Non-true bypass switch popping

Started by MikeH, September 16, 2008, 06:16:57 PM

Previous topic - Next topic

MikeH

So I've built a pedal that uses this switching style:



And it's popping like a mofo.  I'm using a DPDT and using the other pole for the LED.  The switch I'm using is this one:

http://www.smallbearelec.com/Detail.bok?no=23

The bypass is coming from a buffer at the front of the circuit.  I have the 2 68K and the 100K resistors mounted directly on the switch.  All electros are properly oriented.

What am I doing wrong?  Could the switch just be junk?  It has that *pioing!* noise when you switch it.  I haven't checked for DC on the output yet... I'll do that.  It seems this type of switching shouldn't pop at all.
"Sounds like a Fab Metal to me." -DougH

Gus

First disconnect the led to find out if it a step on the power supply line caused by the led current.  Did you try another battery sometimes cheap 9VDC batteries have not so good internal connections to the cells so the internal impedance is not so good.  Did you use a cap across the supply?

What is the circuit above the bypass part of the linked circuit fragment?

Did you try another switch?


MikeH

Haven't tried a different switch, that was my next move.

Quote from: Gus on September 16, 2008, 06:33:59 PM
What is the circuit above the bypass part of the linked circuit fragment?

Do you mean what the "bypass" is connected to?  There's an electrolyte in the path, as well as a resistor, but it just connects to a unity gain buffer- tl072.
"Sounds like a Fab Metal to me." -DougH

MikeH

Oh, and I'm not using a battery, and there is a cap across the supply- I'll try disconnecting the LED too.
"Sounds like a Fab Metal to me." -DougH

grapefruit

Not sure if this will fix your problem, but I think those 68k resistors should be going from switch side of the op amp output caps to ground. You don't really need the one on the effect output because the pot will pull the dc level to ground.


Stew.

kurtlives

My DIY site:
www.pdfelectronics.com

MikeH

"Sounds like a Fab Metal to me." -DougH

ayayay!

The people who work for a living are now outnumbered by those who vote for a living.

MikeH

I suppose another question would be "Can anyone tell me the purpose of those 68k resistors?"
"Sounds like a Fab Metal to me." -DougH

earthtonesaudio

Quote from: MikeH on September 17, 2008, 10:44:02 AM
I suppose another question would be "Can anyone tell me the purpose of those 68k resistors?"

They look wrong to me.  The way it's drawn, the "bypass" and "effect" are both going to the output, no matter the position of the switch.  Flicking it one way or the other just makes one side louder than the other.

MikeH

I left out that the main guts of the circuit (in between the 2 main gain stages) get grounded when in bypass mode, here:



The other side connects the LED.
"Sounds like a Fab Metal to me." -DougH

grapefruit

Yes, but you need a dc path (resistor) to ground on the OUTPUT SWITCH SIDE of C10. The NC, NO and COM terminal of the switch all need to be at the same DC level to avoid thumping. You can check with a multimeter if there is a voltage difference with the switch in either position, but if you have resistors there it won't be a problem.

As I initially said, and earthtones has said, the 68k resistors look wrong. I don't know the circuit but I suspect they should be going to ground, not the output.

Cheers,
Stew.

MikeH

Thanks everyone.  The 68k resistors aren't connected to ground, but they are connected to the 100k resistor which is connected to ground.  That should be allowing them to drain right?
"Sounds like a Fab Metal to me." -DougH

MikeH

Apparently the problem is... I'm a dumbass.   ;D
"Sounds like a Fab Metal to me." -DougH

grapefruit

It's a bit hard to tell exactly what's going on without the whole circuit.
Now looking at the second fragment you posted I can see how the output switching would kind of work. When in bypass the effect signal is shunted to ground (you say between the two main gain stages) so it is not mixed with the dry signal, but when the effect is on a bit of the dry signal is mixed in with the effect signal.

This still looks wrong. I can see a few problems.

1. You say "the main guts of the circuit (in between the 2 main gain stages) get grounded when in bypass mode". well you'll have all the noise of the second gain stage on the output when in bypass mode.
2. The output impedance is higher than it needs/should be. (maybe on purpose?)
3. If you want to mix some dry signal with the effect why not do it pre bypass switch.

Like I said, it's hard to tell without the whole circuit, or knowing where it comes from but I still think it is either drawn incorrectly or just a bad design.


Stew.