CGS Serge Wilson analog delay
CGS208 the Serge Wilson analog delay module is a faithful recreation of the Serge Wilson Analog Delay module. Only basic information is being presented here because the CGS version of the module was a closed release, due to the price and rarity of key chips in the design. This module is designed for +/-12 volt operation.
Construction
Please note that there is an error in the documentation and PCB for this project. The resistor marked 180k (R21) should in fact be 820k. Ken.
Before you start assembly, check the board for etching faults. Look for any shorts between tracks, or open circuits due to over etching. Take this opportunity to sand the edges of the board if needed, removing any splinters or rough edges.
When you are happy with the printed circuit board, construction can proceed as normal, starting with the resistors first, followed by the IC sockets if used, then moving onto the taller components.
Take particular care with the orientation of the polarized components, such as electrolytics, diodes, transistors and ICs.
When inserting the ICs in their sockets, take care not to accidentally bend any of the pins under the chip. Also, make sure the notch on the chip is aligned with the notch marked on the PCB overlay.
Connections to jacks can be established from the schematic diagram and from the Serge Wilson Analog Delay documentation.
Setup
There are two variants of the vintage WAD. Both use the same PCB. The only difference is the value of some resistors. The only official set up instructions refer to the older version. As well, there would have been local knowledge of corrections to which we are not privy.
The following readings were taken from a working late model GENUINE WAD. I am interspersing them with a version of the set up instructions in italics.
Very Important: before power is applied, turn Trimmer B fully clock-wise. Failure to do so before power-on may damage the 3080s. Do not TURN this trim pot fully CCW, or past the halfway point!
The following tests are to be done without the Analog delay chips installed.
Despite the above, as I was taking readings from someone else's working WAD, my tests were performed WITH the SAD1024 still in their sockets. The test document also assumed the board was not connected to any panel controls, where here it very clearly is. The only real differences this makes is instead of injecting external voltages into some pads, you adjust the pots to set the voltages. Initially the pots are set to output 0 volts. For the Filter frequency, this means turning it fully CW. For the Delay pot, it means setting it to the center of its rotation.
VC clock tests
1) check for a square wave at 15000 Hz +/- 40% with a 0 to +12 Volt amplitude at pins 4 and 12 of the 4050. (The period of the wave may be anywhere from 50 through to 70 through to 100 microseconds.)
If you do not get a square wave +/- 10% duty cycle, then check the CA3046.
2) check for pulses about 1.5 microseconds long (at 15000 Hz) at pins 10 and 4 of the 4093.
If these tests fail. check components in the VC clock section of the PCB.
Input VCF tests
1) feed the square wave from pin 12 of the 4050 into pad L (the input). The output at pin 1 of U3 should be almost a sine wave of about 5 Volts peak to peak. Adjust trimmer A if there are flattened spots on the wave. (If test 1. fails and the sine is more or less than 2.5 Vp.p check resistors and caps in the input VCF section.)
2) Input a variable voltage of 0 to -5 Volts at pad P and check that varying the voltage results in a smooth and undistorted attenuation of the sine wave.
I found the sine wave was attenuated to nothing at around -2 to -2.5 volts at pad P.
(If this test fails, first check resistors R57, R58 and R59 (47k, 22k and 1k) and the CA3046. Then check components of the input VCF section.)
3) Connect a voltmeter to the output as above (pin 1 of U3). Disconnect the VCF input and CV. Measure the voltage. Input +12 into pad D. Measure the voltage. The difference in the two readings should be -0.190 volts +/- 10%. If the readings are off, check the values of the resistors in the input VCF section. The voltages here were 4.4 volts and 4.55 volts, a difference of 0.15 volts, well within the required window.
+6 volt regulator check
Check for +6 volts +/- 7% at the junction of R23 and R24. (This can be measured between the 0 volt power supply connection and pin 5 of U5). Check these resistors and the polarity of C6 (10uF) if it is off.
The voltage on this unit was 5.83V, again within the specified tolerance.
Negative volt regulator check
Check that the outside lead of the 79L05 is -5V +/- 10%. If not, check whether Trimmer B is fully clock-wise, as per the Analog Delay Testing section above. Also check the polarity of the associated electrolytic capacitors and the regulator itself. As the voltage specified here is pre-adjustment, it is of little consequence. The tested unit was set to -6.64 volts. Be aware that this is the trimmer that can blow up CA3080 according to what is written above.
Test of the dual S&H sections and the output low pass filters.
