older literature that utilize these and begin looking for them in catalogs. One popular IC manufacturer of these was Mostek. But Mostek and the chips are gone.

I wouldn't fret over trying to find .1% 100K resistors for the 1V/octave input resistors! This is in error. Look for 1% resistors of 100K. Five percenters will do in a pinch. Also look for metal film or wirewound vrs carbon or carbon film as they are more stable with temperature and have lower noise. Look in the various surplus outlets for these as they can get pretty expensive as new parts if you're not a careful shopper.

I dished out the theory here quite a bit - a little more than was really necessary but what the heck....

Fig. 2-6 shows a schematic for a +/-15V supply that is fine for most applications. The 36V center-tapped transformer will be hard to find. You may have to resort to using two

single 18V transformers with rectifiers on each (as in book Fig 2-2b). Remember, for the negative supply, the positive output of the bridge will go to ground and the negative output will go to the filter cap and regulator.

You may want to look for larger "wall wart" supplies at computer swapmeets. Some I have seen provide an amp or so at +/-15 and 200ma or so at 5V.

I did not mention that there are fixed regulators available in smaller packages for low power applications. You may have applications where you would want to place this type locally on your module pcb. Examples of this type are the LM78L15 at 100ma and the LM78M15 at .5 A. The "L" type come in TO-92 transistor packages.

I neglected to mention the LM317 for adjustable positive supplies. They are cheap and great parts. If you need a simple adjustable bench supply this is the part to use. But

pay attention to that pin diagram!! (LM7815, LM7915, and LM317 pinouts below)

Look at National Semiconductor's voltage regulator databook for additional suggestions for regulators and example circuits. Linear Technology and Maxim also make quite a few.

In book Fig 2-11 I give a diagram of the LM399 voltage reference. But I barely touched on where/how this would be used. The exponential converter circuitry used in VCO's and other modules is sensitive to variations in voltage. This variation could be a result of varying load on the supply or variation in temperature inside the synthesizer cabinet causing variations in the regulated output. A voltage reference is best used in these critical circuits to eliminate any variation. The heater in the LM399 heats the part above the expected worst case temperature and thus removes the variation due to temperature factor. Voltage references are spec'd by current output as well as stability with temperature. The LM329 reference is not as stable as the LM399 as it lacks the heater. Analog Devices also makes several voltage references with good specs (as the REF19x series).

Voltage References:

The LM329 and LM399 are described in National Semiconductor's Data Acquisition Linear Devices Databook.

I don't really think output crowbars are necessary for modular systems. They typically short out the output of the supply if an overvoltage situation occurs. This depends on the supply shutting down from the increased load. So the crowbar has to be able to handle the full supply load without frying and the supply has to be able to respond to a short without failing. Sounds like a lot of stress on components to me. Better off just putting fuses on your AC line and maybe the DC supply outputs.

I don't know how many of you will be going to all the trouble of making keyboard interfaces for your system. You don't see many keyboards available as surplus items any more. Rather than have you butcher an old monophonic analog synth just for the keyboard, a better approach would be to make or buy a MIDI to CV converter. Several

examples of these have appeared in the magazines, I remember one circuit in Electronic Musician magazine and another in Elektor. John Simonton of PAIA is putting one together in kit form that should be available soon. There are also some commercial units available as well as mod kits for older non-MIDI synths. In any case, the simplest unit will output a control voltage, gate, and trigger corresponding to a note played on a MIDI keyboard driving it. You can usually find a MIDI synth of some form for less than $250 if you look long enough.

Ok, you want to make one anyway. Book Fig. 3-3 shows a single bus monophonic keyboard interface. First of all, opamp A2 should have the + input connected to ground in the original printing. Next, don't hesitate to substitute the recommended LM307, etc. opamps with alternative parts. The LM307 was a common part back then, a little better than the LM741 that has popular beforehand. Now we are deluged with a huge variety of opamps. When chosing an opamp you need to understand how it will be used. Some opamps are low noise, some have a fast slew rate, some feature rail to rail outputs with no phase reversal, some are low power, some are single supply, some work on 3.3V. In this circuit, the most critical opamps are A5,6, and 7. If its input bias current is high on A6 it will rob current from the storage capacitor and as time goes by the CV will drop. Not a good thing. You may also want to look at CMOS 555 timers for use in this circuit. They will use much less current and reduce noise on the board.

The duophonic circuit in Figure 3-4 is more complex. Note that a 3 bus keyboard is required. These aren't easy to come by. Remember to add filter capacitors on the supply lines to ground near each IC to maintain noise immunity.

