Troubleshooting a double overdrive built

[jackwithoneye]

Hi guys,
i have an issue with my last build, i can't find the cause, maybe you'll have an idea, anyway thank you for reading this!

I built a double overdrive, A side is an Zendrive-like clone, B side a Timmy-like.
I've added dipswitches to chose which one comes first and second in the signal chain

The double Overdrive works pretty damn fine when B side (timmy) is first, and A side (zendrive) is second.

My problem is when A is first and B is second, both activated, and A is in very high gain and voice, B Generates an around 3Khz Whistle when there is no signal (as soon as there is some guitar sound, the whistle disappear until the signal reaches the noise floor, and comes back).


Ic voltages are fine. 


A side


B side



power section :



if you have an idea, or a track...

cheers

[Groovenut]

You'll probably need to post the layout as with any order switching scheme, this is an issue (even lower gain circuits)

The main issue with order switching is that the outputs and inputs are often in very close proximity with each other at the mechanical switch.

[iainpunk]

does the whistle persist if you put a buffered pedal in front? like a boss pedal in bypass mode, or a buffered tuner.

cheers

[jackwithoneye]

does the whistle persist if you put a buffered pedal in front? like a boss pedal in bypass mode, or a buffered tuner.

cheers

It is worst when nothing is plugged in the input.
Actually, it's a better with a buffered pedal before, but it's still there.

the whistle appears on high gain settings, with A and B turned on (signal path A=>B, not present when B=>A), with high gain on A  (with gain&voice are high) or/and B (gain&bass &/or trable are high)

After looking at this on a scope the frequency vary from around 1KHz to 3KHz depending of the gain setting.

on A (first in the chain)
More gain=> Lower Frequency
More Level => Higher Frequency
More Tone => Higher Frequency
More Voice => bigger amplitude

on B
More Treble => Higher Freq




thank you for your help guyz

[jackwithoneye]

You'll probably need to post the layout as with any order switching scheme, this is an issue (even lower gain circuits)

The main issue with order switching is that the outputs and inputs are often in very close proximity with each other at the mechanical switch.

im using 2 Dipswitchs 3 poles to switch from A=>B to B=>A

A=>B
SW1 OFF OFF OFF
SW2 ON ON ON

B=>A
SW1 ON ON ON
SW2 OFF OFF OFF

They're routing the guitar input signal to the corresponding footswitch (each footswitch is connected to its own IN&OUT circuit A or B, and manage the true bypass), and reinject

[composition4]

Try buffering your bias voltage with a opamp. Also, C11, C3 and C4 can go to GND instead of VB (not important to your problem but GND will be at a lower impedance than VB)

Jonathan

[duck_arse]

I'd say you want two separate power sections, so what happens in the timmy supply doesn't get into the zen section. you certainly want separate Vbias sections.

[iainpunk]

I'd say you want two separate power sections, so what happens in the timmy supply doesn't get into the zen section. you certainly want separate Vbias sections.

great advice, but i'd also try a buffered Vbias, which might stabilize the bias voltage.

cheers

[jackwithoneye]

thank you for your advices and ideas guys.

If the whistle would come from a Vb stability problem, wouldn't i be able to track Vb Varations with my oscilloscope?

As far as i'm testing out with my oscilloscope, the Vbias is looking pretty stable (beetween 4.1 and 4.3V, depending if both sides of the dual overdrive are activated or not, and gains value) and has not more signal noise or parasit frequency oscillation when i've the whistle problem versus when the effect is running fine at lower gain.

From what i'm experiencing, it's whistleling when the first effect in chain has a (very) high gain  and attacks the second overdrive with a (very) high level

Your recommendations  are seperate power sections or buffered Vbias, or both?
And in the case of the buffered Vbias, do i have to double it to have real single Vbias per circuit?

What kind of Vbias buffering do you think is the best? LM386 usage?

Thanks!

[iainpunk]

thank you for your advices and ideas guys.