1) Take a 2000Hz sawtooth (0 to +5) straight from a VCO (e.g. the blue jack of a PCO) and feed it into pin 3 of U16 (CA3080). Use a 1k series resistor between the VCO and the chip to protect it. Check for a stepped replica of the saw at pin 1 of U4. (The bottom of the sawtooth wave will be flattened out.)
2) Now connect the scope to pad A and check for a rounded version of the saw. You may have to adjust Trimmer C to center the trace (around 0V). Repeat step 2 of the "Input VCF Test" above, but check for a progressive attenuation and filtering of the sawtooth wave at A.
3) Repeat the tests for the second sample and hold and low pass filter, feeding the saw into pin 3 of U17 (CA3080) and reading pin 7 of U4 for a stepped out, and pad B for a low pass output.
If a S&H doesn't work, check for the presence of the short 1.5 microsecond pulses at pins 10 and 4 of U11, and in much reduced form, near -12 volts, at pin 5 of each of U16 and U17. Check parts in the S&H area if these pulses are missing or incorrect.
If the low pass filters don't work, check construction and component values in the low pass VCF section.
Turn off the power and install the SAD1024 delay chips.
Feed a sine wave of about 200Hz (via a mixer) to input pad L. Monitor pin 12 of U13 and adjust Trimmer A until the (clipped?) sine wave is well centered (around 0V). Using the mixer, drop the amplitude of the sine wave until it is just barely clipping, and readjust Trimmer A for most symmetry.
Now monitor pin 7 of U4. Adjust Trimmer B for the most symmetrical sine wave. As above, you may need to adjust the amplitude of the incoming signal to be able to successfully adjust this trimmer.
If there is no signal at pin 7 of U4, monitor the outputs of the SAD1024 chips, one stage at a time, starting with pins 12 and 5 of U13, and pins 12 and 5 of U14. Also check the resistors in the analog delay section.
Final musical tests should be done with a short transient sound at a 2 per second repeat rate, with a 100Hz saw into a VCA. Feed CVs into the proper inputs, check ranges and proper filtering action, check the flanging output (pad E). If this doesn't work refer to the S&H, LP filter and output flanging sections.
Parts list
This is a guide only. Parts needed will vary with individual constructor's needs.
| Part | Quantity |
| Capacitors | |
|---|---|
| 24pF | 3 |
| 47pF | 2 |
| 82pF | 3 |
| 100pF | 1 |
| 220pF | 5 |
| 1000pF | 2 |
| 10nF (=0.01uF) | 2 |
| 22nF (=0.022uF) | 1 |
| 220nF (=0.22uF) | 1 |
| 1000nF (=1uF) electrolytic | 5 |
| 10uF electrolytic | 7 |
| Resistors | |
| 100Ω 1% | 2 |
| 220Ω 1% | 4 |
| 330Ω 1% | 1 |
| 470Ω 1% | 1 |
| 820Ω 1% | 1 |
| 1k 1% | 2 |
| 2.2k 1% | 2 |
| 3.3k 1% | 2 |
| 3.9k 1% | 2 |
| 4.32k 1% | 2 |
| 4.7k 1% | 4 |
| 5.1k 1% | 1 |
| 6.04k 1% | 1 |
| 10k 1% | 8 |
| 15k 1% | 6 |
| 22k 1% | 1 |
| 33k 1% | 7 |
| 39k 1% | 2 |
| 47k 1% | 12 |
| 56k 1% | 1 |
| 100k 1% | 2 |
| 820k |
1 |
| 182k 1% | 2 |
| 200k 1% | 1 |
| 220k 1% | 3 |
| 680k 1% | 1 |
| 820k 1% | 1 |
| 1.3M 1% | 1 |
| 22M 1% | 1 |
| 10k trimmer | 1 |
| 100k trimmer | 2 |
| Semi's | |
| 1N4004 | 1 |
| 1N4148 | 4 |
| 2N3904 | 2 |
| 4050 | 1 |
| 4093 | 1 |
| 4558 | 3 |
| 5532 | 1 |
| 79L05 | 1 |
| CA3046 | 1 |
| CA3080 | 2 |
| CA3280 | 2 |
| LF353 | 4 |
| SAD1024 | 2 |
| Misc. | |
| Ferrite bead | 2 |
| MTA-156 header 4 pin | 1 |
CC-BY-NC
Readers are permitted to construct these circuits for their own personal use only. Ken Stone retains all rights to his work. Serge Tcherepnin retains all rights to his work.
See also
References
- Wilson Analog delay for music synthesizers. by Ken Stone, 2001, with permission of the author - archived
External links
- CGS Synth discussion group, for discussion of locating parts, modifications and corrections etc.
- CGS208 BOM at Bompiler
- Wilson Analog Delay, Serge Modular Music Systems