Mesa Subway Blues some hum/buzz 100Hz, out of ideas

[iefes]

Chuck you nailed it. That's how the amp is wired right now! Thanks for putting the schematic together. The bright-cap is still the original 120p but that's not important at the moment.

[Chuck H]

Ok... You lifted the components at the 15k, 33k, .02u junction and the noise remained? Are you sure it's the same noise and not just an equal amount of different noise? I'm considering that by lifting the components you leave the PI input grid with considerably less load.

[iefes]

Yes I lifted them BUT I also connected them together to make sure those loop capacitors are not connected to the circuit anymore while retaining the rest of the circuitry. So the PI should still see the same load.

[iefes]

Okay, as per Pedros suggestion I disconnected the cathode capacitor and resistor of V2a and V2b one at a time. The results are about the same as with the anode-capacitors: V2a cathode lifted = hum ; V2b cathode lifted = no hum. Should I go on checking the cathode caps + resistors on the other stages or does this suffice?

I will put together the complete schematic of the amp in the next days during which I won't have time to work on the amp anyway.

Thanks for your help everybody!

[iefes]

Okay, as per Pedros suggestion I disconnected the cathode capacitor and resistor of V2a and V2b one at a time. The results are about the same as with the anode-capacitors: V2a cathode lifted = hum ; V2b cathode lifted = no hum. Should I go on checking the cathode caps + resistors of the other stages or does this suffice?

I'm wondering if the FX-Loop changes the phase of the signal or retains it. With a changed phase this odd 500pF loop capacitor would see opposing phases on either end and I'm wondering if this could induce the hum. However, as the hum was still present with the parts lifted and connected without the loop capacitor that's probably not the source of hum.

I will put together the complete schematic of the amp in the next days during which I won't have time to work on the amp anyway.

Thanks for your help everybody!

[ric]

I'm wondering, in post #11 you mention checking the 47n coupling cap from V1b to V2a for capacitance. Could it be leaking DC?

[iefes]

Ric, thanks for your suggestion. I will try to measure if there's DC on that capacitor during the next day when I'm back home.

I've now put together the correct schematic of the preamp-section with the FX-Loop left out. So this schematic should be a correct representation of what we're dealing with. Hope it helps! :-)

Edit: The 500pF cap forming this loop represents the supposedly two caps that are in parallel on my amp.

[iefes]

I just realized that the coupling cap (C9 in my schematic) that Ric advised me to check for leaking DC is probably fine. The buzz was still unaffected with the coupling cap of the next stage removed (C10). So I suspect that the noise is produced after C10. With C13 removed, there's no hum. That's really strange, there are only very few parts left.

[iefes]

Okay, as ric suggested I checked if the 0.047u coupling cap between V1b and V2a is leaking DC and was able to confirm that the cap is fine. There was some additional buzz coming out of the speaker when measuring but that's probably normal. I also checked the next coupling cap for leaking DC but this one was fine too.

Any further tips somebody? I'd be so happy to get this thing up and running.

[DrGonz78]

Pardon me if this has been checked already... Have you gone through the amp with ohm meter to check all the grounded points? I see an amp that is coated in black and my first thought is to make sure that the mains ground is making good chassis connection. Probably checked but skimming through the thread I was unsure if you had checked the safety ground to see if it is making good contact with chassis. Also, read somewhere that the common ground from OT was floating causing some noise. Easy to run through the amp and confirm, even for a second time.

Edit: BTW this is the thread about the OT secondary ground connection...
http://el34world.com/Forum/index.php?topic=9519.0

[iefes]

Hi DrGonz! Thank you for joining the discussion. I have indeed checked all ground-connections in the preamp-section with an ohmmeter against the common chassis-connection at the input jack and all measured well. The safety ground is making good contact as well. I think, if the latter was the problem the buzz would be audible not only with V2 inserted. Please correct me on this if not.
I came across the thread a few weeks ago and checked the ground connection. In my amp there's a ground wire parallel to the feedback-wire going back to the PCB, so it's not floating. However, I think if this was the problem, we would as well hear the buzz no matter what tubes are pulled in the preamp. From pulling anode coupling capacitors I am quite confident that the buzz is caught somewhere between V2a and V2b. Lifting the Cathode Resistors/Capacitors on these stages revealed the same.

[DrGonz78]

So after re-reading this thread I get that the issue is between V2a and V2b with something in there being the source of the buzz/hum. What about these mojo modules? It really seems that those potted modules might have been sensitive to heat in many other Subway amps. Has anyone ever found these modules to be conductive with other parts on the board or something? Perhaps the amp in this thread is just past the heating up phase to get it to hum and now it does not need heating to hum? Then again I guess those modules were just not shielded or something. So maybe I am way off even thinking it has anything to do with conductive crud.

The silly things I would try would involve pulling up one side of C10 and run a bypass wire bypassing V2b. If that got rid of the buzz then I might think of leaving the amp in that state for the type of gain needed for bedroom levels. This also assumes that V2b is the main problem here. Still would be interesting to know if this lowered the hum or got rid of it at least.

Edit: Just a last thought on those mojo modules. So if Mesa was saying to shield the module as a fix that does not make sense to me. For one thing why would those other amps have to heat up before humming would become an issue?

[iefes]

Hi DrGonz, thanks for your suggestions.
I will definitely try to bypass V2b like you describe it and go from C10 directly to the PI-Input. This will basically make it a ab763 topography until the PI.

The Mojo-Module in my case is completely bypassed. Even the transformer tap that feeds it with power is disconnected. So this only leaves the Bias supply which is still intact and covered with epoxy. The second much smaller epoxy block however is still in the circuit between V2a and V2b as can be seen from the schematic. This thing could cause some issues but I don't see why it would buzz, especially with only being three components in there.

I got the tip to change the tube sockets as they can cause issues in a few cases. If this won't help to get rid of it I'll think about rebuilding the whole preamp-section on turretboard and use a convenient grounding-scheme etc. Some kind of AB763 preamp with the Mesa style PI and Output-stage might be worth a try, this would additionally leave one triode that could be used for a sweet Bias-Tremolo.

But first, I will try more to get the thing running as is. Cheers everybody!

[Chuck H]

The second much smaller epoxy block however is still in the circuit between V2a and V2b as can be seen from the schematic. This thing could cause some issues but I don't see why it would buzz, especially with only being three components in there.

Mesa has always been known to have "errors" in the schematic. Intentional errors meant to throw off cloners. W don't KNOW what's in there. And a bias supply can be a good place for crud to get into the circuit. You could try cathode bias just to see what happens. I can't remember board specifics, traces and board mounted sockets may complicate the effort. But you just need to remove the grid leaks from the bias supply and reference them to ground and then lift the cathodes from ground and reconnect them through a shared 220R/10W (ish) resistor parallel with a 100uf/50V cap. That would eliminate the bias supply questions.


EDIT:Oh yeah.?. We already eliminate the bias supply with removal of V2. Nevermind.

[iefes]

Thanks for your suggestion Chuck but you're right, the bias shouldn't be the problem because there's no hum when V2 is pulled.

I tried DrGonzos suggestion and connected C10 directly to the Input of the PI, completely bypassing V2b and the periphery around it. This didn't help with the hum unfortunately but the volume was very high (all those voltage dividers jumpered) and thus the noise was higher than before.