problem: getting sound with volume set to 0

[Pete]

I have been using my amp since I built it some years ago.
Never noticed this before but I can still hear quite a bit of sound with the volume all the way off (Master all the way up).
Changing the pot made no difference. Tone (TMB) settings have no effect on the problem.
Is this normal? Do I have something wired wrong?
Any thoughts appreciated.
Pete T

[jaysg]

This is usually caused by wiring. Remember there's no volume knob until there's been some gain. The first stage boosts the signal and despite knob settings, a later stage is picking it up. Parallel stretches of wire can behave like transmitter/receiver RF pairs. Study the wires from B+ to the plate resistors and from the plate resistors to the tubes.

[George Adzich]

Pete,

I have also found this problem in amps using dual triodes. I think the problem can also be caused by the interaction of the triodes. Even though you turn the volume down or even short the output to ground, the signal is still "inside" the tube and can bleed out through the other triode.

I cannot remember the term for it right now, but it does have an explanation. I think that ther are actully some dual triodes that have an inner shield to help out with this problem.

I have asked about this problem on other boards and not found a better answer. I think a lot of amps exhibit the behavior, but no one worries about it or even checks to see if happens in their amp. If the output of the second tride is grounded, you should see this problem go away, if it is not a wiring issue.

I am actully going to build a Dumble clone using all single triodes (6AV6). This is really overkill, but when you can pick up vintage NOS RCA 6AV6 tubes for 50 cents, it is hard to pass up.


Good luck and keep us posted on what you find

[ElectronAvalanche]

Hi there,

I think grounding has a lot to do with this.

Make sure you have the "proper" grounding for your amp.

In my 75W ODS with Reverb I am still struggling with the same phenomenon.

Changing the grounding a bit helped me get it to the point, where without the Reverb, there is no "bleed through" with the MV all the way down.

To pinpoint the offending stage, try grouding specific points of the ckt. Ground the Clean output cap and see if you still get this bleed through. Start from the PI input cap and work your way to the input of the amp.

Cheers,


Dominik

[ayan]

Pete,

I have also found this problem in amps using dual triodes. I think the problem can also be caused by the interaction of the triodes. Even though you turn the volume down or even short the output to ground, the signal is still "inside" the tube and can bleed out through the other triode.

I cannot remember the term for it right now, but it does have an explanation. I think that ther are actully some dual triodes that have an inner shield to help out with this problem.

I have asked about this problem on other boards and not found a better answer. I think a lot of amps exhibit the behavior, but no one worries about it or even checks to see if happens in their amp. If the output of the second tride is grounded, you should see this problem go away, if it is not a wiring issue.

I am actully going to build a Dumble clone using all single triodes (6AV6). This is really overkill, but when you can pick up vintage NOS RCA 6AV6 tubes for 50 cents, it is hard to pass up.


Good luck and keep us posted on what you find

While pots and grounds might have an effect on this, as per my experience it's been the 12AZX7 cross talk that causes this in amps -- I've experienced this in Fenders and Boogies too. As George says, tube cosntruction has a lot to do with it, and if there is properly internal shielding, things work well.

Cheers,

Gil

[ElectronAvalanche]

Hi Gil,

AFAIK no double triode tubes have any shielding between the two systems. The anodes would keep any stray electrons from one system entering the other. I doubt this is really the problem.

When swapping tubes makes this phenomenon go away, this could easily be that the tubes have different gain, thus affecting the ground bleeding or bleeding in the Power supply.

Sometimes this bleeding can be caused by a bright cap. Try to remove it and see if the problem goes away.

Cheers,

Dominik

[llemtt]

no double triode tubes have any shielding between the two systems

Dominik, this isn't completely true, there are chinese 12AX7 that have a shield between the two halves (I own a few) and also other tubes do, anyway cross capacitance between the two plate of a 12AX7 is less than 1.6pF and cross capacitance between other elements of the two halves is less than 0.01pF...

