Ampeg Gemini I - hum hum hum !

[pula58]

Man I have tried everything trying to quiet this amp down!

60 and 120 Hz hum. Even if I remove all the preamp tubes! If I ground the input to the PI (oin 7 of 7199 PI tube) at the resistive summing point of the two dry channels and the reverb the amp goes completely without hum. The hum is independant of swapping two new PI tubes.

It seems to be that this hum is coupling into the circuit right at the summing junction because if I ground the "other side" of all the summing resistors (150K, 270K and 270k) the noise does not get any better (By other side I mean the opposite end of the resistors, the ends NOT at the summing junction). But if I ground the summing junction itself all the hum goes away.

I was measuring the bias of the output tubes to see is possibly some imbalance in biad current is bothering the OT. I used an ammeter across each half of the primary and measured 50mA and 45mA (@ 393V plate voltage thats ~ 19.5W each). Lo and behod, the hum goes away almost compltetly when I was shorting each half of the OT primary (one at a time) with the ammeter.

I note that he OT is located on the bottom of the chassis directly underneath the three-resistor summing junction.

So, any ideas on what is going on here? Mismatched output tubes? bad OT? Output tubes running too hot?
Bad Layout? In which case: Do ALL Ampeg Gemini I's have this hum??
I am going to try to move the OT...but..the darned Ampeg folks use mounting screws with a rectangular head depression and none of my driver heads fit it...a trip to the hardware store......

[9pins]

sounds like you've isolated it to the power side. Old OPT may be, but check the first couple of filter stages in the PS up to the power tubes, does the PS or the OPT tranny look rusty over the lamination's?

[Firestorm]

So if you pull all of the preamp tubes, you still have hum. But if you ground the PI input, the hum goes away (I'll bet if you pull the PI tube, the hum goes away, too). That tells you the hum is not in the output stage, nor in the preamp, but in the PI itself.

The pentode section of the 7199 is sensitive to induced noise, so you want to get rid of possible sources. Check the quality of your grounds: the ground reference for the PI input is through those summing resistors back to the input jacks (the whole ground scheme in the Gemini leaves something to be desired and I have sometimes had to rework it completely). The 7199 is also notorious for picking up heater hum. In your version of the Gemini (the schematic you posted), the heater center tap is grounded through a cap; later versions used a hum balance pot, which suggests that Ampeg found resistive balancing to work better. Better still might be to DC-elevate that heater reference; you could start with the output tube cathodes and see if that helps.

On the off chance there's a leaky coupling cap somewhere, you might check DC at Pin 7 of the 7199 (pull the 7199 first, because the grid will have a little DC on it in normal operation).

[pula58]

So if you pull all of the preamp tubes, you still have hum. But if you ground the PI input, the hum goes away (I'll bet if you pull the PI tube, the hum goes away, too). That tells you the hum is not in the output stage, nor in the preamp, but in the PI itself.

The pentode section of the 7199 is sensitive to induced noise, so you want to get rid of possible sources. Check the quality of your grounds: the ground reference for the PI input is through those summing resistors back to the input jacks (the whole ground scheme in the Gemini leaves something to be desired and I have sometimes had to rework it completely). The 7199 is also notorious for picking up heater hum. In your version of the Gemini (the schematic you posted), the heater center tap is grounded through a cap; later versions used a hum balance pot, which suggests that Ampeg found resistive balancing to work better. Better still might be to DC-elevate that heater reference; you could start with the output tube cathodes and see if that helps.

On the off chance there's a leaky coupling cap somewhere, you might check DC at Pin 7 of the 7199 (pull the 7199 first, because the grid will have a little DC on it in normal operation).

I found no leakage if any coupling caps. I tried a few other 7199's and no differenc ei nnoise.

Finally, I looked at the grounding scherme. Yuch, not the best.

I moved the ground of the cathode bias resistor. That helped. I also moved the ground of the preamp power supply filter caps to a location near the input jacks. That helped too (especially in getting rid of harmonics of 60Hz seen at the output).

Moving the center tap of the PT's heater output to the output tube(s) cathodes (+12V) made no difference in noise.

It's quiter now, but still not as quiet as some of th eamps I have built.

I tried moving the OT around and it made no difference in output hum

The question is: Why would shorting-out one half of the OTprimary (with an ammeter, to measure the plate current) reduce the noise to near zero?

[Dr-Joned]

Shorting out one half? What happens if you do the other half? That sounds like you might have a unbalanced transformer, meaning that it is partially shorted. If that is the case, it would make sense why the amp has lots of hum. The transformer is not able to correctly cancel out the 60hz.

[pula58]

Shorting out one half? What happens if you do the other half? That sounds like you might have a unbalanced transformer, meaning that it is partially shorted. If that is the case, it would make sense why the amp has lots of hum. The transformer is not able to correctly cancel out the 60hz.

Shorting-out either side of the OT primary (with the Ammeter) at a time quiets the amp down a lot.

The OT seems to work fine though: If I apply a sinusoidal input to the amp (@1kHz)), the output is a nice sine wave with equal pos and neg arcs. 10% THD at 24.5W output into 8ohms.

[Firestorm]

The question is: Why would shorting-out one half of the OTprimary (with an ammeter, to measure the plate current) reduce the noise to near zero?

Shunting either end of the transformer very nearly shuts down the whole thing because the two ends of the transformer are not separate from each other. Although they are in series electrically, they are magnetically parallel (since they couple inductively to a single secondary winding). So when you shunt current around one end, it affects the other end, too. That, unfortunately, is the limit of my E&M understanding of such things.

[Davejurasic]

I too have been working on an old Gemini I amp circa 1966 that had a persistent hum even after reworking the power supply. The grounding scheme of this amp is rather haphazard and it became evident that the offending components are the input jacks grounding to the chassis. I replaced the input jacks with Switchcraft long bushing jacks, used isolation washers to prevent the sleeve from grounding to the chassis, then soldered a wire from the jacks’ sleeve terminal to a ground point on the circuit board. The amp’s hum completely disappeared. There are other types of jacks that have a nylon bushing that are designed to isolate the sleeve from the amp chassis that work just as well.
And while inside the amp, one might as well redo the layout/lead dress of the AC heater supply because as is it is a mess. For this refer to the Heater/Filament Supplies discussion provided online by The Valve Wizard.

[slembones]

The input jacks are where the ground is supposed to originate. the ground loop is actually caused by a white and green wire that comes from the output transformer that is attached to the chassis of the transformer. it was sometimes connected to the filter cap ground which causes the ground loop. Ampeg traditionally in those days only used one connection to chassis ground (input jacks).

[Stevem]

I don't have time this morning to read thru this whole string, but if you pull out the power tubes of any amp and still have a light hum then your first filter node cap(s) are bad.

I see this all the time in Fender amps like the pro jr, hot rod deluxe and hot rod deville .

[Bev Wood]

Just a quicky- had a similar thing with a Selmar combo. persisting hum, after renewing caps, output valves and rectifier. There had been a nasty fault which I won`t go into - so it was good to replace these anyway. Everything seemed to be in order. However, the earth busbar struck me as possibly carrying enough current to affect the early stages. I cut the busbar and gave the pre-amp stages their own - grounding it at the same point as the existing one. Hum cured. I was rather pleased when the owner said,"Oh yes - it has always hummed a bit since I bought it new"!