If you have become somewhat familiar with analog to digital and digital to analog converters you may want to try replacing the sample and hold in the above circuits with them. This way you will be eliminating any voltage droop inherent in the analog sample and hold. You could also use a microcontroller along with one or more digital to analog converters to make a duophonic or even polyphonic keyboard interface.

The AR and AD circuits of book 3-10 and 3-11 need a 3.9K resistor just before the junction of the two diodes (between "Q" and D2/D3 on fig. 3-11). On figure 3-12, the Q and /Q labels are reversed.

SSM2050 Fig. 3-14

The SSM2050 is no longer available. The SSM product line has ended up as part of Analog Devices. Analog Devices makes only a few remaining SSM products - mainly a few VCA and preamplifier chips.

Fig. 3-16 CEM3310 Envelope Generator

Feel free to substitute alternate opamps for the listed LM348 and LM1448. Almost anything will work in these locations. The supplies of CEM3310's are drying up so

good luck with this one.

Fig. 3-17 10 Step Sequencer

The top of R1 (1M Rate pot) should be tied to pin 3 of the 555, not +15V.

Fig. 3-19 the noise source transistor is connected incorrectly. Ground the base of the transistor and connect C1 to the emitter (not the base). The collector remains floating.

Feel free to substitute op amps here too.

Fig. 3-21

I believe the National MM5837 is no longer made. Instead try the MM5437 which runs

off a positive supply in the range of 4.5-11V. It is described in National's Linear Application Specific IC's Databook.

Fig. 3-25 Sample and Hold

Note that the external input comparator A4 is set up for a signal varying about ground (like a VCO +/-5V output). You may want to connect the "+" input of A4 to the wiper of a pot connected between 5V and ground. You could then adjust this pot for your desired trigger level to accomodate 0-5V control voltages/signals.

Fig. 3-32 - The 4050 was misdrawn here - it should look like a non-inverting buffer.

Fig. 3-35 - I didn't go into how this circuit is used. This circuit is the heart of an external signal processor. It will derive a trigger, gate, and envelope from an external signal. So if you want to make an audio input act as a trigger or control voltage for a filter or something, this is the circuit to use as a front end. It does not do pitch to voltage conversion so you can't input a guitar or voice signal and use it control VCO's. But connected to a high resonance VCLPF it will give you an envelope controlled wah-wah kind of effect (just what you need!).

Accurate stable VCO's are critical in your modular synthesizer project. I recommend that after you finish the power supply a VCO be your next module to build. This is because it is very satisfying and encouraging to get something going that actually makes sound right away. The custom SSM or Curtis/OnChip Systems VCO's are your best bet if you can find the parts. Unfortunately many of the book circuits are obsolete because of unobtainable parts. So what do you do? Well, I have not yet experimented with substitute discrete VCO designs but I will give you some suggestions if you are hot to try something.

Q81 Tel Labs Resistors

Starting with Figure 4-2 you see a part listed as a Q81 Tellabs temperature compensating resistor. Tellabs has gone away as a source for these but they are still available. See the source listing in further along in this booklet for details but resistor manufacturers like Vishay and others still make these. You will end up having to live with a minimum order that may be a bit on the high side. Raise a stink on Analogue Heaven and someone may

sell you a few.

Fig. 4-3 Thermostat Circuit

The CA3046 has been used in commercial analog synthesizers with a heater/thermostat circuit much like this one. The substrate pin (13) must be at the lowest potential of the circuit (ground). So study this circuit and plan to apply it or something similar in just about any exponential converter circuit where drift is undesirable. The theory here is to heat the subject transistors to a temperature high enough that the ambient temperature will never approach it. The thermostatic circuit can easily and accurately maintain this temperature. So after this is established you don't want cooler air currents to test things so it would be advisable to put a thin layer of Styrofoam around the transistor array. When seeing this circuit you (as I) may initially think it a good idea to tie the critical transistors etc. to a common heat sink so a common temperature is maintained. The problem is that this heat sink takes time to stabilize at the high temperature. You are heating the components as well as the heat sink. It is better to isolate the heated assembly from any heat sinking components/surfaces and surround it with a thin Styrofoam. This way it will heat up rapidly and the Styrofoam will isolate it from the environment. Maybe

use thin longer leads on these parts so the pcb doesn't act as a heat sink. Again, I haven't

tried this yet but it sounds like a good plan.

