Orange Squeezer is killing me

[swever]

I've breadboarded it following both ggg's and tonepad's shemos and it didn't work. I spend some 4 hours trying to debug it and it didn't work. Then disassembled the whole thing and started from scratch using this schematic (which appears to be the same):

I used 1n34a diode.

I used 2n5457 that I bought off ebay. They worked fine in every other circuit I tried.

Here's what I got. It's very wrong, obviously. The ranges are min and max position of the bias pot.
Component         Location        Voltage
9V battery                                9.53
Q1                       Gate              0 to 0.66
                           Source           0.16 to 0.84
                           Drain             9.53
Q2                      Gate              -0.23 to -0.50 (with some wobble)
                           Source        0 to 0.66
                           Drain            0.66
IC1                     Pin 1              5.27
                           Pin 2              5.27
                           Pin 3              5.16
                           Pin 4              0
                           Pin 5              2.28
                           Pin 6              3.34
                           Pin 7              7.46
                           Pin 8              9.51

Sounded very bad - nasty overdrive with some sort of modulation. The bias pot did not make much difference other than intensify the modulation a bit.

Then I though I would try other JFETs. I put j202's there. Less modulation and the voltages at jfet pins did not respond to the bias pot at all. Still sound very wrong.

Then I thought I should try other jfets and pulled the 202s with the whole thing powered and plugged in and found out that there was NO difference at all whether the jfets are there or not.

Please help me before I go completely nuts 

[swever]

http://i.imgur.com/aAVtFSM.jpg

The diode is the other way around because I wanted to test if it was going to make any difference. It didn't.

[antonis]

Start with properly terminating the unsed half of 4558..

[swever]

That would be pin 5 to gound and pins 6&7 linked, right?

I think I did that the the first time and it did not have any effect but I have to try again I guess 

[EBK]

Can't fully make out all the connections to that 2nd transistor from your pic....

[swever]

http://imgur.com/a/M9PU1

It's messy. Painful to look at especially on this pictures 

I think the connections are correct though:
Pin1 (drain) connects to 82k, 47n, 2.2n
Pin2 (source) -  to q1 gate and bias pot max lug
Pin3 (gate) - to two 470k

[EBK]

What's connected to the other end of the 470k resistor at the top?

[swever]

Damn! 

Gotta make that connection and see if it works. Will report in a minute.

[swever]

I guess I missed that connection both times
Seems to work now! Thank you so much, sharp-eyed Sir!

[EBK]

I guess I missed that connection both times
Seems to work now! Thank you so much, sharp-eyed Sir!

Glad it's working.
Happy to help. 

[swever]

It does work but now I am not sure it works right. At lower bias pot setting it passes through a long attack phase then crushes the level down. Then is releases slowly. That makes a very prominent ducking effect that is not rhythmically related to what I play.  Reminds me the dynamics processing in cheap digital cameras/phones etc.

The higher the "bias", the less prominent is the ducking, but it starts to sound "choked" overall. At about 80-90% of the pot rotation in quickly dies out completely - to silence. Is that how it's supposed to work?

Also, it's way more sensitive to lower frequencies/strings/notes than high.

It sounds like some random volume variations triggered by the loudest/lowest notes instead of the tidy dynamics leveling that I heard in demos. Maybe it does not like my active pickup. Maybe it's just the way I hear. I have never had any experience with stompbox guitar compressors before (although I have plenty of experience with software mixing and mastering compressors - maybe I'm spoiled by that).

[EBK]

Here is how it should work:
At one end of the bias sweep, you should hear silence.  Starting there, slowly turn it until the sound comes through.  Should be kind of transparent sounding at that point.  Turn it further and you will start to notice the compression eventually (also picks a bit of distortion as you keep going).

What are voltages at now?  We should probably check that first.  Did you remember to flip the diode back the right way?

[EBK]

I have never had any experience with stompbox guitar compressors before (although I have plenty of experience with software mixing and mastering compressors - maybe I'm spoiled by that).

