Squealing RAT proco 2 strip board build

[tom1970]

Hi
I made a Proco RAT 2 pedal using this stripboard layout. https://1.bp.blogspot.com/-aglaFjEaC...Proco-Rat2.png
It is working but it starts squealing if you crank the gain beyond about 75% and is worse with the filter down low. It's doing it with all three of my guitars on various amps (both tube and tranny amps). It's a high pitched tone that changes as you shift to different pick ups or as you move the gain between 75-100%. This does seem to be a common issue with homespun RAT pedal builds (see this blog http://tagboardeffects.blogspot.com/...proco-rat.html ) but I found another post elsewhere where someone suggested putting a buffered pedal in front of it so tried doing that by putting my Ibanez tube screamer in front of the RAT and it worked a treat - squealing gone completely. However I really want the RAT pedal to work well as a standalone so I tried taking a the output buffer from the tube screamer and adding it to the beginning of my RAT. So right now it looks like the attached diagram. Problem is the squealing hasn't gone.
Any idea where I'm going wrong here? The tube screamer has both an input and output buffer but I don't have room to put all that on my RAT stripboard. Here is the tube screamer circuit...https://www.electrosmash.com/tube-screamer-analysis

btw I have added in the 1N cap to ground (C4) at input 3 on the op amp as it was missing on the stripboard layout post and this doesn't make any difference to the squeal.

Thanks in advance for any help/advice...
Cheers
Tom

https://www.electrosmash.com/tube-screamer-analysis

[Enzo]

Just idle thinking, but that emitter follower buffer alone does not invert the phase of the signal. When you use a whole TS I bet it does.

[tom1970]

Just idle thinking, but that emitter follower buffer alone does not invert the phase of the signal. When you use a whole TS I bet it does.

Thanks for the insight! Having the tubescreamer in front of the ratt works even if the tube screamer is in bypass. I don't know whether that would invert the phase (not really something I know much about). This guy on another forum suggested that the squealing might involve phasing... https://forum.pedalpcb.com/threads/s...555/post-55230

[Helmholtz]

The Rat circuit is close to instability and likes to self-oscillate around 3kHz.

In the early 80s I bought an original Rat and used it for a while, but never was really happy with it.
Sound was too dark/midrangey for me and way too dull when turning back the guitar volume.

Reason is the 1nF cap at the opamp input. It acts like adding something like 15m of guitar cable.
So I removed it and also tried lower cap values, but then it oscillated most of the time.
I rearranged wires, used shielded wire, increased filter cap values but nothing cured the instability without the 1nF cap.
So I put it back in but switched to a TS 808 instead.

Later I found that the Rat actually sounds quite good when used after a buffered pedal or a separate buffer.
Actually no surprise as the buffer isolates the guitar from the cap, and the low buffer ouput impedance makes the cap mostly ineffective.

I think the main reason for the instability is the exceptionally high gain of the unit and likely too little separation between input and output wiring.
Input and output of the Rat are in phase, which is actually worst case regarding instability. A buffer or outside unit won't change that.

Never saw signs of instability with the Rat driven from a low impedance source.
So maybe your assembly is just too squeezed.

[Leo_Gnardo]

A squealing rat? Isn't that what they ... neverrrr mind

[tom1970]

The Rat circuit is close to instability and likes to self-oscillate around 3kHz.

In the early 80s I bought an original Rat and used it for a while, but never was really happy with it.
Sound was too dark/midrangey for me and way too dull when turning back the guitar volume.

Reason is the 1nF cap at the opamp input. It acts like adding something like 15m of guitar cable.
So I removed it and also tried lower cap values, but then it oscillated most of the time.
I rearranged wires, used shielded wire, increased filter cap values but nothing cured the instability without the 1nF cap.
So I put it back in but switched to a TS 808 instead.

Later I found that the Rat actually sounds quite good when used after a buffered pedal or a separate buffer.
Actually no surprise as the buffer isolates the guitar from the cap, and the low buffer ouput impedance makes the cap mostly ineffective.

I think the main reason for the instability is the exceptionally high gain of the unit and likely too little separation between input and output wiring.
Input and output of the Rat are in phase, which is actually worst case regarding instability. A buffer or outside unit won't change that.

Never saw signs of instability with the Rat driven from a low impedance source.
So maybe your assembly is just too squeezed.

Thanks so much for that. It's weird as on mine the 1nF cap seems to make absolutely no difference. Also, I still don't really understand why the tube screamer buffer works but my internal one doesn't. I have made it identical to the tube screamer's output buffer so it should be lowering the output inpedence, right?

[Helmholtz]

It's weird as on mine the 1nF cap seems to make absolutely no difference.

Not weird, because after a low impedance buffer output the cap has no effect in the audio range.

Also, I still don't really understand why the tube screamer buffer works but my internal one doesn't.

I think the difference is if the buffer is inside or outside the Rat's case.
Inside it will increase input to output capacitance/coupling - outside not.

[tom1970]

Not weird, because after a low impedance buffer output the cap has no effect in the audio range.




I think the difference is if the buffer is inside or outside the Rat's case.
Inside it will increase input to output capacitance/coupling - outside not.

Got it. Thanks Helmholtz. So you think that there's no way that I can stop the squealing/oscillation by changing something inside the pedal itself? How about if I re-made it using one of these... https://www.ebay.co.uk/itm/313987958...b-ab8971394cd6

[Mick Bailey]

I've had a similar issue, but with one of my own builds rather than a Rat. I ended up replacing the socket connection leads with screened cable, including those to the foot switch. If you decide to do this make sure the screens are only grounded at one end. The PCB may give an improvement, but if part of the problem is lead dress or proximity, then you may still have issues.

[Helmholtz]

So you think that there's no way that I can stop the squealing/oscillation by changing something inside the pedal itself? How about if I re-made it using one of these

Can't say if the PCB would make much difference.
But using shielded wire will reduce unwanted couplings and might be a solution here.

[tom1970]

Thanks so much everyone!

[tom1970]

Hi,
Just a quick update. As a last ditch effort before completely rebuilding I ended up getting a new LM308 from a different ebay supplier and whatdoyouknow it works perfect! After all that pulling stuff out and redesigning, it was just the chip! I think the first chip that I had in initially had way to much gain and a tendency to resonate. I was also noticing with the first chip that the sound would drop out when you hit the low strings hard on full gain. This is not happening now. The new chip is slightly less gain - and this is noticeable when playing muted metal stuff but it sounds very nice. Thanks again for all your help.
CHeers
Tom

[g1]

Hi,
I ended up getting a new LM308 from a different ebay supplier and whatdoyouknow it works perfect!

LM308 is an extinct chip. Fake copies of extinct chips are common now. Chances are the first one you got on ebay was bogus.

[Mick Bailey]

An op-amp gain is set by the associated external component values, not by the opamp itself (under normal conditions). It may be that a defective chip has some internal resistance that's skewing the gain calculation or it's suddenly going to open-loop gain if the feedback loop is internally compromised.