Fig. 4-7 and 4-9 Discrete VCO (AD821/818 transistor pairs)

The AD821 and AD818 are no more. Even their temporary recommended replacements from MicroPower (MP310/11/12) are out of production. So what to use? Well, some circuits have used the CA3046 (NPN) but most builders report rather poor performance. Check out National Semiconductor's AN-299 and Linear Technology's LT1055 data sheet for examples design examples. I would suggest trying the matched transistor arrays from Analog Devices like the MAT-03 or SSM-2220 (PNP) or the MAT-01, -02, -04, or SSM-2210 parts (NPN). All but the MAT-04 are dual transistor pairs so if you want to make a heater circuit you may want to use discrete parts epoxied to the matched pairs and again surrounded by Styrofoam. (Don't get it so hot it melts the Styrofoam!) You may want to try wrapping the transistors with a few wraps of aluminum vent-sealing tape and then covering this with the Styrofoam. See the table in a following section listing several transistors and their characteristics with further suggestions for replacements.

AN-299 Audio Applications of Linear Integrated Circuits

LT1055 Data Sheet/Application Note 14

Jim Williams of Linear Technology wrote both of these application notes. I wrote him about them and he gave me a little background on them. The AN-299 VCO utilizes the CA3046 transistor array. He stated that even if the exponential transistor was ideal the

performance would be degraded by the ramp reset time. His second circuit in LT1055/AN14 notes utilize a charge pump IC - the LT1043 - to eliminate the dead time errors. He thinks this circuit has about a 2X improvement in exponential performance.

One idea suggested to me (that I haven't seen applied) is to use the VCO square wave

output to trigger a monostable that feeds an integrator circuit. The output of the integrator could be fed back to the control voltage circuit as a form of negative feedback to help counteract effects of reset delay, etc. I'll have to play with it!

Custom VCO Chips

The custom VCO chips really are the way to go if you can find them. Several popular synthesizers used these (and some current units still do). OnChip also had the 3374 Dual VCO and the 3394 voice-on-a-chip in their product line. Check with Paul Schreiber at www.synthtech.com for possible VCO kits.

Fig 5-15 SSM2040 VCLPF

Although the SSM2040 may be hard to impossible to find, take a look at the VC resonance circuit using the LM13600. You should be able to adapt this to alternative circuits if you desire. Note again that the Q81 Tellabs resistor can be obtained from

Vishay and a few other sources.

Fig. 5-18 SSM2044 VCLPF

C1, R2, R3, and C2 should tie to ground.

Fig. 5-21 Phase Shifter

Vactec is now an EG&G company. These parts are still available (try Newark Electronics). I got a copy of the Optoelectronics Data Book from Vactec and found that these parts show quite a bit of distortion as the signal level rises. To be honest, my circuit sounds kind of plain (not swirly like the MXR100 effects box). This could be partly due to a slightly different signal path. Also, way back then, a Vactec sales guy told me that they sold selected and matched parts to MXR. The circuit is definitely unique sounding and worth building. Fixed settings sound similar to cabinet simulator circuits. I have been playing with this circuit on a circuit simulator and will be trying a few new configurations later on. Use quiet op amps in this circuit as noise gets quite objectionable if you don't. The vactrols exhibit slow response and decay times. For slow sweeps you won't notice any problem but at faster rates the depth will lessen. FET's or OTA's can also be used for phase shifter circuits. I plan to build some of these in the future. Quiz people on the AH list for suggestions in the meantime. Siliconix has a data sheet on using FET's as voltage controlled resistors. (AN105) Call their faxback number at 408-970-5600 for faxback code number 9105 or voice at 800-554-5565 to have it sent in the mail.

Fig. 5-24, 5-26 VCF's using OTA's

You might want to try using 3280's here instead of 3080's. They are lower noise and

have better performance/specifications but maybe that is part of the charm(?) of 3080's. Again, the AD821 is no longer available. Look at matched pairs from Analog Devices or other alternatives. Frequency tracking may not be as critical in filters (unless you are increasing resonance to oscillation and using them as auxiliary oscillators). To get going you could match a pair of 2N3906's and later look into replacing them with a better alternative.

Fig. 5-28 CEM3320 VCF

R3 ties to -15V, R1 goes to the diode bias pin - pin 2, R50/51 go to pin 12 of the CEM3320, not pin 7.

B&B Audio 1538

I haven't heard or seen anything on this chip since then. These chips were handled by Aphex but the number I have for them is no longer good.

Reticon is now an EG&G company. Unfortunately the SAD512 and 1024 are no longer available. Reticon superseded these with the RD series of chips and finally, Reticon has ceased production of analog delay line IC's. Panasonic still makes several analog delay parts (they are quite popular in surround sound decoder circuits) and I guess has the market all wrapped up. Their chips are listed in the Digi-Key Electronics catalog.

Note that the vc clock of Fig. 7-5a is pretty cheap and useful. (Just wanted to bring it to your attention)

I think these things have some great possibilities. Could be as useful as a VCF. You may want to build one or more of these right into your VCO module.

