Orange Squeezer: Momentary signal drop when engaged?

[Russ]

My diy Orange Squeezer (GGG board and wiring) has worked flawlessly for a couple of years, but recently has developed a problem:

when switched on, there is a slight momentary drop in volume (or signal) until the effect engages. It reminds me of the old Mk. IIB Boogie amplifiers- when the lead channel was engaged, there was a momentary drop in volume till the lead channel swelled in.

I also notice that this seems to generally occur after the effect has been off for about 15 or 20 seconds, that is, it doesn't seem to occur when I quickly switch the pedal on and off repeatedly (after the initial drop after the first switching).

I wonder if this could be a power supply cap losing charge after the effect has been off for several seconds.

The pedal is wired for true bypass with the effect input grounded when off. The 3PDT switch and LED seem to function perfectly.

Help please, any ideas?

Thanks, Russ

[Processaurus]

Interesting problem, here's a test that might help pinpoint where it is, when you're playing the OS, stop playing for the 15-20 seconds, and then play and see if it swells in like when you switch to it, that'll tell you if it is a problem with the compressor circuit when it's been silent for a while, or something to do with the switching.

[coelacanth]

Mine has a similar problem, but it only does it when using a 9v battery. Try it with a power supply to see if you still have the problem. It may not be too hard to troubleshoot.

[Russ]

Mine does it with a power supply (and I've tried several). Upon looking at the schematic, I notice that there are no filter supply caps.

Could it possibly be my switch? Can only one leg of a 3PDT become glitchy?

Thanks, Russ

[Russ]

Replaced the switch, and that's not it. All connections in the pedal are good, and the supply voltage.

I notice that the problem seems to occur a few minutes after the pedal has had power connected to it (like it needs to warm up to exhibit the momentary drop when engaged).

Could one of the electrolytic caps, or a transistor, or the IC chip be glitching?

Thanks, Russ

[Mark Hammer]

I yanked mine out last night and it does the very same.   Switching from bypass to OS takes about 3-5sec for sound to come back.  Hmmmmm.   Verrrrrry interesting.

Postscript:  Now that I look at it again, it would seem the question to ask is "Why is the FET being held to a low resistance until the pedal is turned on for a bit?".  Remember that the device works by reducing the drain-source resistance of that FET hanging off the 82k input resistor.  If the pedal starts up with the gate voltage such that the D-S resistance is at min, then you won't hear anything until the D-S resistance goes up a little bit at least.  Hmmm, now I'm wondering if this is one of the things one sets the trimpot to avoid. 

Experts, what say you?

[Stellan]

Mine too... and it didn´t use to

[Russ]

My momentary drop is only about 1/2 of a second before the circuit comes to life. I've adjusted on the trim pot throughout my testing, and in my case, it hasn't effected the problem at all.

Any ideas?

Thanks, Russ

[ayayay!]

Anyone else have a take on this?  I'm really surprised to see two different pedals do this seemingly randomly after a long period of time.  I was just about to build one so I'm curious...   

[Mark Hammer]

I wish I could remember if the phenomenon I get on mine was there from the get-go or was only more recent.  All I know is my reaction was "Where in the heck did THAT come from?".  I initially thought it was because I had done some smartass thing with recovery-time mods, but when I went back to stock, it still did the same thing.  I'll look into it further this weekend and report back.

[ayayay!]

That's very kind of you Mark.  Thank you! 

[Russ]

Thanks Mark, it's got me perplexed. I'd love to get this figured out.

Russ

[head_spaz]

Try adding a temporary jumper as shown below.
When you go into true bypass mode, the input signal to the compressor circuit is removed and the compressor re-adjusts to the zero input level.
Then when you switch it in, it takes a few seconds to recognize, resample and adjust to a "new signal" before it levels off.
This mod will keep the compressor tracking your inpiut signal at all times (even in bypass), so it's up and running, ready whenever you are.
This mod will also defeat the concept of "true bypass", but the input impedence on this circuit is so high that's unlikely to "tone suck" from the input source.
And then pick which poison sets best with you, because all things in life demand tradeoffs.

[Mark Hammer]

Brilliant!  Of course.  Why didn't I think of it?  The original would not have used TB, so the rectifier circuitry was aways kept busy and charged up.  This also means I can use a cheaper DPDT X-wing switch for the pedal and still get a status LED!   I do believe I owe you a big one.

[Russ]

I guess I still don't get what has caused this to occur in these pedals all of a sudden. Like I stated initially, mine worked flawlessly for several years, and I thought the concept of the input to the effect being grounded during true bypass mode was supposed to result in the circuit being charged and ready when switched on.

Russ

[ayayay!]

Brilliant!  Of course.  Why didn't I think of it?  The original would not have used TB, so the rectifier circuitry was aways kept busy and charged up.  This also means I can use a cheaper DPDT X-wing switch for the pedal and still get a status LED!   I do believe I owe you a big one.   

Yeah, but wouldn't there be a simple way to keep the rectifiers busy all the time, yet still maintain TB?  I guess not since then you'd have to flip THAT noise out of the way when you engage your effect.  ...Not that it seems to matter that much anyway.   Not having true bypass on this one probably won't hurt anything.  I was so close to mentioning this method too, but of course I didn't trust my gut.   Yeah my thought was just to use a typical SPDT, a la GCB-95.  That would work. 

I guess I still don't get what has caused this to occur in these pedals all of a sudden. Like I stated initially, mine worked flawlessly for several years, and I thought the concept of the input to the effect being grounded during true bypass mode was supposed to result in the circuit being charged and ready when switched on.

Russ 

+1