Orange Sqeezer (OSQ) Problems

[Fleetdog]

I just populated the circuit board for a General Guitar Gadgets Orange Squeezer (OSQ) kit.  I made no modifications or substitutions.  I have not yet done all the off board wiring so I am using alligator clips to make connections to my prototyping board.  When I play through it, I can't tell any difference between the dry signal and the effect's output except for the extra noise coming from the gator clip wires.   The trim pot sends the whole thing into oscillation if I get it more than half way clock-wise.   Here are my voltage readings followed by what the chart on GGG says they should be.
       measured    suggested
V+       9.11            8.8
Q1
  G       2.46             2.4
  S       3.95             4.0
  D       9.11             8.8
Q2     
  G       0.04             0.6
  S       2.46             2.6
  D       1.08          2.1-2.6
IC
  1       8.49             8.3
  2       8.49             8.3
  3       0                  0
  4       0                  0
  5       4.1               4.7
  6       4.98             4.8
  7       4.95             4.8
  8       9.11             8.8

While inserting the diode, I did chip a bit of the glass off one end; however, I measured the diode with my multimeter (on it's diode setting) and it read conductivity only in one direction then a 250mV drop the other. I expected that if I damaged the diode, it would either not conduct in either direction or it would conduct in both.  Is it likely that I did actually damage it and it is causing the low voltage on Q2's drain?
     

[Fleetdog]

After doing a bit more research, I found that the datasheet for the 1N100 says the forward voltage drop is 1V.  While it seems like that could cause problems, looking at the schematic, I think any problems with the diode would present voltage problems to Q2's gate as well as it's drain.  Could this just be a faulty transistor in Q2?  I don't think I have more of that transistor laying around.

[PerroGrande]

Your Q2 gate is pulled down substantially lower than the recommended GGG voltages, too.  The 1N100 is a Germanium diode, so it will forward bias at around 0.25 volts, give or take. 

Check the soldering around Q1 and Q2 very carefully for bridges.  The unit should not be oscillating when you adjust the trim pot.  Oscillating is frequently the sign of a dodgy solder connection (frequently a bridge) that is spilling output back to input.

It is *possible* that the diode is damaged.  The purpose of the diode and capacitor (D1 and C7) form a sample-and-hold circuit that is discharged by R12.  This drives the gate of Q2 which is acting like a voltage-controlled variable resistor.  This controls the "volume" of the circuit (which is what produces the compression).  If you can measure the DC voltage on C7 while providing the circuit input, you should be able to tell if your diode is doing anything.  If possible, put the positive clip of your voltmeter on the cathode lead of D1 (which connects to the positive side of C7) and ground the negative lead.  Set the meter to DC voltage.  Provide the circuit some input with varying loudness levels (i.e. plug in your guitar and strum loudy, pluck quietly, etc).  Observe the DC voltage on the meter (might change, so it could be hard to read.  Change is what you want to see, though -- and it should correspond with the loudness of your playing.  If this looks good, then your diode is okay.

[Fleetdog]

I checked the soldering and it all looked fine to me.  Definitely no bridges.  So I get it all hooked up to try and measure the voltage on C7 and suddenly the thing works fine.  I have no idea what was going on yesterday.  So I tried to bias it through an amp then headphones (napping wife) and I can definitely tell the effect is there though subtle.  I had to bias it significantly lower than the voltages on that chart, but I have noticed that hose headphones distort early so I'll try again when I can make some racket.  Thanks for the help Perro. 

[PerroGrande]

No Problem, Fleet --  let us know how it goes when you have the opportunity to test is further.

[Mark Hammer]

1) Because the stock OS does not aim for prolonged "sustain" (i.e., slow recovery), the action of a properly-functioning OS does tend to be more subtle than many expect.  If you have not worked with a compressor before, it can sometimes be hard to tell if its on. 

2) FETs always seem to just "know" what the most counter-intuitive pinout is.  Verify your pinouts for that FET for that manufacturer.

3) The trimpot needs to be set right.  You will still hear a clean sound even when the trimpot is not set right, so tweaking it to the sweet spot can sometimes be tricky.

[Fleetdog]

1) I *think* my OS is working, but as you stated it is very subtle.  Really, it sounds more to me like a clean boost with a very slight tone coloration.  I went to guitar center today and messed with a couple of their compressors to get a better idea of what to listen for in the OS.  I like the way it sounds right now, but I'd like to squeeze a bit more squishiness out of it (pardon the pun, I couldn't help myself).

2) I am using the fets and PCB that came from GGG as an OSQ kit.  While I haven't looked at the datasheet for the particular manufacturer (I'm not sure who it is from the tiny markings on them), I the pinout is correct for the manufaturer of that part number I did look at.

3)  I think this is going to be the key to it all.  One thing that bugs me here is that when I look at the build instructions, it mentions that all the way counter-clockwise there will be no sound (this is true for my unit) and that all the way clockwise will likely cause distortion (not true at all for me).  The only obvious result of turning up the bias in my box is that it increases the max volume, but the sound is clean as a whistle all the way to the top.  Even at that setting, the Drain on Q2 doesn't make it to 2V though.  I'm seriously considering ordering a replacement for Q2 just to see if that helps. 

I did try Perro's trick for checking the DC voltage across C7 and it appears the diode is functioning fine.  It was actually really cool to watch the voltage follow my average playing volume.  It would be cool to put a small analog voltmeter on the top of the pedal like a tachometer and have it back lit instead of an LED.  I'm not going to do that on mine, but it would be cool.

[Fleetdog]

So, I understand now that in the schematic http://www.generalguitargadgets.com/pdf/ggg_osq_sc.pdf?phpMyAdmin=4a28f86a515b7883e7bc35a68d4e7b6d R12, C7, and D1 form the envelope follower that sets the gain of the circuit.  I assume then that varying the resistance of R12 would change either the attack or decay time of the compressor (more likely both). Since the effect is already all soldered together, could anyone make an educated guess as to whether changing this value would likely help make my OS a bit more obvious when it is on?  I'm pretty new to envelope followers and compression, but I think it's starting to click for me.