What would cause a lag switching from bypass to effect mode?

[Paul Marossy]

In an inverting opamp scenario? I have a circuit I'm messing around with that I like, but it has a little problem: when you switch it on, the signal level dips to like 50% of what it should be for about 1/2 second before it comes on full strength. This is weird, never encountered this before and I don't have any idea what could be causing it.

Schematic is here: http://www.diystompboxes.com/smfforum/index.php?topic=102162.0

The only thing I've done different is to add a 100uF cap to the 9V supply, a 22uF to the Vref supply and a 1M pulldown resistor on the input.

Any ideas?

[R.G.]

Any time there is a time delay on sound coming through a circuit, it's either a result of something having to heat up (as in tubes...  ) or a cap or inductor charging. The inductor charging scenario is there for completeness; in reality, it never happens for a variety of reasons. So you have some issue with a capacitor charging til the circuit is at normal operating voltages.

I personally would bet on the output cap. I would try using resistors of about 4.7K to both +9V and ground from the output of the opamp to force it to nearly Vbias with DC current. There are several reasons I'm suspicious of that, but the easiest way to find out is to hang the resistors on it.

[Paul Marossy]

I was thinking it had something to do with either the input or output. I tried adding a little bias to the input and that did nothing, but I didn't think to try it on the output. I'll give that a try and see what that does.

[amptramp]

Do you have a pulldown resistor on the output cap?  You may have leakage that misbiases the subsequent stage.

Also, that's a lot of stuff connected to the non-inverting input of the op amp.  If the capacitance to ground is high enough, the stage may act as a differentiator and oscillate at a frequency above the audio range.  Check the effect of changing the pot setting.  If it changes the delay time, you are almost certainly oscillating.  A small capacitor from output to non-inverting input will fix that.

[Paul Marossy]

Do you have a pulldown resistor on the output cap?  You may have leakage that misbiases the subsequent stage.

Also, that's a lot of stuff connected to the non-inverting input of the op amp.  If the capacitance to ground is high enough, the stage may act as a differentiator and oscillate at a frequency above the audio range.  Check the effect of changing the pot setting.  If it changes the delay time, you are almost certainly oscillating.  A small capacitor from output to non-inverting input will fix that.

I had a pulldown resistor on the output at one point, but it seemed to load down the circuit. It sounded better when I removed it.

As a matter of fact, I did notice that the lag did change depending on how the pot was set. But that was when I had it miswired after I first built it (forgot to make the connection between the transistor emitters and the input). Might still be behaving that way, I'll have to verify that. Adding a small cap is simple enough, it probably ought to have one on there anyway...

[duck_arse]

I just had another look at the magazine, and it shows 47uF on Vref, and the output cap is 1uF non-polar.

have you tried a balance pot between the 2 bases yet?

[Paul Marossy]

I just had another look at the magazine, and it shows 47uF on Vref, and the output cap is 1uF non-polar.

have you tried a balance pot between the 2 bases yet?

Aha! The output cap on mine is 4.7uF. Guess that schematic is wrong then. I like it as is though, I'd rather get it to work rather than change it. I guess I could try it both ways and see what I like better.

I haven't tried a balance pot between the two bases.

[Paul Marossy]

OK, so I verified that no matter what position the gain pot is in, there is still a lag. Plus I am using grounded circuit input true bypass, not sure how it could be oscillating. A small cap between inverting in and output does not stop it, it just cuts a little bit of the highs.

I also tried adding some bias to the output, that doesn't fix the problem either. It just lessens the amount of overdrive and seems to make the problem sightly worse.

All of these results are with a 1uF non-polarized output cap which the original schematic apparently indicates.

Still baffled...

EDIT: Maybe a better description is that for maybe 100mS the signal sounds normal, then it dips for maybe 750mS then comes back up. Not that noticeable when playing guitar but very noticeable when using a 1kHz test signal. Maybe it's just the nature of the beast?

[PRR]

Put voltmeter on opamp output pin. Should stay _AT_ 4.5V. Watch carefully as you switch in/out.

That plan does not show any in/out switching. Maybe you need to show us how you do that?

[Lurco]

You got a 4.7µF cap into a 100k input like in that eti schematic? Make that 47n, and it doesn`t need ages to charge up while still letting >34Hz through.

[duck_arse]

I was wondering about the original post: are you getting lag each time you bypass, or are you switching the power on/off? they are 2 different problems.

[Gus]

Have you tried the circuit with the 100Ks and transistors removed to hear if the time delay goes away?
Are you using a 741?


That circuit need an buffer before it IMO. Something like this or an opamp buffer or Jfet buffer etc.
You can have this bias the opamp.  
You can change the emitter voltage with adjusting R3
R6 is the 100k resistor to the - input.
I would also maybe add a resistor like a 100k or lower from Vref to the + input.  Look up input offsets in opamps.

[Paul Marossy]

I was wondering about the original post: are you getting lag each time you bypass, or are you switching the power on/off? they are 2 different problems.

Every time I bypass the effect. The power is always on.

[Paul Marossy]

Put voltmeter on opamp output pin. Should stay _AT_ 4.5V. Watch carefully as you switch in/out.

That plan does not show any in/out switching. Maybe you need to show us how you do that?

It's just a grounded circuit input true bypass, nothing special...

[R.G.]

It's just a grounded circuit input true bypass, nothing special...

Does that input grounding ground the input for **DC**, forcing the circuit to come back up to operating voltages on the input each time? Is the input cap possibly backwards, leaky, shorted, etc.?

[Paul Marossy]

It's just a grounded circuit input true bypass, nothing special...

Does that input grounding ground the input for **DC**, forcing the circuit to come back up to operating voltages on the input each time? Is the input cap possibly backwards, leaky, shorted, etc.?

I don't know for sure. It's just grounding the wire that goes to the input of the circuit. I tried adding a 1.5K bias resistor at the non-inverting input and it seems to have improved it slightly, but only very slightly. I checked the voltage on the input with a multimeter before I added that resistor and I did not see a dip in the voltage when switching the circuit on. Not sure if input cap is backwards, I put it in the way that schematic shows it, with neagative terminal towards the input, positive towards the opamp.

Have you tried the circuit with the 100Ks and transistors removed to hear if the time delay goes away?
Are you using a 741?

I bypassed the 100K resistors, they appear to have no affect on the lag. I'm using 1/2 of an OP275.

[R.G.]

Does that input grounding ground the input for **DC**, forcing the circuit to come back up to operating voltages on the input each time? Is the input cap possibly backwards, leaky, shorted, etc.?

I don't know for sure. It's just grounding the wire that goes to the input of the circuit.

Perhaps your DMM could watch the input and/or the output of the opamp while you switch and see if the DC levels there change when you do that.

[gjcamann]

What if you simplified the circuit and removed the transistor feedbacks, leaving just the 1M Ohm feedback in place. That may isolate the problem.

[R.G.]

What if you simplified the circuit and removed the transistor feedbacks, leaving just the 1M Ohm feedback in place. That may isolate the problem.

That's a good simplifying test, too.

[Gus]

What did you do with the other half of the dual opamp?  How did you wire it?

As myself and others posted did you remove the 100Ks AND transistors?