60hz hummmm Signal Corps AM-186

[Dmichel123]

I've posted on my recent hum problem in an old "drawing board" thread on the same unit, but I think the current discussion fits here.

We received the original PT back from a rewind and got the limiter going with it's original internal PSU, rather than the bench supply we'd been using with it.

The unit now has considerably more hum than with the bench supply. It's 60hz, so I first looked at the heater supply. I tried elevating the heaters, putting in a hum balance pot, and even powering the heaters from the bench supply. The hum seems pretty consistent in all configurations.

Shorting the input transformer primary leads kills most of the hum. It seems the PT is inducing a great deal of hum either mechanically, magnetically, or into the chassis ground.

Should I send it back to the rewinder? This was a very expensive rewind! 60hz hum is at -48dB, and that's with 15dB of attenuation at the output!

 

[Dmichel123]

Hooked the limiter back to the bench supply. Very low hum. Simply applying power to the rewound transformer, with it bolted to the chassis but not connected to the circuit causes all of the hum to appear.  I'm going to contact the rewinder to see if he has a solution.

 

[emrr]

These appear to be a 3 chassis limiter, like the 1126.  Can you separate the PSU chassis from the rest and test that way, and with greater distance? 

 

[Dmichel123]

I think the chassis will unbolt from the other two. I would definitely have to make a cable to carry all the connections up to the amp chassis. You'd think a $400 rewind of a transformer for an audio amplifier would produce a transformer that doesn't require me to modify the vintage unit.

 

[radardoug]

The rewinder can only rewind with what you supply. The unit as designed may have had a different chassis for the supply. You already know separating the supply gives you low hum. So dont dis the winder, he did you a service within the limits you set.

 

[emrr]

I think the chassis will unbolt from the other two.

Maybe if it's isolated in terms of chassis ground, without moving it any distance at all.

 

[Dmichel123]

The rewinder can only rewind with what you supply. The unit as designed may have had a different chassis for the supply. You already know separating the supply gives you low hum. So dont dis the winder, he did you a service within the limits you set.

The unit is still in the original chassis as designed. Yes, I knew hooking the limiter up to my bench supply gave low hum. We used it like that for 9 months or so before opting to have the original PT rewound and use the factory PSU.

 

[Dmichel123]

Maybe if it's isolated in terms of chassis ground, without moving it any distance at all.

Everything but the transformer case was isolated from chassis ground when I tested with the bench supply. We will have to use it on the bench supply for sessions the next 4-5 days before I get a chance to try isolating it further.

 

[emrr]

Everything but the transformer case was isolated from chassis ground when I tested with the bench supply.

I mean PSU chassis from the control and amp chassis.

Is the rewind in the original transformer case shield?

 

[emrr]

front

 

[emrr]

schem

 

[emrr]

60hz hum is at -48dB, and that's with 15dB of attenuation at the output!

I see the 1126 data sheet claims

Output noise of -45dBm unweighted.  At point where compression starts output level is approx +24dBm resulting in an operating S/N ratio of at least 69dB. 

53dB total gain 600 to 600 ohms, attenuators at zero attenuation.

Besides the transformer contribution, add anything else that might not be optimized to that noise figure; loose connections in the ground plane, connections that shouldn't be there, proximity of other stuff radiating fields, etc.    The Signal Corp version may have a different noise spec too, assume it's not known from a document. 

 

[Dmichel123]

Been a while since I've had a chance to revisit this thing. The -48dB hum figure I gave is dBFS, so -26dBu? It's loud hum. Definitely unusable humming like that. I am fairly confident that the noise isn't coming from anything else. It only happens when powering the limiter from the power trans. ...or with power applied to the rewound transformer primary with no other taps connected to anything and the limiter powered from the bench supply. I'm also fairly confident that this thing didn't hum like this when built. Why go through all the trouble of a regulated supply only to use a power transformer that radiates a ton of hum into the input transformer that's 18 inches away?

We did send the transformer back to the rewinder, who had connected an internal screen differently than what he saw when rewinding. He said he had never seen a screen connected how he found it and was certain that it must have been a mistake. ??? When we sent it back, he had added a terminal to the case for that screen. I tried connecting that screen to a few different ground points with absolutely zero change in the amount of noise. I have no idea what the difference is between what he found, what he originally did, and adding the tap to the case.

Today, I was poking around in the PSU and found that the "Y" filament winding is measuring weird. I measure ~1.1 ohms end to end and ~4 ohms from either end to the center tap. That can't be right can it? I would expect half of the full winding resistance to the center tap. 

Anything jump out at y'all as to what the problem may be?

 

[Dmichel123]

I isolated the PSU chassis from the rest. It still hums, and there is 104VAC and 94VDC potential between that chassis and ground.  :

 

[scott2000]

Today, I was poking around in the PSU and found that the "Y" filament winding is measuring weird. I measure ~1.1 ohms end to end and ~4 ohms from either end to the center tap. That can't be right can it? I would expect half of the full winding resistance to the center tap. 

is that out of circuit?


There are certainly a lot of shield designations around.....

also kinda goofy how they swapped numbers where E18 connects to E12...

 

[CJ]

did you try swapping the primary leads?

did the winder rewind the xfmr exactly like it was built?

there are tricks used for winding power transformers, probably to keep flux contained better,

ask the rewinder to send you the print he used,

if you sift thru these transformers you will notice some different techniques,

https://vacuumbrain.uk/power-transformers

 

[Dmichel123]

I think he rewound it like it was built... There was one coil that was almost completely burnt, probably this one? I just got an email back from him that he wound a spare coil for it and he will measure it tomorrow to see if it measures like this one. All the windings are putting out the proper voltages. I am measuring the filament taps with the tubes pulled, which I think is effectively out of circuit? I will try swapping the primary leads today.

 

[scott2000]

I am measuring the filament taps with the tubes pulled, which I think is effectively out of circuit?

The y looked to be elevated from memory so I'm not sure......

 

[Dmichel123]

The battery in my DMM was weak, but after changing it and with the weird winding completely disconnected from everything I'm still getting odd readings: 0.6 ohms end to end and 1.3 ohms either end to center tap.

 

[emrr]

The battery in my DMM was weak, but after changing it and with the weird winding completely disconnected from everything I'm still getting odd readings: 0.6 ohms end to end and 1.3 ohms either end to center tap.

That would suggest it's not laid out as one would assume.  Don't recall if you are measuring voltages that make sense there.