When I first noticed this leakage I immediately thought to internal tube coupling and I swapped a shielded tube in V1 with no effect, the cause is another one.

Pete your signal leakage is somehow "normal" because the grid of CL2 isn't grounded when you turn volume to 0 but instead it connects to ground through the 220k grid resistor. Do a test, ground directly the grid and every leakage should disappear.

cheers
Teo

[ElectronAvalanche]

Hi Teo and Gil,

ok I found some info on the shield. I have never seen that before and only one chinese 12AX7 has some sort of shield (metal element connected to the getter, but no connection to any external pin).

I stand corrected...

Cheers,

Dominik

[VacuumVoodoo]

Hi guys,

my 1st post here...been lurking for a while but had nothing constructive to contribute since I'm not building any TW or D-clones, prefer to go my own way

There are several dual triodes with shield between the halves, these are mostly of russian origin, some are still in production. The types are 6N1P, 6N2P etc. Shield is brought out to pin 9 and should be grounded. Heaters are internally connected in parallel and heater syupply is 6.3V only connected between pins 4 & 5. You can not plud these directly to a 12A*7 socket without rewiring it for shield and heater supply.

Bilingual datasheets:

http://www.mif.pg.gda.pl/homepages/fran ... 6/6N2P.pdf
http://www.mif.pg.gda.pl/homepages/fran ... 6/6N1P.pdf

6N1P is an excellent 12AX7 alternative, milityary version with EV suffix has guaranteed operational life over 5000hrs.

Reagards
Alex

[Dai H.]

wasn't the Sovtek 12AX7WA a re-based version of that tube (I think those have a mu of 90 versus 100 for 12AX7 though). I also have an old Tesla 6CC41 which also the same basing (equivalent to 12AX7 w/internal shield between sections). Maybe they were common in E.Europe or something.

[Pete]

Thanks to all for the comments .. I think most of my wiring/lead dress is typical of these models (similar to pics I've seen) not sure what to look for around PS/B+.
I did take some resistance measurements from ground to some obvious places like at the ground side of the vol pot, and didn't see anything out of the ordinary (IMO)

from George:

If the output of the second tride is grounded, you should see this problem go away, if it is not a wiring issue.

Shorting the output of the second triode kills all sound, as does shorting the PI input (basically the same point in the circuit right?)
Also,
- other than Bright switch, there are no bright caps in the amp.
- there is no sound with the master all the way down.

from Teo:

Do a test, ground directly the grid and every leakage should disappear.

OK, I just tried this (grid shorted directly to ground) I got a small decrease in sound (and noise) level, but for the most part, no change.

So, is there room for improvement? I suspect something is "not right"
Pete T

[Bob-I]

Every tube amp I've ever owned had some bleed through when the volume was a 0. I've always suspected crosstalk in the dual triodes as the cause. Since it's so minimul I've never spent any time trying to correct this.

[pedro]

FWIW - I have used the 6N1P in my 2 headphone amps and its a great tube.
Would recommend it for sure - maybe worth looking into for guitar applics.

As for the channel bleed - just dont worry about it - I simply suggest playing at more than "0" volume settings

cheers
Pete C

[MarkB]

Foolish question time - is this a Dumble amp? If so, then the bleed-through would be out of phase, no? I'd want to get rid of that as a matter of principle.

Even when not playing with the volume on 0.

[Dai H.]

This is usually caused by wiring. Remember there's no volume knob until there's been some gain. The first stage boosts the signal and despite knob settings, a later stage is picking it up. Parallel stretches of wire can behave like transmitter/receiver RF pairs. Study the wires from B+ to the plate resistors and from the plate resistors to the tubes.

yeah think about that. Which stages are the volume(s) affecting or not affecting in the chain. It should be possible to find the source. Or otherwise, maybe a pot that isn't shorting the wiper enough to ground when in the zero position.