Ray Wilson authored this content while he was actively running MFOS as the founder and resident genius.
We retain the content because it reflects a valuable point of view representing that time and place.
| Introduction |
|---|
| I will be putting answers to emailed questions and requested help here. I imagine that eventually everything will be covered. The main thing to realize is that when wired according to the documentation I have provided on this site the circuit works perfectly. When you get it put together right and make sure you don't have any bad active components or misconnected or unconnected circuitry it will work. General trouble shooting info: Trouble Shooting 101 |
| Oscillator Calibration |
While the Sound Lab is primarily a sound generator and the oscillators are not stable or precision enough for many octaves of "in tune" 1V/octave range they can be calibrated so that you get an octave or two of "in tune" 1V/Octave range.
|
| VCF Control Range Change |
| The filter will hit the bottom of its range at various settings depending on what you are controlling it with. If you find that it makes the passband too low to pass anything of interest
when you still have several degrees of pot rotation no matter what you have connected to it then increase the value of R41 (resistor from pin 3 of the Cut-Off Frequency pot to -9V). That will raise
the bottom of the control range. Try something between 4.7K and 10K to start. It won't hurt to go higher than that in resistance if you want to try.
Signal bleed-through? A two pole filter will always allow a bit of the original signal to pass through but just a itsy bitsy teeny bit so if you have a lot of bleed-through at the bottom of the range look over all of the components associated with the filter to make sure everything is the correct value and connected properly.
|
| VCA Voltage Controlled Amplifier |
| If the the VCA is acting totally strangely make sure everything is wired correctly and that you have a good LM13700. Check the panel wiring to make sure that a ground or power connection
hasn't been missed.
The input to the VCA goes through R21 (33K) and is dropped across R22 (1K resistor to ground). Any patched in signals should go (via 100Ks) to the junction of IC3-B pin 14, R22 pin 2, and R21 pin 1. All this part of the circuit does is feed an attenuated signal to the non-inverting input of the LM13700 (the inverting input is grounded). Now on to the control voltage portion of the VCA. As you can see we are only applying current to pin 16 of IC3-B from the various control voltage inputs and we have biased the input toward the low end by R17, 100K to -9V. The VCA initial gain only applies a voltage of between -9V and +9V to pin 16 of IC3-B (via R18 470K). All of the control voltage inputs to the VCA are adding to or subtracting from the current arriving at pin 16 of IC3-B. So this is one of the simplest modules and it definitely works when everything is connected properly. Another thing to check is the average voltage at the junction of S6 pin 1 and R21 pin 2. This should be near ground and if the VCF has something connected improperly it could be applying a DC offset to this input thus causing the VCA circuit to misbehave so check that. All I can suggest is that you check all of the wiring, soldering and component values (including the components in the filter circuitry). Last but not least try another chip (LM13700) if something is amiss with this circuit module. R26, R42, and R43 must be tied to -9V or the bias of the circuit will be wacky. Also the biasing of the output buffer in the VCA is very critical make sure R24, R25, and R26 are all connected properly.
|
| Pulse Width Modulation Circuit |
| The output of IC6-D needs to be present for the pulse width modulator (IC2-A) to work. Check pin 14 of IC6. A sawtooth wave should be present and the frequency should vary when R58 is
adjusted. The same saw tooth signal should be present on pin 1 of R79 (the side not connected to pin 3 of IC2A). The sawtooth should also be present on pin 2 of IC2-A (it may look a bit attenuated
depending on the impedance of your scope probe). Make sure S15 is turned off for the following checks. When you adjust R85 the voltage present at pin 1 of R86 (the side not connected to pin 2 of
IC2-A) should vary in the same manner as the voltage seen at the wiper of R85. If it does not then check that the wiper of R85 is connected to the circuit board via a wire. If the voltage at the
wiper of R85 does not vary when it is adjusted then the resistive element terminals may not properly connected to both +9V and GND via R84 and R104 respectively. If it is connected then you may have
a bad pot. If the voltage is varying properly on R85 then check the output of IC2-A (IC2 pin 1). You should see a pulse wave whose frequency is the same as the sawtooth you saw on pin 1 of R79 and
whose pulse width varies in response to adjusting R85. If you have checked everything so far and still don't have a pulse wave on the output of IC2-A you may need to replace IC2. If you have the
pulse wave on pin 1 of IC2 then make sure you have it at pin 3 of S10 (it will be attenuated by a factor of about 5). If you have it there then make sure that the schematic point O2 is properly
connected to R38. You should see the same attenuated pulse wave on pin 1 of R38. The wiper of R38 should be connected to R33 on the PCB. IF you have the signal to here you are good to go as far as
the Pulse Width Modulated waveform is concerned.
