Throbbing 5F1 any theories?

[leevc5]

Hi again, here I am in a mess once more.
5F1 build when volume at 9.5 to 12 play a chord and sounds good but as sound diminishes it begins to oscillate, it sounds like throbbing.
I have checked and double checked wiring, did the chopstick and found nothing.
When I substitute a 12AY7 for the 12AX7 it goes away. If I substitute a 6L6 for a 6V6 it goes away. Hmmm, for some reason the 12AX7's higher out put must be doing someithing to the 6V6. Have a 5751 on order to see if it will handle that preamp tube. But what and how can I fix it so I can run a 12AX7/6V6 configuration.
Pictures included for your examination and a shor movie for your entertainment have been included.
Any and all comments, recommendations, jokes or whatever will be most appreciated:
https://drive.google.com/file/d/0B0Hw7ihDOmlmVTd1VUg0cXlUd1E/view?usp=sharing

[sluckey]

Temporarily disconnect the NFB wire from the speaker jack. Does that help?

[pompeiisneaks]

One suggestion, nothing major, but using wire nuts in amps is usually not a super safe thing to do.  I'd suggest getting some shrink tubing, and connecting the wires end to end and twisting them together tightly in a linemans splice and then melt hte shrink tubing over the wire, rather than wire nuts.  Sluckey's got a great idea there, test it out and see.

Welcome to the forums!

~Phil

[Tony Bones]

Low frequency oscillations are usually a sign of inadequate decoupling, sometimes between cathodes that share a common bypass cap, but in this case I think through the power supply. Coupling between the first stage plate and the output tube screen would do it, and have the side effect of sounding like a trem.

Try clipping more capacitance across either / both of the 8uF filter caps. You didn't reduce the size of the 22k dropping resistor, did you?

http://el34world.com/charts/Schematics/files/fender/Fender_champ_5f1_schem.pdf

[leevc5]

Temporarily disconnect the NFB wire from the speaker jack. Does that help?

Thank you for the suggestion but I just did and it didn't help.

[leevc5]

Low frequency oscillations are usually a sign of inadequate decoupling, sometimes between cathodes that share a common bypass cap, but in this case I think through the power supply. Coupling between the first stage plate and the output tube screen would do it, and have the side effect of sounding like a trem.

Try clipping more capacitance across either / both of the 8uF filter caps. You didn't reduce the size of the 22k dropping resistor, did you?

http://el34world.com/charts/Schematics/files/fender/Fender_champ_5f1_schem.pdf

How would I go about "clipping more capacitance across either / both of the 8uF filter caps?"
Resistor is 22k as specified.
Thank you for looking.

[pompeiisneaks]

Get some alligator clip jumper wires, connect one end to the positive side of a cap, and to a new cap, and the same for the negative being very careful to set it where it can't move etc.  That or tack solder one to the leads temporarily enough to test.  I prefer the alligator clipped leads for temporary testing like this. 

~Phil

[leevc5]

Get some alligator clip jumper wires, connect one end to the positive side of a cap, and to a new cap, and the same for the negative being very careful to set it where it can't move etc.  That or tack solder one to the leads temporarily enough to test.  I prefer the alligator clipped leads for temporary testing like this. 

~Phil

Thank you!!! That stopped the amp from oscillating.
Now, if it's no trouble could you explain in simpleton terms what happened and does it affect the performance of the amp in any way?

[pompeiisneaks]

Per the previous explanation by Tony Bones, you're getting coupling through the power supply due to insufficient filtering.  You can just increase the capacitors there a bit, (if you added another 8uF then make it 16uF) so that it filters better and decouples the power supply more effectively.  You could try just one instead of all of them, (i.e. just replace the one used in the power tube screens with a 16uF).  or whatnot.

~Phil

[Tony Bones]

You mean I was right? 

Anyway, the simpleton explanation is that the filter caps are supposed to shunt any currents that appear on B+ to ground. But, as you know, a capacitors act like a resistor that depends on frequency. At lower frequencies they have higher resistance.

If a PS filter cap is small then at very low frequencies, like 1 Hz, the cap has enough resistance that it doesn't shunt the current to ground without developing some change in voltage in the B+ at the same time. Other circuits that share B+ (namely, the whole rest of the amp) see that change in voltage and might react to it.

So here's the fun part: when you couple a late stage in an amp with an earlier stage you create a feedback loop. If the feedback happens to be positive, then oscillation can be the result. In the case of too small filter caps (or old and ineffective filter caps) the coupling is only at low frequencies, so the oscillation is at low frequency. Possibly one particular frequency depending on how the phase shifts line up.

To be honest, I'm not completely sure where the feedback loop is in your case (what stage is coupled to what) but the output tube coupled to the input makes some sense.

So, to fix your problem you can increase one or both of the 8uF filter caps so they act like smaller resistors. It might also work to reduce the coupling cap at the power tube grid from 0.02 to 0.01. (Remember, there's a loop.) Clip another 0.02 cap across the existing one to find out; two 0.02 caps in parallel = 0.01.

Question for the student: why would reducing a coupling cap kill low frequency oscillation?

[leevc5]

You mean I was right? 

Anyway, the simpleton explanation is that the filter caps are supposed to shunt any currents that appear on B+ to ground. But, as you know, a capacitors act like a resistor that depends on frequency. At lower frequencies they have higher resistance.

If a PS filter cap is small then at very low frequencies, like 1 Hz, the cap has enough resistance that it doesn't shunt the current to ground without developing some change in voltage in the B+ at the same time. Other circuits that share B+ (namely, the whole rest of the amp) see that change in voltage and might react to it.

So here's the fun part: when you couple a late stage in an amp with an earlier stage you create a feedback loop. If the feedback happens to be positive, then oscillation can be the result. In the case of too small filter caps (or old and ineffective filter caps) the coupling is only at low frequencies, so the oscillation is at low frequency. Possibly one particular frequency depending on how the phase shifts line up.


To be honest, I'm not completely sure where the feedback loop is in your case (what stage is coupled to what) but the output tube coupled to the input makes some sense.

So, to fix your problem you can increase one or both of the 8uF filter caps so they act like smaller resistors. It might also work to reduce the coupling cap at the power tube grid from 0.02 to 0.01. (Remember, there's a loop.) Clip another 0.02 cap across the existing one to find out; two 0.02 caps in parallel = 0.01.

Question for the student: why would reducing a coupling cap kill low frequency oscillation?

Thank Mr. Bones!!! I spent almost a week going back and forth with tech support at TubeDepot (providers of the board and components). They had me going round and round insisting it was a bad ground contact somewhere. After I did everything they recommended and more they finally said I should return the amp to them to troubleshoot.

You knocked it out of the park first time up.

[Joel]

This right here is exactly why I come to this forum.  Golden nuggets of useful information given out for only the price of asking nicely.  Thanks guys for being awesome gurus.