Artificial Center Tap

[zendragon63]

Hi gang--I got a filament artificial center tap question that I can't find an answer for. I just finished a subminature tube build inspired by the Lil Devil build by UR12 from another amp forum. Fired right up and sounds every bit the part--except that it had a little higher 120 cycle background noise than I  wanted to accept. Thought it was a ground loop and reworked the grounds, then checked the power supply caps and several other components. No change. Until I looked closer at the 2x 100 ohm filament artificial center tap. I subbed in the humdinger pot thing and could dial away most of it BUT it essentially--and happily--disappeared with a single 100 ohm on just one of the legs to ground. Still 6.2 VAC on all the filaments... I just have not seen only a single resistor so do you think this is going to be ok on the long term?


Kinda a cool little build and great tone even if I dial it down enough to keep the household and my ears happy. Thanks in advance for any insights on the filament center tap issue. Regards.


dennis

[sluckey]

Having only one side of the filament winding referenced to ground through a 100Ω resistor is equivalent to tying one side of the filament winding directly to ground. I'm surprised that reduced more hum than the humdinger pot, but if you like it then that's what counts. There's nothing dangerous about doing that. It would be interesting to measure the ac hum voltage at the speaker with your single resistor and then again with the humdinger pot set for minimum ac voltage at the speaker.

BTW, ac filament hum is 60HZ, not 120hz.

[zendragon63]

Thanks Steve. Yeah, that low 60 cycle hum is what I would expect out of the filament line but this sounded like a mild ground loop of sorts. Like the input plug wasn't shorted to ground (and why I spend a few days chasing it). It was unusual to me because I just hadn't experienced this situation. But the result is that it is acceptably quiet now. Thanks and regards.


dennis

[zendragon63]

Steve's reply made me do some thinking about this and came to the conclusion that this noise--the 120 cycle noise--may not be directly related to the artificial center tap. So I am flummoxed and request some help on this one. 


I have built many amps and they are all--save the typical hiss at full volume-- pretty much dead quiet. The noise which I am combating sounds like a ground loop of sorts (though I could be wrong). I have the understanding that 60 cycle hum is filament related and 120 cycle noise is power supply related. This is what I have done so far:

• Humdinger pot. I can dial out virtually all of the hum and a significant amount of the 120 cycle hash toward one extreme of the pot.
• Rearranged the grounds along the buss bar--no change in noise.
• I have elevated the artificial center tap to the +16V of the power tube cathode. No change in noise.
• Reflowed all even slightly suspicious solder joints. 
• The first volume controls the noise level entirely.
• I have replaced the V1 tube with a second NOS 6112. No change in noise level.
• Replaced the diode rectifiers and all filter caps in the power supply and preamp--one by one--and no change in noise level.
• If I ground V1b grid and subsequent tube grids (V2 and V3 PI--dead quiet with the expected hiss just a hint of 60 cycle hum at full volume.

The noise seems to be right in the V1a. I don't think I have any higher voltage lines near the input lines; I have chopsticked and moved around the plate feeds to V1a with no change in noise levels.


Other than the noise, the amp sounds great--it is just that the noise is unacceptably high as you turn up the volume. It is likely something simple that I am looking beyond and is stupid simple. Many thanks in advance for your insights. Regards


dennis

[pdf64]

...that low 60 cycle hum is what I would expect out of the filament line but this sounded like a mild ground loop of sorts. Like the input plug wasn't shorted to ground ...

A 'non-shorted' input would pick up mains frequency EM signal, 60Hz.
Due to the use of 'ground' to cover mains ground and circuit 0V common connections, 'Ground loop' is a non specific term for frequency here, as it would be at mains frequency if between the loop was different items of mains ground connected equipment, or twice mains frequency if between different circuit 0V common connections of a single item of mains powered equipment with full wave rectified power supplies.

...
   

• Humdinger pot. I can dial out virtually all of the hum and a significant amount of the 120 cycle hash toward one extreme of the pot.

...

Hash, if used to indicate higher frequency components eg buzz rather than hum, often indicates that the noise source is the heaters. May be due to heater - cathode coupling of some sort within a (early stage) tube.

My guess is that there may be more than one source of hum and buzz in this amp, perhaps with some degree of cancellation between them, such that eliminating one may make the other more apparent.

[zendragon63]

My guess is that there may be more than one source of hum and buzz in this amp, perhaps with some degree of cancellation between them, such that eliminating one may make the other more apparent.

I agree.--and many thanks for the comment pfd64. It is just that finding the offending source of buzz is that is apparently riding on top of the 60 cycle is proving to be a formidable challenge. FWIW I took the VVR out of the circuit with no change and reversed the filament wires on V1 with no change. It occurred that the PT could be defective or possibly the OT is not positioned correctly to cancel out PT/OT coupling buzz--though I did the headphone trick for placement and I would think you would detect that buzz in stages subsequent to V1?


Any other thoughts or theories are humbly welcome. Regards


dennis


 

[zendragon63]

Thanks 8wattjack. The grounds have been moved around but to no avail. I don't get this one. I have replaced every preamp part, power supply cap, taken the tone stack out of the circuit and the replaced V1 3 times. What I do notice consistently is when I short the outside lug volume control to ground, its doesn't affect the buzz significantly (unless at max--then quiet) and when I short to ground the center (leading to V1b), it goes quiet, I have come to the conclusion that the interference is right at the V1b side of the tube. Both sides of the triode do share he same filament--that could internal filament phase be a factor? It has really turned out to be a great sounding little amp--except for that buzz...


Any other ideas anyone? Regards


dennis

[PRR]

> when I short to ground the center (leading to V1b), it goes quiet

How long is that wire? How many buzz-sources does it rub against?

[zendragon63]

Ok , because the humdinger got me so far and nothing else got me anywhere, I got convinced last night that the filament supply was causing the buzz. I did the cheap fix DC-on-the-first-tube thing that I lifted from the Merlin book. 6.2VDC on V1. Now quiet, real quiet and I hear a slight more 60 cycle when everything is dimed but WAY better and overall about as quiet as other builds.


It doesn't explain why I had to go to this length on this one--first time on the DC thing for me--but hey, we arrived.


Thanks to PRR, 8wattjack and pdf64 for your interest and responses and hope this situation can help others. Attached is the revision. Again, thanks. Regards


dennis