distortion on leading edge of note

[surfsup]

Is your PT secondary CT connected to chassis star or to the first "resevoir" filter cap directly?

[ampgeek]

Can you quickly check for any "wiggle" in your bias supply circuit?
Although it would likely be 60Hz if you are following the standard plan, it couldn't hurt to eliminate that as a possible source.

Good luck!
Dave O.

[TUBEDUDE]

If all else fails, you can try this: Drive the input with a signal generator at a medium level. Scope it with channel A. Hook scope channel B to the output. Display both signals, adjusting the scope channel levels so they are identical on the scope display. Now invert the second channel and combine the channels. The display should be a flat line, Now plug in the guitar, crank up the amp till the offending distortion is heard and the scope should show you the actual distortion signal, ie. the difference from the guitar itself as the guitar signal is phase reversed and about equal (this may require some fine tuning on the scope to account for minor level changes when the guitar is cranked up).
Anyway, the resultant signal shape can give you some clues as to the nature of the distortion, its frequency, period, harmonic relation to the original signal etc.

[pula58]

Is your PT secondary CT connected to chassis star or to the first "resevoir" filter cap directly?

The "-" lead of the first cap goes through about 5" of # 20 (maybe # 18)wire to the star ground. The ground connection of the bias circuit filter cap joins this 5" trace about 1/2" from where the first filter cap is.

like this: ( == indicates wire)


star================== "-" terminal of bias cap=== "-" term of filter cap

By the way: I tired increasing the capacitance from 47uF to 77uF on the first (main) cap. It made no difference. I also tried increasing the filter cap on the screens from 22uF to 52uF, that too made no difference.

[tubesinside]

I have had similar issues with AB736 50 watt build (reverb channel) where I had strange sounds at near max volume into a speaker but nothing evident into test load. Try scoping the output with the amp cranked into a speaker (mind your ears/speaker/neighbours) and see if there are any parasitics on the waveform. The extra inductance may be causing instability.
I sorted mine by dressing the output tube grid properly. Quick test is to bridge the anodes of the driver tube with a 120pF mica and see if the issue disappears.
Cyril

[surfsup]

have you tried connecting the CT directly to the (-) of the resevoir cap?

[pula58]

I have had similar issues with AB736 50 watt build (reverb channel) where I had strange sounds at near max volume into a speaker but nothing evident into test load. Try scoping the output with the amp cranked into a speaker (mind your ears/speaker/neighbours) and see if there are any parasitics on the waveform. The extra inductance may be causing instability.
I sorted mine by dressing the output tube grid properly. Quick test is to bridge the anodes of the driver tube with a 120pF mica and see if the issue disappears.
Cyril

Do you mean to bridge the anodes of the phase inverter?

[tubesinside]

yes,
This is just a quick test to see if the instability disappears .....real work begins when you have to find the lead dress problem.
C

[pula58]

yes,
This is just a quick test to see if the instability disappears .....real work begins when you have to find the lead dress problem.
C

I tried a 250pF silver/mica cap between the anodes of the PI tube. Result: No change

I neatened-up the leads to the grids of the output tubes: no change

I gave the bias circuit capacitor its own wire to the star-ground. Previously, it was sharing the wire of the (-) end of the first (main) filter cap to the star-ground.

I still have the distortion artifact.

I have not yet tried connecting the CT of the PT HV secondary directly to the main cap (-) terminal. Seems to me that if I do that I need to move the output tube cathode cinnections to the same point. Currently the output tube cathodes and PT secondary HV CT connect to the star ground. The (-) terminal of the 1st filter cap connects to the star through a 6" piece of # 20 (maybe # 18) wire.

[pula58]

I just recently tried changing the ground-connection of the filter cap for the screens away from the preamp (input jack) ground. Made no difference.

Seems to happen when (looking at o-scope) the output (as measured at output jack, or speaker) starts to clip.

[martin manning]

I'm thinking this may be "ringing" that appears when the waveform squares-off. Got pics to post and a schematic?

[pula58]

seems to be something having to do with the phase inverter outputs getting clipped when the grid of the output tubes gets driven above 0V.

[pula58]

see schematics here....

[Firestorm]

I originally thought this might have been the "gain-spike" effect Merlin discusses, but your PI tail seems too small. Maybe an NFB issue? If the PI is significantly unbalanced (more in terms of current than transconductance), the NFB may be trying to correct the signal. Have you tried varying the feedback? Or increasing the gridstops on the output tubes? Sometimes you can achieve nice things by slightly reducing the grid loads from 220K to 200K or lower.

[cxx]

I had something like this that turned out to be the power tubes. Some strange sounds that were transient around the note. I chased it around for a while before finding it. Swapped them out and it went away. Afterwards I found I could tap those tubes and get the effect.