Hum After Zener Diode Voltage Reduction

[Chuck H]

Good observations. The bias circumstance was mentioned earlier with a link. The lamp could be an issue. Notice that there is no resistance where it taps into the HV supply at the PI tube. If that filter isn't 100% (and even new ones aren't) that neon lamp noise could be an issue. Maybe try pulling the lamp and see if that doesn't kill the noise. It should be pretty easy to change to a standard incandescent lamp on the primary side if this turn out to be the case.

And, as mentioned twice, DO CHECK BIAS VOLTAGE AND CURRENT!

[Daver]

Good observations. The bias circumstance was mentioned earlier with a link. The lamp could be an issue. Notice that there is no resistance where it taps into the HV supply at the PI tube. If that filter isn't 100% (and even new ones aren't) that neon lamp noise could be an issue. Maybe try pulling the lamp and see if that doesn't kill the noise. It should be pretty easy to change to a standard incandescent lamp on the primary side if this turn out to be the case.

And, as mentioned twice, DO CHECK BIAS VOLTAGE AND CURRENT!

I have already removed the lamp from the circuit. I did measure the bias voltage, but I'll do it again and report back. I don't remember the exact numbers, but it wasn't as bad as you would think. Even with the standby off there is some bias voltage and some B+ being generated. It's kind of and odd amp. Thanks for the suggestions.

[Daver]

OK, grounding the center tap for the heaters (bypassing the cap) didn't change anything. Without the zener the bias voltage at the bias filter cap/diode junction is -22.52 volts and with the 47 volt zener in it's -23.81 volts. Not a huge difference. Any other ideas? Thanks!

[dstrat]

ok That is unexpected... where are you taking your bias reading ?

schematic shows -50v at the 270k resisters.

seems like the zener is not doing anything, amp is not pulling enough current to work right.

http://music-electronics-forum.com/t44099/

gives some details.

[Chuck H]

Well the Zener is dropping it's voltage. So it IS doing something. -22/23V bias should be a concern if the design calls for -50. But even at high-ish plate volts an assignment of -50V bias surprises me because 7591's are sort of like uber 6bq5's in that they don't usually run a big bottle bias voltage. Bias should be adjusted by current anyway. Not a schematic voltage or a circuit component value.

I have NO freeking idea why the amp is humbuzzing at 120Hz. At this point I'm thinking there is another issue with the amp and the Zener isn't the problem, but it's letting us know there is one. We've ruled out the bias circuit (whether the bias is off or not). We've ruled out the neon lamp. We've ruled out the filament CT. We've ruled out a hot chassis. So...

Something is missing from the equation. I know the amp supposedly has a grounded AC cord (that I assume has the green wire bolted to the chassis) so WHY IS THE DEATH CAP STILL IN PLACE??? Check to be certain the chassis is grounded and snip out the death cap.

How old are the filter caps and bias caps. I don't recall this being covered in the information yet.

[dstrat]

Chuck your right , the schematic must be wrong.

agree that cap on line ac should not be there.

[Daver]

Yeah, the schematic is incorrect for the negative bias voltage. I'll try removing the death cap. The filter/bias caps look very recent. The green AC cord wire is connected to its' own lug to ground on the chassis. I agree that the zener may be something that is revealing another problem. It's just weird that the phase inverter has something to do with the noise only when the voltage is lowered. Thanks!

[Daver]

I removed the death cap and that didn't change anything. I guess I'm going to look into other ways of reducing the voltage.

Edit: I removed the negative feedback resistor. That made the hum noticeably louder. The NFB resistor connects to the cathode of the 7199 pentode half and that seems to be where the noise is coming from. But only with the zener in there.....

[Daver]

FIXED! I connected the can cap ground that feeds the power tube plates (OT) and screens to the "top" of the zener putting it at the same potential as the center tap. I moved the OT secondary ground tap that was connected there to the power tube cathode/bias circuit ground. The hum is GONE and the voltage reduction is still there. Joy! Thanks for all the help!

[Jazz P Bass]

Cool.

[Chuck H]

Hmmmm...? I also think that's cool. but... So the OT secondary termination was elevated at the Zener voltage? I wonder if that was the trouble this whole time.?. If this is so I might try grounding the power supply filters back at 0V just to see what happens. I would think it's best to have those filters at 0V because they not only smooth DC as reservoirs, they also decouple AC from the the supply rail. I would be inspecting hum levels between the current wiring and with no Zener and the old wiring.

[Daver]

No, no, no. The secondary was NOT elevated before or now. It was connected to the can cap ground originally. When I moved the can cap ground to the Zener, I grounded the secondary elsewhere so it wouldn't be elevated.

[Chuck H]

Ah, ok. Well... That would mean that the screen grids are no longer fully bypassed and the plates 0V reference must be elevated while the peak voltage is reduced? I wonder how this will affect performance. I can't wrap my head around it yet.

[g1]

I can't wrap my head around it yet.

Me neither. A pre and post op schematic help. The other thing is that can type cap's grounds are almost always physically connected to chassis. Is this one of those rare ones that is isolated, or did you have to insulate it from chassis Daver?

[Daver]

It's a JJ can cap, so there is a separate ground lug that is not connected to the case. Here is where I got the idea to try it. What is Back-Biasing?
I didn't need to generate a bias voltage, but if you read down, he connects the filter cap ground to the Zener. I figured it was worth a try.