distortion on leading edge of note

[pula58]

new build.,
Fender-ilke, ab763, but single channel, no "normal" channel. Sounds great if plucking the string lightly, but, with notes with a hard attack I am hearing a non-harmonic distortion of some kind...very low pitch.

If I put in a 100mV test signal and crank the amp up (into a 8 ohm resistive load) I see nothing strange. With the output just at the edge of clipping I see 50W RMS (just like I planned) and no distortions anywhere that are out of the ordinary. I have not found any oscillations either.

I tried testing the amp, with resistive load, using my guitar and hitting the strings real hard, but, the guitar signal is very erratic, and all the signals in the amp are similarly erratic, so, its hard to spot what might be going-on,

Any tips as to how to find what is going on here?

P.S.
Seems like the power supply filtering is working right. grounding scheme as follows:

star ground with:
HT center tap
chassis ground
1st power supply cap
bias circuit gnd
cathodes of both output tubes (6L6GC).
speaker jack ground

then, screen cap, and all other supply caps are grounded near the input jack. I have a ground bus extending out from the input jack ground, it is isolated fro mthe chassis except at the input jack, where it connects. All the cathode bypass caps in the preamp are grounded to the bus. Anything in the tone stack that needs ground also gets it from the bus.

[JazzGuitarGimp]

I expect someone with a better understanding will chime in, but this kind of sounds like blocking distortion to me. You might try setting your signal generator to a high frequency (say 5KHz) and use a momentary switch to inject a burst of this tone into the amps' input. Use a dummy load on the output, and then use your 'scope to look for low-frequency artifacts associated with the 5KHz burst.

[pula58]

Thanks Lou

here is link to a recording. You have to listen carefully. It is a low note, that pulses twice, with each attack of the main note.

http://www.seanet.com/~pauls/amp_distor ... tifact.WAV

[JazzGuitarGimp]

Nope, not blocking distortion - that's 120 Hz, most likely ripple from the power supply. Do you have enough filtering in the design? Are you using a choke for the screens and preamps?

[JazzGuitarGimp]

And does this only occur at high volume / power levels?

[pula58]

1st power supply cap is 47uF 500V F+T.

then, a choke (Hammond 194B, 4H, 90mA, twin reverb size)

then, 22uF 500V F+T screen supply and reverb power supply
* reverb gets power from screen supply through a 2W 5K resistor

then, 2.2K 2W resistor

then, another 22uF 500V F+T

then, 7.5K 2W resistor

and, the last 22uF cap (preamp power supply), another 20uF 500V F+T.

[pula58]

And does this only occur at high volume / power levels?

seems to only be at loud volumes. If I turn down the master volume (pre PI master volume) the effect does not seem to happen.

[crbowman]

Have you tried a different guitar?
Listening to the sample, almost sounds like it could be fret buzz farther up the neck.

[Structo]

Is this with a single coil type guitar?

If it is, if you have a humbucker guitar try that.

What I am thinking is maybe the pickup is too close to the string.
Stratitis

Also, check all filter caps for leakage.

Where is your output secondary grounded?

[pula58]

bowman, structo: It's a Strat ...but...my other amp, with same guitar, same speaker, in same room, does not exhibit this behavior.

[pula58]

Is this with a single coil type guitar?

If it is, if you have a humbucker guitar try that.

What I am thinking is maybe the pickup is too close to the string.
Stratitis

Also, check all filter caps for leakage.

Where is your output secondary grounded?

The output tranny secondary is grounded at a star connection along with:
1) Main filter cap (47uF)
2) cathodes of output tubes
3) ground of bias supply
4) power tranny heater center tap
5) power tranny HT center tap
6) power tranny hum shield.
7) chassis ground connection to ground of ac power cord

[JazzGuitarGimp]

I'm wondering if 4H is too little inductance. The effective resistance of 4H at 120 Hz is 3,016 ohms and the 22uF that follows it has an effective resistance at 120Hz of 60 ohms, This gives you about 34dB of ripple rejection. Do you have a bigger choke you could temporarily swap in for a quick test?

[pula58]

HI Lou,
I don't have another choke around...but..isn't this a typical value (4H, 90mA) for a two 6L6 fender blackface/silverface style amp?

[JazzGuitarGimp]

HI Lou,
I don't have another choke around...but..isn't this a typical value (4H, 90mA) for a two 6L6 fender blackface/silverface style amp?

Yes, 4H is correct, and Fender typically followed this with a 20uF cap (at least in the 45W BF Super Reverb). But I really think you've got an issue with ripple. Another thing you could try, which would be the equivalent of doubling the choke to 8H, would be to double the cap right after the choke to 44uF (you could just tack solder another 22uF across the 22uF that's already in there).

Another thing to try, would be to put your scope on the screen supply node, AC couple the input, set the timbase to 10mS and the input sensitivity to 5V/div. Center the trace before you connect it to the power supply screen node. You will be able to see the ripple and measure its amplitude. Then crank the amp up, and play the single note like you did in the sound clip and see if the ripple gets really large when you hear the hum along with the guitar note.

[Diablo1]

You could try to add some caps in parallel to your first and 2nd filter cap on your Pi filter. That would tell you if you have a ripple problem or not.