TL071 buffer strange behavior

[Silvio55]

I made this buffer and put inside my DS1 clone, between the input jack and the bypass switch, so is always in line, I also removed the DS1 original input buffer and added a new pull down resistor at the input of the DS1 circuit:


I used 2M2 resistors instead of the 1M for the bias.
What happens is this: every time I push the bypass switch I get a loud pop that wasn't before I added the buffer.
And the strange part is that sometimes when I bypass the DS1, being only the buffer in line, I get hiss, sometimes is quiet and this happens at random pushing the bypass switch!
I should try a pull down resistor at the output of the buffer, any suggestion for the value?
And what about the random hiss?

[Vivek]

Pull down resistor at output of Opamp, for C2 : Anything from 100K to 500K

You might need one for C1 too, maybe 2megs

There could be noise created by the 2 large resistors. There are ways to bias the Opamp that create less noise

Read about noiseless bias here :

http://www.geofex.com/Article_Folders/modmuamp/modmuamp.htm

Use 10k + 10K and 47uF for the divider, then 1Meg from the centre of the divider to the Non Inverting input of the Opamp.

[Silvio55]

Ok thanks! The thing is the hiss happens at random, sometimes it does it and sometimes doesn't.
What would be a better way to bias the IC?

[Silvio55]

UPDATE: The pull down resistor at the buffer output cured the hiss problem, zero noise every time now, but the pops are still there every time y hit the bypass switch, not as loud as before but very noticeable. I don't know what the problem could be, I have pull down resistors on the buffer and the DS1 (270k at buffer output, 2M2 at DS1 input and 100k output). It's not noise caused by the led either and it wasn't before I installed the buffer.

[iainpunk]

the pull down should be to 4.5v (Vbias) not ground,

cheers

[antonis]

Wire ON-BYPASS 3PDT switch like below:

[GibsonGM]

the pull down should be to 4.5v (Vbias) not ground,

cheers

I accept this, I have seen it enough to know it's true - but why?  At Vbias, you just 'gain' a cap to go thru to get to ground...AC ground.  Does it reduce resistor noise or something?   I mean - in a tube circuit, it would go to 'normal ground'...

[Vivek]

It ensures that both sides of C2 are at same DC voltage

Hence no DC potential on cap

[duck_arse]

the pull down should be to 4.5v (Vbias) not ground,

cheers

terminology, fellas. you don't pull down to bias, you pull down to ground. you don't pull the outer ends of caps to bias, it defeats their intended purpose.

[Silvio55]

Wire ON-BYPASS 3PDT switch like below:

I have the led wired like that but don't have the input and output grounded, I thought it was not necessary since the pulldown resistors. I'll give it a try!

[GibsonGM]

It ensures that both sides of C2 are at same DC voltage

Hence no DC potential on cap

Ha ha, I know why you want a pulldown resistor there, but not the 'dark art' of sometimes 'a thing' goes to Vref (aka, AC ground) and sometimes 'regular ground'.   Like, you often see stuff in the feedback loop of an opamp grounded at Vref, and others they use 'regular ground'. 

I have personally had circuits oscillate if grounded one way, and be fine if done another way. I just wonder if there are hard & fast rules.

[iainpunk]

the pull down should be to 4.5v (Vbias) not ground,

cheers

I accept this, I have seen it enough to know it's true - but why?  At Vbias, you just 'gain' a cap to go thru to get to ground...AC ground.  Does it reduce resistor noise or something?   I mean - in a tube circuit, it would go to 'normal ground'...

because its a boss style circuit.
jfet switches seem to work better with less popping noises when both sides are biased to the same voltage level.

cheers

[GibsonGM]

Ok, I can see that I'd never considered 'why', but now looking at a TS schematic, it makes sense.  Never thought to ask! Thanks.

[antonis]

I have the led wired like that but don't have the input and output grounded, I thought it was not necessary since the pulldown resistors.

Pull down resistors aren't necessary since IN-OUT grounded..

[Silvio55]

I have the led wired like that but don't have the input and output grounded, I thought it was not necessary since the pulldown resistors.

Pull down resistors aren't necessary since IN-OUT grounded..

And grounding the input and output would be a better way to reduce pops than pulldown resistors?

[antonis]

jfet switches seem to work better with less popping noises when both sides are biased to the same voltage level.

The issue with popping sound is the DC voltage level on left side of input cap..
(low level signal bumps into a high voltage wall so the idea of a "pull down" resistor (sorry Stephen  ) is to drain the residual charge out of the effect when bypassed..)

By routing that resistor to Vbias, you also inject DC on guitar pickups..

[antonis]

And grounding the input and output would be a better way to reduce pops than pulldown resistors?

IMHO, yes..

RxC time constant is reduced down to zero hence immediate input cap left plate discharge..

[iainpunk]

jfet switches seem to work better with less popping noises when both sides are biased to the same voltage level.

By routing that resistor to Vbias, you also inject DC on guitar pickups..

no, the pull down resistor is after the buffer, before the Jfet switching.

cheers

[antonis]

I'm talking about present buffer..