Fig. 7-12

Oops! The floating "-" input of the opamp (in the dashed line box) should be connected to ground. This circuit can get to be a lot of parts. You may want to build it on a breadboad first to hear how it sounds. You may want to use less sections to save cost.

Bernie Hutchins

1 Pheasant Lane

Ithaca, NY 14850

Bernie Hutchins began publishing Electronotes back in the early '70's. A full collection almost takes up a full bookshelf of material. Many a synth-builder and future engineer/technician got his/her start with EN. Bernie continues to sell complete collections of EN for around $300 and a "Preferred Circuit Collection" of the most popular stuff for around $35. Well worth getting.

6400 Hollis Street #12

Emeryville, CA 94608

510-653-3307

This magazine has a DIY article once in a while. Check out the recent PAIA

Fatman and tube preamp projects.

411 Borel Ave., Suite 100

San Mateo, CA 94402

415-358-9500

I don't remember any DIY articles here. The articles are much like those in EM - reviews

of the latest technologies and instruments.

Future Publishing Ltd.

FREEPOST(BS4900)

Somerton, Somerset TA11 6BR (U.K.)

In the states this magazine goes for $6.50 - 8.00 if you can find it. For some reason they don't have very good distribution here. This magazine has more emphasis on older equipment than the U.S.magazines. It also comes with a CD with sometimes interesting demos of new equipment and overseas bands.

PO Box 30, St. Ives

Cambs PE17 4XQ (U.K.)

This magazine also puts more emphasis on older equipment. A lot of ads too.

MIT Press Journals

55 Hayward Street

Cambridge, MA 02142-9902

Unfortunately this magazine is displayed in the stands in a plastic wrapper that you are not supposed to take off to see if it is worth the $10+ asking price. As its title indicates, many articles on computer music systems and theory. Many university libraries carry this magazine. If this magazine went for under $7 a copy I'd get it every time it comes out. But alas, it is pretty expensive and an extravagant purchase for me.

L333

Quadrant House

The Quadrant, Sutton

Surrey SM2 5AS (U.K.)

This magazine has very high quality articles on electronics theory. Often has schematics you might be able to use for music purposes. Goes for around $6.50 in the states.

These magazines focus on beginner and intermediate level electronics interests. Occasionally you will see an electronic music related project like some of those

designed by John Simonton of PAIA. You can find these magazines at just about

any US supermarket magazine stand.

4 Park Street

Suite 20

Vernon, CT 06066

203-875-2751

If you want to get into microcontrollers this magazine deals with them quite often. I get the feeling that all of the authors are really digital types and analog is a little bit of a pain in the butt for them. Good bunch of guys though. Correspond with them on their BBS at

203-871-1988. It is a great resource if you are having trouble with the theory behind something or want to know how best to approach a design problem.

430 Princeland Court

Corona, CA 91719

909-371-8497

909-371-3052 fax

This is a large (13" x 10 1/2") format magazine that originally just had electronics related want ads but has grown in popularity (at least here) and now has some pretty good construction articles dealing with the PIC microcontroller, laser projects, robotics, and more. Don Lancaster has a column in it with worthwhile source lists. Also, this is a great place to look for surplus component buys.

This is a bible you NEED. There have been several editions over the years and I'd say look for the OLDER ones first as they had material that has been deleted in the more recent editions. But the newer editions have some new parts listed (but not as much new material as he could have put in). Walt is pretty busy with Analog Devices as Applications manager. He travels around the country (world?) giving seminars on analog design. Analog Devices then sells the seminar manuals seperately and you may want to get ahold of one or two of these. I have the following:

Practical Analog Design Techniques (just out)

Analog Devices also has new product applications guides that are worthwhile. Get a copy of their Non-linear Circuits Guide before it goes out of print. It has several discussions on exponential and log conversion circuitry, multipliers, dividers, and other neat concepts. Analog Devices' worldwide headquarters phone number is 617-329-4700.

Analog Devices has quite a bit of support literature/books on their various DSP chips. Perhaps you want to dabble with one of these? Well, they have demo/development kits called the EZLAB kits that allow you to use a pc for control and program development. Also, they have developed a sound card design that several companies have manufactured under their own label. This sound card uses the 2115 DSP chip. Analog Devices has support software for this card/chip on their DSP BBS at 617-461-4258. If you have not been exposed to high level calculus you are probably better off not attempting to experiment here - I know it is beyond my capabilities anyway.

Texas Instruments

These were given to me by a TI sales guy. They have some pretty good data on linear applications theory and use of the latest analog chips in real world applications. Not a cookbook but some good stuff for free if you talk to the right guy at TI. The Advanced Linear book describes the various types of op amps, where they are best applied, and the

noise and performance characteristics of each.