That could be a significant factor here.  The OS is not a very sophisticated compressor.  People tend to like it because it is small, cheap, and has a bit of dirt, but you have very little no control over how it actually compresses. I had one for a while, but eventually recycled the parts and went on the hunt for something more adjustable. 

[swever]

That desription of how bias works sounds rather close to what I hear. The only thing is that its not really transparent neither at the "past silence" area nor further. There is some sort of distortion. I think it sounds like crossover distortion - it's most noticable at note releases/fades. There is also some clipping. I blame the opamp chip but I can be wrong. Tried a few different ones and they seem to make little or no difference.

I will measure the voltages as soon as I find a minute to spend in my man cave

And yea I did flip the diode. And tried a few different ones, including some very sexy big black russian germaniums. Not that I hear any difference in sound, but they defenitely looks "mojo" 

What did you end up building, Erik?

I want to build a compressor for bass - a "order" from an old pal of mine. Any recommendations?

[EBK]

What did you end up building, Erik?

I want to build a compressor for bass - a "order" from an old pal of mine. Any recommendations?

I feel like a walking billboard for this circuit (Wayne Kirkwood's One Knob Squeezer) because I've mentioned it so many times, but it really is my favorite pedal I've ever built, and the only one I could happily remove the stomp switch from:
http://www.diystompboxes.com/smfforum/index.php?topic=53097.msg946007#msg946007
I have a Vero layout for it if you have any interest (it uses one surface mount IC...).  It's very clean, and you can't dial in any "squishiness," so it won't be to everyone's liking.

If you want another suggestion, and on balance, I'd probably recommend building this one, The Engineer's Thumb is quite popular here and has many control options:
http://www.diystompboxes.com/smfforum/index.php?topic=97123.msg845753#msg845753

You could also take a look at the Ross/Dynacomp compressors. (sort of a throwaway suggestion...)

But, let's get this Orange Squeezer working first (hopefully, you agree).

[swever]

An update. I decided to build an orange squeezer on a pcb. I used this layout. After a few hours of agonal debugging yesterday late at night I was about to quit and go to bed. Switched off the iron and light. Gave the board a final glimpse and noticed there was a solder bridge.  It worked once I had it fixed.

Didn't have time to play with it for long enough yet, but I can say that it sounds nothing like the stuff I had on breadboard. Now it's very efficient and gentle at the same time - predictable and sensible dynamics processing. It has a MUCH faster attack now and does not let the initial transients through, and does that quite gracefully. No terrible ducking either now.

It must be that there was either a bad connection somewhere or another mistake when I had it on breadboard.


Anybody have any experience with the bass mod? What kind of change can I expect? I read somewhere that some people prefer it stock for bass. And I haven't got a bass guitar for testing.

[antonis]

Anybody have any experience with the bass mod? What kind of change can I expect?

I haven't the slightest experience but I should start by oversizing caps in series with signal rute and undersizing shunt caps..

[swever]

I mean the "bass mod" that's mentioned on effectslayouts and tonepad:

• Swap the output cap from 4.7µ to 1µ (between the 390k and 1.5k resistors, below the IC)
• Swap bleeder cap from 4.7µ to 10µ (cap connected to the 1N34 diode)
• Reduce the feedback resistor from 220k to 200k (between Q2 and the IC)

I find it weird that the output cap becomes a smaller value btw...

[Passaloutre]

I'm having similar problems with an Orange Squeezer on my breadboard. The "ducking" description seems to fit mine, it only responds to the loudest notes, takes half a second to react to them, and then the sustain "steps" down to a lower volume after half a second. I've played other Orange Squeezers that were properly tuned, and they sounded much better (smoother response) than this. I had some luck smoothing the response by replacing the 100k resistor to ground after the diode with a 500k (I guess this drains the 4u7 cap slower?)

[swever]

Apparently OS is not the most breadboard friendly circuit.
I highly recommend to try if it turns out better on a PCB.