If the whistle would come from a Vb stability problem, wouldn't i be able to track Vb Varations with my oscilloscope? might be, but not always

As far as i'm testing out with my oscilloscope, the Vbias is looking pretty stable (beetween 4.1 and 4.3V, depending if both sides of the dual overdrive are activated or not, and gains value) and has not more signal noise or parasit frequency oscillation when i've the whistle problem versus when the effect is running fine at lower gain.

From what i'm experiencing, it's whistleling when the first effect in chain has a (very) high gain  and attacks the second overdrive with a (very) high level sounds plausible

Your recommendations  are seperate power sections or buffered Vbias, or both? i'd first try the separate power supply, then try the buffered Vb
And in the case of the buffered Vbias, do i have to double it to have real single Vbias per circuit? you can use 2 buffers both originating from your single Vb voltage divider

What kind of Vbias buffering do you think is the best? LM386 usage? a TL072 or similar dual opamp, so you can do 2 Vb outputs

Thanks!

hope this helps

[jackwithoneye]

thank you for your advices and ideas guys.

If the whistle would come from a Vb stability problem, wouldn't i be able to track Vb Varations with my oscilloscope? might be, but not always

As far as i'm testing out with my oscilloscope, the Vbias is looking pretty stable (beetween 4.1 and 4.3V, depending if both sides of the dual overdrive are activated or not, and gains value) and has not more signal noise or parasit frequency oscillation when i've the whistle problem versus when the effect is running fine at lower gain.

From what i'm experiencing, it's whistleling when the first effect in chain has a (very) high gain  and attacks the second overdrive with a (very) high level sounds plausible

Your recommendations  are seperate power sections or buffered Vbias, or both? i'd first try the separate power supply, then try the buffered Vb
And in the case of the buffered Vbias, do i have to double it to have real single Vbias per circuit? you can use 2 buffers both originating from your single Vb voltage divider

What kind of Vbias buffering do you think is the best? LM386 usage? a TL072 or similar dual opamp, so you can do 2 Vb outputs

Thanks!

hope this helps

Ok Thank you very much i'll to give it a shot with seperate power supply first

[antonis]

Before power supply seperation, just try to connect all Vbias gain branch points (like C11 in A side and C3 & C4 in B side) to GND instead of Vbias..

[jackwithoneye]

i tried separate power branchs (same 9 V but different voltage divider Vbias for each side), and it's still whistling at very high gain/level.

Before power supply seperation, just try to connect all Vbias gain branch points (like C11 in A side and C3 & C4 in B side) to GND instead of Vbias..

I'll try that, i have to re-breadboard everything from scratch to do that,

I dont really understand how it works and why grounding these caps should help, because i'm dumbly using the Zendrive principles, which are using the Vbias at these points. Could you enlight me about the fundamentals of these changes? I want to learn as much as i can.

[antonis]

I dont really understand how it works and why grounding these caps should help, because i'm dumbly using the Zendrive principles, which are using the Vbias at these points. Could you enlight me about the fundamentals of these changes? I want to learn as much as i can.

These caps are already AC grounded (via Vbias filter cap) but they might pass "garbage" into Vbias..
By DC grounding them, you let Vbias be as clean as possible..

P.S.
I think duck_arse suggested you to implement completely seperated power supplies..

[jackwithoneye]

I dont really understand how it works and why grounding these caps should help, because i'm dumbly using the Zendrive principles, which are using the Vbias at these points. Could you enlight me about the fundamentals of these changes? I want to learn as much as i can.

These caps are already AC grounded (via Vbias filter cap) but they might pass "garbage" into Vbias..
By DC grounding them, you let Vbias be as clean as possible..

I think duck_arse suggested you to implement completely seperated power supplies..

thanks for the explaination, i understand, it makes sense. I'll give it a shot and tell you if it improves the whistling noise.