Texas Instruments

Like National's linear applications book, this book covers a variety of linear applications and discusses the advantages of the BiFET type of op amps.

TTL Cookbook and CMOS Cookbook by Don Lancaster

These books are must have's although technology is advancing beyond the material within them. Don needs to give them a serious rewrite but still the basic stuff is there and you need it! All the basic circuits for digital things you want to do. Trigger circuits, gate circuits, dividers, counters, 555 timers, etc.

Troubleshooting Analog Circuits by Robert A. Pease

This book is really good reading. If you want to be a future analog "guru" here is the master! Loads of hands-on theory, tips and test methods, things to watch out for, construction techniques, etc. If you have seen his column in EDN then you know this is worthwhile having. This is a book you can read over and over again.

Analog Circuit Design by Jim Williams

Another book by an analog master! In the same series as Bob's book, this too is full of great information and tips. And Jim likes to dabble with voltage to frequency converters (which have some directly applicable stuff for our modules).

The Art of Electronics by Horowitz and Hill

This book is probably the first one you should get if you want to learn electronics theory. I really don't know how they did such a good job on it. It is just full of great information and theory. It covers discrete components, semiconductors, op amps, digital IC's, microcontrollers, sensors, whatever. There is also a Student Manual that goes along with it that has added information - get both.

National Semiconductor

Customer Response Center 800-272-9959

This book has a slew of application notes dating back to day one covering just about any linear type of circuit you could wonder about. Lots of material relating directly to modular system design. In fact AN299 has a VCO, VCA, mic preamp, panner circuit, and bandpass circuit designs.

Linear Tech.'s data books and applications books are not completely updated and replaced year to year. So you may need databooks dating back to 1990 to get info and app. notes

on their full product line. Jim Williams (see book listed above) has authored many of the app notes. Several of the circuits are state of the art - spare no expense - types of circuits to get the best performance possible. Great articles on test equipment use, prototype construction, and neat pictures of old Tektronix scopes and analog point to point circuits the way they should be done (you'll be amazed). Headquarters: 408-434-0507.

I want to give you some advice on obtaining databooks. You may see them in bookstores and may have actually purchased one or more there for $10+ each?? Well, if you talk to the right person they will be free. First you need to do a bit of research to find out who the distributors are in your local area as well as the company's local sales office. I have just walked in to the various local sales offices and asked for databooks and received them. Get your story straight before you go in though. You don't have to work for a large company to get them but it sure helps. Usually if you have taken the effort to appear in person at their office they realize you are really serious in learning of their company's products and you'll get what you want. What motivates them to supply you with books is the possiblity of making money from your efforts (duh!). So don't just say you want something to read. Instead, say you want to make voltage controlled oscillators and sell them worldwide over the Internet. You've heard they have great parts and want to get data on them. Or say you want to do an article for Electronic Musician on mixer design and want to research suggestions for op amps to use. Don't come in looking like a slob. Also, you could call them up and get a list of local distributors. Go to the distributor on your/their lunch hour and ask to see the databook room. They will usually just point it out and then leave for lunch (with you in there piling up books). I've come out with a stack of books 4-5 feet tall before! Things are rough for the salespeople as there are not as many new businesses popping up as before. There just isn't the call for databooks as there used to be. Here are some vendors to target, in approximate order of importance:

National Semiconductor - Linear Appl. and data books, Voltage Regulators, Linear Appl. Specific Databook

Linear Technology - all databooks and appl. books.

Analog Devices - Linear Design In Reference Manual, anything else you can get

Texas Instruments - Linear Circuits databook

Burr-Brown - Appl. Handbook and Integrated Circuits databooks

Harris - Linear and Telcom ICs, CMOS Integrated Circuits, OEM Schedule (pricing!)

Philips - General Purpose Linear Databook

Maxim - New Releases various volumes

Raytheon - 199x Data Book

The following is a list part suppliers/distributors that deal in new and/or surplus parts and usually have catalogs. Be sure to look through the various magazines like Nuts and Volts, Popular Electronics, Electronics Now, etc. for additional sources.

P.O.Box 567

Van Nuys, CA 91408-0567

800-826-5432

818-904-0524

Good surplus electronics catalog. Switches, pots, IC's, connectors, and small

pcb's that do weird things.

7410 Pebble Drive

Fort Worth, TX 76118

800-433-5700

A large electronics distributor with several regional office locations. 1000+ page catalog.

Good source for Analog Devices IC's and Vactrols.

P.O. Box 3047

Scottsdale, AZ 85271-3047

800-528-1417

602-464-2485

Catalog of a variety of components, tools, computer boards, etc.