|
| Attack Release Generator |
| As always, check the associated part values and orientations to ensure they are correct. Disconnect the external gate if you are using the external gate feature. With no external gate
connected, S3 (AR Mode switch) set to "Gated" and S1 not pushed you should see +8 to +9 volts (depending on your meter's impedance) between pin 1 of C2 and ground. In this state IC1-A pin
2 should be low. When you press S1, you see about 1.5 volts between C2 pin 1 and ground and IC1-A pin 2 should go high and return low when you release S1. You should see a positive 8V to 9V pulse
across R4 whenever you press S1. Set both Attack and Decay to the fully counter-clock-wise position (minimum attack and decay times). When you press S1 the voltage between C3 pin 1 and ground should
go to between 8 and 9 volts (IC2-B pin 7 should also go to 8 to 9 volts). When you release S1 the voltage between C3 pin 1 and ground should go to 0 volts (IC2-B pin 7 go to between -8 and -9
volts). You should see the same voltage on pin 1 of R15 as you see on IC2-B pin 7. When IC2-B pin 7 is greater than about +6 volts IC1-C pin 6 should be at 9 volts. Set S3 (AR Mode switch) to
Trig'd. Pressing S1 should cause IC1-E pin 10 to go to ground and then almost immediately return to +9V. Pressing S1 should cause IC1-F pin 12 to go to +9 volts and then almost immediately return to
ground. Turn R10 a bit clockwise. Pressing S1 should cause IC1-E pin 10 to go to ground and then return to +9V after a delay. Pressing S1 should cause IC1-F pin 12 to go to +9 volts and then return
to ground after a short delay. The further clockwise you turn R10 the longer the delay should become. The voltage envelope on IC2-B pin 7 should have an attack curve and a very very short decay
curve. The further clockwise you turn R10 the longer the attack curve should become. Advancing R11 clockwise should cause the decay curve to lengthen. Return to a very short attack and decay time
setting and set S2 (AR Repeat) to on. You should see a repeating attack decay envelope at IC2-B pin 7. Changing R10 or R11 should cause the envelope to change appropriately.
Some causes for problems include:
|
| White Noise Generator |
| As always, check the associated part values and orientations to ensure they are correct. Once you've done that make sure that you have the emitter and base of Q7 oriented correctly on the
PCB. You should see a positive voltage at the emitter of Q7 of between 4.5 to 6 volts (depending on the impedance of your meter or scope it could be lower but it should at least be a couple of volts
above ground). Check the output of IC7-C (pin 8) you should measure a voltage that is near ground (depending on the leakage of the IC). You should see white noise at a level of at least 100 mV or
more at pin 8 of IC7-C when you look at it with an oscilloscope. If you do, great, if you don't then hold one end of a short wire and touch the other end of it to IC7-C pin 10. The output of IC7-C
(pin 8) should oscillate at 60 hertz at or near the rails (+/-9V). If you don't see this oscillation or IC7-C (pin 8) is stuck high or low replace IC7 and try again. If you see the oscillation but
you don't see noise you probably have a quiet 2N3904 and you are going to have to dig through your box of general purpose NPNs and test them until you find a noisy one. YOU DO NOT NEED TO CUT
THE COLLECTORS OFF JUST TO TEST THEM. When you finally find the noisiest one cut its collector off and use it. The reason you do this is because the collector acts like an antenna and unless you
WANT radio signals in your white noise you probably want to reduce the size of this impromptu antenna to the shortest possible length. OK once you have noise at pin 8 of IC-7 the rest of the circuit
should work. You should see rail-to-rail digital looking noise at the output of IC7-D (pin 14). If you don't see that then make sure the associated components (R95, R99 and C17) are soldered to the
board and that you have ground on pin 2 of C17. If you have noise at pin 8 of IC7 and you don't have noise at pin 14 of IC7 try replacing IC7. The point marked NS should see 1/6 the level of noise
you see at the output of IC7-D. The most likely cause of no noise is the transistor not being noisy.
|
| Overall Signal Check |
|
1) Review the panel wiring thoroughly. Once you have verified that both the panel wiring is correct and all component values are correct try observing the following circuit points. A good place to start looking is at the Sound Lab Mini-Synth VCA (Voltage Controlled Amplifier) and VCF (Voltage Controlled Filter) schematic. You will notice several circuit points (circles with letters in them) throughout the page. O1 - Raw Oscillator 1 output. As you test O1 sweep the Oscillator 1 frequency knob to insure you are listening to (or observing the right point). Do the same for O2, additionally flick the Ramp/Rect switch up and down a bit to make sure the waveform changes properly between ramp and square. Point NS should be the raw noise output (very noisey). LP is the output of the lowpass filter. Sweep the cutoff frequency as you observe this point. Turn the resonance down all the way. As you advance it the sound should become more resonant (contain more harmonic overtones). The filter may oscillate when resonance is all the way up (that's OK). Observe the signal level at pin 9 of IC3. This should be the main output prior to the output pot (which is wired as an adjustable voltage divider). The signal at R27's wiper should go from 0V to the same signal seen at Pin 9 of IC3 as you advance it. LFO and AR should go from 0V to the same levels as observed at LFS (on low frequency oscillator schematic) and pin 7 of IC2 (on Attack Release envelope generator schematic) respectively as you advance their respective controls (R92 and R15).
|