P.S.
I think duck_arse suggested you to implement completely seperated power supplies..

i actually did that, i think..., unless you meant that i need to use 2 seperate external 9 Volts isolated power supplies. I actually used a single 9V coming from a single Ciocks power supply to feed the 2 seperated power circuits, one circuit per channel of the pedal. ( not common Vbias)

[jackwithoneye]

I dont really understand how it works and why grounding these caps should help, because i'm dumbly using the Zendrive principles, which are using the Vbias at these points. Could you enlight me about the fundamentals of these changes? I want to learn as much as i can.

These caps are already AC grounded (via Vbias filter cap) but they might pass "garbage" into Vbias..
By DC grounding them, you let Vbias be as clean as possible..

P.S.
I think duck_arse suggested you to implement completely seperated power supplies..

The thing that i don't understand by doing this, is how the removal of the 4.5V Vbias voltage from the inverted inputs side of the opamps will not affect the reponse on these opamps?

[iainpunk]

I dont really understand how it works and why grounding these caps should help, because i'm dumbly using the Zendrive principles, which are using the Vbias at these points. Could you enlight me about the fundamentals of these changes? I want to learn as much as i can.

These caps are already AC grounded (via Vbias filter cap) but they might pass "garbage" into Vbias..
By DC grounding them, you let Vbias be as clean as possible..

P.S.
I think duck_arse suggested you to implement completely seperated power supplies..

The thing that i don't understand by doing this, is how the removal of the 4.5V Vbias voltage from the inverted inputs side of the opamps will not affect the reponse on these opamps?

there are really big capacitors from ground to 4.5v
those capacitors are a short circuit for AC signals but a extremely high resistance for DC
your signal is either ''effectively shunted to ground'' or ''actually shunted to ground''
resulting in almost the same response.
the only difference in terms of signal is the chance of a small bit of ''garbage'' that could be on your 4.5V node, ground doesn't really have that ''garbage''

cheers

[duck_arse]

I dont really understand how it works and why grounding these caps should help, because i'm dumbly using the Zendrive principles, which are using the Vbias at these points. Could you enlight me about the fundamentals of these changes? I want to learn as much as i can.

These caps are already AC grounded (via Vbias filter cap) but they might pass "garbage" into Vbias..
By DC grounding them, you let Vbias be as clean as possible..

P.S.
I think duck_arse suggested you to implement completely seperated power supplies..

nuh-uh. what I was aiming at was a single DC adaptor feeding 9V into the case, then a pair of R//C supply filters, one going left, one going right, feeding an effect each. then you make your pair of bias sections off those isolated supplies.

[ElectricDruid]

The thing that i don't understand by doing this, is how the removal of the 4.5V Vbias voltage from the inverted inputs side of the opamps will not affect the reponse on these opamps?

I never got this either until I realised that DC can't pass through a capacitor, so it doesn't really matter from a "DC bias" point of view what voltage the bottom of a cap connects to. You're not removing the 4.5V bias by moving that connection, because *it was never getting the 4.5V bias from there anyway*. If you look closely, there'll be another source of bias for the op-amp.

[jackwithoneye]

thank you for your answers guys, it helps a lot, i try to improve my knowloedge on each build.
I'll do a feedback when i'll have tested out all of this.
I hope that this whistle is coming from this Vbias reinjecting noise, I've tested with seperated Vbias per channel but it's still there, maybe when those cap will be directly grounded, it'll solved the thing, and it will be a big thanks to you.

The thing that i don't understand by doing this, is how the removal of the 4.5V Vbias voltage from the inverted inputs side of the opamps will not affect the reponse on these opamps?

I never got this either until I realised that DC can't pass through a capacitor, so it doesn't really matter from a "DC bias" point of view what voltage the bottom of a cap connects to. You're not removing the 4.5V bias by moving that connection, because *it was never getting the 4.5V bias from there anyway*. If you look closely, there'll be another source of bias for the op-amp.

damn, you're right, i feel so dumb to have missed that
Thank you for your message, i'm on another build using your wonderful TAPLFO 3C. I may have questions on it in the future, but i keep it for later