701 Brooks Ave.

P.O.Box 677

Thief River Falls, MN 56701-0677

800-344-4539

Very much like Mouser in that they have periodic catalogs, are a good source for discrete components and IC's and ship promptly.

PO Box 5408

Scottsdale, AZ 85261

602-451-7454

Fun catalog of miscellaneous surplus items. Often has great deals on switches, pots, IC's. Often sells complete subassemblies at good prices.

17921 Rowland St.

Industry, CA 91748

800-669-4406

Stocks a large variety of inexpensive printed circuit protoboards.

18 Canal St.

P.O. Box 122

Bristol, PA 19007-0122

215-788-5583

Interesting catalog of electronic and electromechanical stuff. Some power supplies.

1355 Shoreway Road

Belmont, CA 94002-4100

415-592-8097

A lot of computer related stuff but also several pages of IC's/semiconductors.

1850 South 10th Street

San Jose, CA 95112-4108

Catalog of IC's, components, and various electronics items.

P.O. Box 160113

Cupertino, CA 95016

408-996-3240

Looks like OEM surplus components. Several pages of IC's and discrete components.

Good pricing on trimpots and resistors.

10 Hub Drive

Melville, NY 11747

516-756-1750

Catalog of a collection of semiconductors, IC's, etc. Very inexpensive digital meters.

P.O. Box 12685

Lake Park, FL 33403-0685

407-848-8236

Good catalog of various surplus components and assemblies. Good source for power supplies.

650 Congress Park Dr.

Centerville, OH 45459-4072

800-543-4330

Good source for audio components and jacks. Stocks several hundred replacement type semiconductors. Stocks several types of replacement speakers.

2401 HWY 287 N.

Mansfield, TX 76063-4827

800-346-6873

817-483-4422

Reasonable prices on many parts. Several regional offices. Good source for new caps and resistors and other discrete components. Catalogs 4 or 5 times a year. Prompt shipping.

4801 N. Ravenswood Ave.

Chicago, IL 60640-4496

312-784-5100

Another large distributor with a 1000+ page catalog. Also carries Analog Devices parts.

Several regional offices.

3200 Teakwood Lane

Edmond, OK 73013

405-340-6300

405-340-6378 fax

One of the few kit manufacturers left that has musical electronics projects in their product line. Back issues of Polyphony (an old analog synthesis mag. for DIY'ers). Good source for proto-boards for the remaining Curtis/Onchip custom chips and also for inexpensive rack mount panels and enclosures. PAIA will be coming out with a new MIDI to CV product as well as a new modular system line.

P.O. Box 886

New Hyde Park, NY 11040-0311

A bunch of what looks to be OEM surplus parts.

10000 Canoga Ave., Unit C-2

Chatsworth, CA 91311

818-341-8833

800-824-3432

Smaller distributor. Good variety and pricing on IC's.

The following are current manufacturers of analog and digital IC's that you may wish to use in your system design. I give the headquarters address/phone number and related customer service information as well as one or more US distributors. Also I include notes on any interesting IC's they may have.

Allegro makes semiconductors and IC's intended mainly for industry and machine control type applications but there are some that maybe we could use. There are several hall-effect sensors intended to sense changes in magnetic flux fields cause by field interruptions by switches, gear teeth, etc. These could possibly act as keypress switches, velocity switches, etc. The model 3501 looks interesting as it is a linear response sensor - meaning it outputs an analog signal corresponding to the amplitude of the disturbance. Could we use this as a guitar pickup? Or a sensor for a theremin? In any case, if you are interested in hall effect sensors in any way, this databook has several pages dedicated to the theory and application of them and it is worth getting just for that.

Allegro Microsystems, Inc.

Integrated and Discrete Semiconductors AMS-501 (Databook)

115 Northeast Cutoff, Box 15036

Worcester, MA 01615

508-853-5000

508-853-7861 fax

street address:

One Technology Way

P.O.Box 9106

Norwood, MA 02062-9106

tel: 617-329-4700 fax: 617-326-8703

distributors: Hamilton/Hallmark, Allied Elect., Alliance, Pioneer Elect., Bell, Future

The UAF-42 is a monolthic implementation of the state variable analog filter (no voltage control though). Burr-Brown has a floppy disk available that aids in selecting circuit component values. Burr-Brown also has a Collection of Applications Bulletins as well as an Applications Handbook. More of the same theory/suggestions as in the competitive sources. Burr-Brown makes several high performance op amps, digital audio converters (used in Korg X/05rw series), and a few light to voltage converters.

