Killing the hash PO

[BobW]

Structo,

Your op-amp supply should have .01 ceramics to decouple the noise. You could also try decoupling the noise at the input w/ a .001 - .02 and or adding a series input resistor to slow down the response time. Also what resistor values are you using on your op-amp loop? You don't want any caps in the loop, doing so will cause the opamp to act like an integrator.

[Structo]

Heheheheh,
Yeah, all of the above.

Just an ongoing problem I've had with this amp.
Overall it sounds great but it has a bit of fizz or hash that rides on the notes.

Eventually I'll track it down but I have been picking the brains of the smart guys here.

Deric, on the last time I was trouble shooting it seemed to be clear with the loop bypassed.
It could even be a pedal that is doing it.
I just need to devote some serious time in a day to better track it down.

I'm too lazy to go back and read all the posts - Are the O.T. leads bleeding to some signal wire in the build - I've had problems with the OT secondary wires (going to the Z selector) being too close to the PI grid wires causing "hash".

I can't remember off hand but I will certainly give that a look.
That is probably the weak point in my lead dress because it is a bit cramped where the primaries and secondaries are routed.
Thanks.

Yes I have considered building a Kleinulator but as I mentioned I would really rather not have an outboard unit.
I'm sure the FET's would probably do a better job on the high voltage swings than the opamp.
BobW,
The way the Ironsouds loop is hooked up it sends to the .02uf cap on the PI input and is fed off of the Master vol.
Are you saying to decouple it further with a .001uf in series with the in and outs?
Or to put those in parallel with the coupling cap on the PI?

Or are you talking about the loop power supply?

I included the Ironsounds schematic.

[BobW]

Are you saying to decouple it further with a .001uf in series with the in and outs?
Or to put those in parallel with the coupling cap on the PI?

Or are you talking about the loop power supply?

I should have been clearer, here's a few suggestions:
1. The 2 decoupling caps on the power supply should be as close to the op amp as possible from +vcc, pin 8 to Vgnd, and from -vcc, pin 4, to Vgnd.

2. Decouple the noise at the non-inverting input w/ a 47 - 100pF across R10. This cap and R14 form a LP filter with the knee at approx. 3kHz. to 7kHz.

3. You probably don't need C4.

4. Make sure your voltage rails are below +/- 18V.

5. Keep your lead lengths to the op-amp inputs as small as possible.

I see the voltage rails have no current limiting (series resistors). You may want to add resistors (1k) in case you fat probe across any pins.

[Structo]

Installed the new 2134P opamp yesterday.
Seems just a bit better.

I found out that the Ironsounds loop is nothing more than a Marshall FX loop used in some of the later amps.

Bob, not sure what you mean by keeping the voltage rail below 18v .

That was one of the problems that Steve at Metro said was an issue with this loop.
He said average signal to the loop is over 30v so the opamp is clipping anything above what the opamp can handle.

His loop uses some higher voltage FET transistors that will not clip in the range that the PI sees.

I suppose it couldn't hurt anything to try a couple .001uf ceramics on the + and - to Vgnd.

Thanks for the tips.

[odourboy]

That was one of the problems that Steve at Metro said was an issue with this loop.
He said average signal to the loop is over 30v so the opamp is clipping anything above what the opamp can handle.

FWIW, I don't completely agree with this statement. First, the zeners on the opamp inputs are going to protect the opamp from clipping. (Opamp clipping can get pretty ugly, but zener clipping can be somewhat musical.) Second, while it's true that the signal levels into the loop can exceed the +/- 12V headroom of the loop with the master up high, I think you'll find the PI/Power section is already well into clipping at that point, so the lack of headroom in the loop is somewhat moot.

[Structo]

That was one of the problems that Steve at Metro said was an issue with this loop.
He said average signal to the loop is over 30v so the opamp is clipping anything above what the opamp can handle.

FWIW, I don't completely agree with this statement. First, the zeners on the opamp inputs are going to protect the opamp from clipping. (Opamp clipping can get pretty ugly, but zener clipping can be somewhat musical.) Second, while it's true that the signal levels into the loop can exceed the +/- 12V headroom of the loop with the master up high, I think you'll find the PI/Power section is already well into clipping at that point, so the lack of headroom in the loop is somewhat moot.

OK, that sounds reassuring.
I don't know enough about this stuff to make a judgement either way.
So I have to take what others say in consideration.
I know what I hear and that is all I can go by.
I messed around with it and after taking the preamp out of the equation and the power stage out, I'm thinking the noise is most likely in the preamp.
I probably just need more tubes to roll through the amp to find the best combination.

Funny, I cranked it up yesterday afternoon and man, does this sucker get LOUD!
I don't turn it up very often and haven't really cranked it since I got the other EVM 12L.

In overdrive it just kills!
I don't think in a gigging situation that the fizz would be that noticeable.
It's just that we tend to be our own worse critic and overanalyze the amps we build.

With two EVM 12L's cranking, the punch and chugga chugga of those drivers will weaken the knees of any mortal being.
Of course my neighbors might disagree.

Power chords through humbuckers were blissful.
And some say these amps aren't for rock.......