Supa sustain build - is this what it's supposed to do?

[saxtim]

I've just built a supa sustain from GGG and I'm a bit confused as to what it's supposed to do.  I've read a few threads in the archives, and the general consensus is that it acts more as a fuzz pedal than a compressor.

schematic:

http://www.generalguitargadgets.com/v2/diagrams/cssussc.gif

Well, I've got a couple of problems.  Firstly I tested it by running my keyboard through it and I got a signal through ok - turning the sustain pot simply boosts the signal and then distorts it the more you turn it.  Couldn't really here any compression going on.

Then I tried my guitar and I had nothing, zip.  I couldn't figure out the problem untill I maxed the sustain control and I got a bit of signal.  I think the output of my keyboard is much, much hotter than my guitar, so I guess that's the reason.  So obviously I must have a problem in the circuit.

I used two dual gang 100k log pots as I had them lying around - i just wired up the first set of terminals on each (so they're like single gang) this shouldn't matter no?

I used a high brightness LED - it comes on, however nothing changes how bright it is.  Rotation of the volume pot or sustain pot don't effect it, nor does incoming signal.  I'm sure something should change it, or else you'd just have a normal resistor instead of the LED/LDR combo.

Maybe I should've used a RED led instead of the high brightness one?  Could that be the problem?

Tim

[RickL]

Once you get it working it will work as a regular compressor. I built one a year or so ago and after some fiddling I got it to work. IIRC I had to go through several different combinations of LED/LDR before I got a response that worked for me.

If you leave the LDR uncovered, point a light at it and move your hand between the light and the LDR does the output volume change? If not, your problem is in that section of the circuit. It sounds like the envelope detector part is working correctly. The LED lights up, then fades as the guitar signal dies down, correct?

[saxtim]

It sounds like the envelope detector part is working correctly. The LED lights up, then fades as the guitar signal dies down, correct?

No - the LED is permantely on, it is not effected by signal (LED is on even when there is no signal).

Tim

[RickL]

There's at least part of your problem then. The LED should change brightness with the volume of the input signal. As it changes brightness it changes the resistance of the LDR which changes the gain of the circuit. I don't remember if it gets darker with increased volume or lighter but it should change. You might have to fiddle with the sensitivity pot to get it to react correctly depending on how hot the signal from your guitar is (humbucker vs single coil pickups or active pickups).

Did you try the suggestion about waving your hand in front of the LDR to change the amount of light reaching it? What were the results?

[saxtim]

Ok - moving my hand in front of the LDR makes some difference (only a little, but because it's daylight, I'd expect much more with full darkness to full light.)

I did some transistor readings:

Q1 (labelling 1-4 from left to right in the schematic)

E .015 B .5 C 2.35

Q2

E 1.73 B2.35 C 7.92

Q3

E 1.95 B 1.61 C 9.30

Q4

E 1.32 B 1.97 C 9.30

Only Q3 looks unhealthy.  It's collector is attached to D1.  

I've checked the 220k resistor on Q3's base, and the 47nf cap.  Both are correct value and have continuity to Q3.  The 100ohm resistor is correct at Q4's emitter.

I don't really know what to make of this now - could it be the LED?  Maybe it's not working because it's a high brightness? On a side note, the LED has got 9V across it, I would've though that was very high?

Any more suggestions, maybe sub a regular old red LED in?

Thanks

tim

[RickL]

I'm 'bout running out of suggestions. The envelope detection part of the circuit seems to be the part that isn't functioning. Have you tried using an audio probe to see where the signal gets lost? I think you should be getting a fairly hot signal at the base of Q3.

If I had to guess, I would expect the LED to be brightest with the lowest signal, lowering the resistance of the LDR and providing the highest gain. I wonder if the first section of the circuit isn't amplifying the signal enough to dim the LED even at it's highest amplitude. That might result in both low output volume and no envelope detection.

Since you have a keyboard, try sending it a sustained sound and varying the keyboard volume from zero to max. See if this has any effect on the LED. If it only dims the LED at very high volume it might indicate that the amplification of the first part of the circuit isn't enough. If so, do the usual checks for transistor polarity, cold solder joints, parts values, etc.

It would be nice if Mark Hammer would chime in here. He seems to be the resident expert on envelope detectors, usually in reference to envelope controlled filters.

[saxtim]

RickL

Different volume in keyboard makes no difference to the LED - it just stays on permantely bright.


I traced with an audio probe and this is what Ive noted:

Goes through Q1 ok, clean

Comes out of Q2's collector distored - it's this signal that I get out at the jack.

I've got signal at the base of Q3 - when I cover the LDR (Ie make it dark) this signal increases in volume.  There's no signal at the emitter or collector of Q3 when the LDR is in it's normal state.  When i cover it there is a extremely faint signal at the collector of Q3 - only just above the noise floor though.

I'm not sure if this helps anymore.

RickL - does your functioning unit remain clean throughout the rotation of the sensitivity pot, or does it distort.  I know mines not working correctly, but in the Archives several had commented that the original unit distorted the signal.  I figure if it's supposed to distort, then the area around Q2 is probably ok and it's only the envelope section as you commented.

thanks
tim

[RickL]

I don't remember any distortion once I got it working the way I wanted. There might have been distortion with the sensitivity pot at maximum but it's been a while since I've played with it (at least a year) so I could be mistaken. It's buried in one of 10 or 12 boxes full of pedals right now so it might take me a while to find it and report back.

I'll be out of town Monday and Tuesday so don't expect a reply until Wednesday at the earliest.

[saxtim]

RickL,

I've got the envelope part of the circuit working now - spotted a mistake.  The LED now lights at an incoming signal and fades as the signal does.

It's still acting as an all out fuzz pedal though - be interested to know when you get a chance to check if your's is the same.  Also be interested to know what transistors you used, though I don't think that should make a difference.

It doesn't actually sound too bad, just as I said it's basically a fuzz box that adds a little sustain, not the compressor I was thinking it would be.  

thanks
tim

[jsleep]

Transistor circuits can be a real pain in the neck.  Most likely your transistors aren't quite biased, so you get the distortion.  Very common for this sort of circuit (in fact it sorta almost looks like a fuzz face).  Hopefully you are using BC169 or some very low hFE transistors.  I'm not exactly sure the what the best way to bias those first two stages.  If anyone has any suggestions, jump in.

JD