Burr-Brown Corporation

6730 S. Tucson Blvd.

Tucson, AZ 85706

Mailing address:

P.O. Box11400

Tucson, AZ 85734-1400

602-746-1111

800-548-6132

602-741-6132 fax

street address:

6730 S. Tucson Blvd.

Tucson, AZ 85706

Distributors: Northstar Electronics, Insight Electronics

Crystal Semiconductor has several A/D and D/A converters that are useful for audio signal conversion. But also they have a series of chips utilizing wavetable synthesis intended for sound card applications. In the databook the application note for the CS8905 and CS9203 gives a pretty detailed description of such an implementation. The app. notes section of the databook covers delta-sigma conversion theory as well as the AES/EBU digital audio interface. There are a couple notes describing MIDI specification theory.

Crystal Semiconductor Corp.

P.O. Box 17847

4210 S. Industrial Dr.

Austin, Texas 78744

512-445-7222

800-888-5016

512-445-7581 fax

As with Xicor, Dallas has several digital pots in various configurations, single and dual. The DS1267 is a dual, no EEPROM, at 10K, 50K, and 100K values (256 positions). The DS1666 is a 128 position, single, at 10K, 50K, and 100K. The DS1667 is a dual 256 position 10K, 50K, or 100K with 2 on board op amps. The DS1668/9 is a single 10K, 50K, or 100K 64 step pot with onboard mechanical pushbutton switch and EEPROM. The DS1802 is a dual 65 step audio taper pot with pushbutton inputs. The DS1867 is similar but with EEPROM. A few more exist in the product line but you get the idea. They are described in the System Extension Data Book 1994/5.

Dallas Semiconductor

4401 South Beltwood Parkway

Dallas, Texas 75244-3292

214-450-0448

214-450-0470 fax

Manufacturer of vactrols - LED illuminated photocell modules. Very informative Analog Optoisolators databook.

street address:

10900 Page Ave.

St. Louis, MO 63132

tel: 314-423-4900 fax: 314-423-3956

Harris Semiconductor bought out RCA and so they are the source for the CA3080, CA3280, and similar CAXXXX parts. In addition to these, there are several two and four quadrant multipliers, and the ICL8048/9 log/antilog amplifiers (expensive).

Harris Semiconductor

1301 Woody Burke Road

Melbourne, Florida 32902

407-724-3000

800-442-7747 literature dept.

407-724-3937 literature dept. fax

407-724-3818 faxback service

street address:

1301 Woody Burke Road

Melbourne, Florida 32902

distributors: Arrow/Schweber Electronics, Hall-Mark Electronics, Wyle Laboratories,

EMC (Electronics Mktg Corp.), Falcon Electronics, Gerber Electronics, Newark Electronics

Source for a variety of matched pair transistors.

310 South Milpitas Blvd.

Milpitas, CA 95035

408-263-8401

408-263-7280 fax

Linear Technology makes several high performance op amps and has very good applications literature.

Linear Technology

1630 McCarthy Blvd.

Milpitas, CA 95035-7487

408-432-1900

408-434-0507 fax

Maxim makes a variety of IC's but the following are of special interest. The MAX274/5 are linear filter building block IC's. Applications literature shows methods of digitally controlling them with a DAC. The MAX275 has a higher max. frequency and a slightly different configuration. There is a software development/evaluation kit available for the MAX274. Maxim also makes several switched-capacitor filter circuits. The MAX038 is a waveform generator chip with selectable sine, square, or triangle waveforms selected by digital code. Internal and external sync is supported. The part has a 200ppm temperature accuracy. I haven't heard of anyone using this as the heart of a VCO yet. It was just released a few months ago. Note that it does not have internal support for exponential frequency control - you will have to build this yourself.

Maxim Integrated Products, Inc.

120 San Gabriel Drive

Sunnyvale, CA 94086

408-737-7600

408-737-7194 fax

800-998-8800 free samples and literature line

MX-COM makes IC's intended for voice and data communications, as in wireless and cellular phone networks. Their Mixed Signal IC's Product Handbook lists several IC's that may be of use to us tinkerers. There are sub-audio tone encoder-decoders that maybe could be used for switching effects over audio lines, voice filters and codecs, digitally controlled amplifiers (octal, dual, and quad), and finally the MX406 Phase locked filter/oscillator.

MX-COM, Inc.

4800 Bethania Station Rd.

Winston-Salem, NC 27105-1201

919-744-5050

800-638-5577

919-744-5054 fax

Motorola makes a wide variety of semiconductors and IC's. They have a few transistor arrays that I have not investigated, as of yet. Many applications notes are available on their products. They also typically have low-priced development kits for their microprocessors if you are feeling "digital".

literature dist. center:

P.O.Box 20912

Phoenix, AZ 85036

"Design line" - help line tel: 800-521-6274

National is one of the best sources for linear IC application support literature. They make a wide variety of op amps, references, voltage regulators, etc. Notable special function IC's include:

The LMC835 is a digitally controlled graphic equalizer chip

The LMC1982/83/92 are digitally controlled stereo tone and volume circuit IC's with

switchable inputs. BTW, unfortunately the Audio/Radio Handbook is no longer printed.

National Semiconductor

2900 Semiconductor Drive

P.O. Box 58090

Santa Clara, CA 95052-8090

800-272-9959 sales support

street address:

2900 Semiconductor Drive

P.O.Box 58090

Santa Clara, CA 95052-8090

tel: 800-272-9959

distributors: Hamilton/Avnet, Arrow Electronics, Zentronics, Anthem Electronics

What was once called Curtis Electromusic is now OnChip Systems. They continue to manufacture musically related IC's but desire to do business with large volume customers.

Stock of the most popular chips for our purposes (VCO's and VCF's) are pretty much

unobtainable. They gratiously allowed me to reprint the data sheets for their full line of EM-related chips as well as the two Synthesource brochures (good insight into discrete VCF and VCO design) if I withheld their address/phone number from this update booklet.

Please honor their request and look for alternatives (like discrete) for new designs. To

obtain parts to repair a synthesizer that is down, place a message on the AH or Bode lists and you may find a recommended source. But if you are going to make the next TB303

clone and want to buy thousands of these chips, search them out and give em a call. Paul Schreiber has some of these IC's available and plans to sell kits for them in the near future.

www.synthtech.com

Philips makes a variety of IC's including an 8051 family of microprocessors, various standard linear IC's and some interesting video IC's. The Philips TDA3810 is an IC that performs stereo spatial expansion. I guess it is used in "boom boxes" to extend the stereo imagine. Might be useful for a studio effect.

Philips Semiconductors

Literature Dist. Center

800-234-7381

811 East Arques Ave.

P.O.Box 3409

Sunnyvale, CA 94088-3409

800-227-1817

street address:

350 Ellis Street

Mountain View, CA 94043 7016

tel: 800-722-7074 fax: 415-966-7742

TI has a full range of linear and digital IC's. Check them out for their high performance

rail to rail op amps and be sure to ask for applications literature.

The TSL220 chip has a photodiode and a current to frequency converter all in one chip. It's output is a squarewave whose frequency is directly proportional to the light intensity it sees. You may wish to use this as the heart of some kind of unique light-sensitive controller or signal source. The TSL250/1/2 have voltage outputs that correspond to light intensity. Here to they may have some use in controllers.

Texas Instruments SN76494/6

These chips are discussed in my 1992 Linear Circuits Data Book Volume 3 but are I believe out of production as I had heard/read this before and there is no mention of them in my 1994 TI Pricing Guide. Unfortunate, as the chip was a cheap "noisemaker" as it had onboard tone generators and digital noise generator.

Customer Response Center 800-336-5236 214-995-6611

Xicor has come out with 8 pin DIP package chips that effectively are 32 step digitally-controlled potentiometers. One digital input is step up, the other step down. The X9511 are linear pots (1K or 10K) and the X9514 is a 10K log. pot. An onchip EEProm can store the setting and restore it on power application.

Xicor Inc.

1511 Buckeye Drive

Milpitas, CA 95035

408-432-8888 408-432-0640 fax

I called a few resistor vendors to find sources for the Q81 Tempco resistors used in the exponential converter circuits of several VCO's/ VCF's etc. I found these two (if you find additional vendors please pass on the information). Note that you may see reference to Vishay-Ultronix in Grand Junction, CO in National's LM394 datasheet as a source for Q81 resistors. It turns out that if you call them they will refer you to Vishay-Dale as below (same part #).

KRL

160 Bouchard Street

Manchester, NH 03103

603-668-3210

603-624-0634 fax

As of May 23, 1994 (now out of date):

P-2AQ-1001F TC Wirewound Resistor Quan. $ each

3 watt, 1K ohms, +1%, +3500 +300ppm/°C 25 $4.25

100 3.75

250 3.25

500 2.75

1K+ 2.25

$75.00 minimum order

Vishay (Dale Electronics, Inc.)

1122 23rd Street

Columbus, NE 68601-3647

402-563-6511

402-563-6418 fax

RS-2B-255 3W 3500ppm +350ppm 1Kohms Quan. Price

10-99 $3.72

100+ 3.25

2Kohms 10-99 3.25

100+ 3.53

I got the quote from the western regional office (May 1994) at 714-544-7024.

The following is a list of some dual and single transistors I have found that may be useful for exponential circuits. For the best results, the Analog Devices and Linear System parts

(see separate datasheet) should be considered. (Corrections and additions to this table

invited)