Humdinger with center-tapped heater winding

[Ken Moon]

Sorry if this is a repeat, but I couldn't find the answer to this particular nuance on humdingers:

I have a TW-style amp, with a center-tapped heater winding. The center tap is connected to a voltage divider (tapped off the screen B+), giving 70Vdc of elevation (250k/47k resistors, with 100uF/150V cap across bottom (47k) resistor).

Even with this setup, I still get some heater hum amplified by the preamp.

If I want to try a humdinger, should I disconnect the heater center tap entirely, connect the two outside lugs of the humdinger pot across the heater winding, and connect the wiper to the 70V elevation point?

I think that's right, because the humdinger pot is creating a virtual center tap through the wiper, and adjusting the pot balances the 2 (virtual) halves of the winding. But I figured it was better to ask before I create any smoke and fire

[Badside]

I believe DC elevation sorts of nullify the need to balance both sides. I could be wrong, but it seems logical to me.

Did you try the lamp battery trick to make sure the hum is coming from the filaments?

[Ken Moon]

Thanks for the reply.

The elevation helps, but I don't think it's directly related to heater winding balance.

Putting in the humdinger is sort of a brute-force, see-if-it-works proposition, but it's worth a try.

A separate 6.3V filament tranny helped, as did a regulated and filtered DC heater supply using an LM7806 and a 12V filament tranny.

But I'm going for low parts count and simple here

[Firestorm]

Do you know what stage of the amp the hum is coming from? For example, does it go away if you pull the PI tube? V1? V2?

[Ken Moon]

It's the typical TW hum - goes away completely if PI input grounded.

The third gain stage is so sensitive on every express-type amp I've built that you just have to think about touching any of the wires/tube sockets/etc in this area and you'll hear it on the output

I haven't reached humvarna (or hisshalla) yet, but the journey goes on, and probably will, until I reach those pearly noise gates

[Gaz]

Mr. Moon,

Here's what I found over at humdinger central (AKA AX84):

The simple solution is not to use the centre tap - just tie it out of the way.

If you feel that you really MUST use the CT, connect the variable resistor across the heaters as you have shown and then use a capacitor to connect the wiper to the CT / elevation point. You will need to get the value of the cap right so that the basic 50 / 60HZ waveform is blocked by the cap, but the buzzy spikes are passed.

Personally, I would just ignore the CT.


I thought your question was familiar... and that was because I asked the same one

A humdinger miraculously fixed a 'buzz' in an amp of mine. The 3rd stage of an October, actually. I replaced every part of that stage's circuity (and much more) before trying the humdinger. Can definitely be a 'buzzdinger' as well. It worked for another guy who was having a nightmare too with a HO (I think). In his case and mine, DC heater via a lantern battery fixed the noise too, which is what initially lead me to thinking the filaments were the culprit.

[Ken Moon]

Thanks, Gaz.

Yeah, the lantern battery helps immensely on mine, too - so I'm hopeful.

I went ahead and got some of those pot grounding lugs, and I'm going to experiment a bit with grounds and shields around the tone stack area while I'm in there.

[Gaz]

Through that same debugging process I ended up shielding almost every signal wire as well, connecting the shield to chassis ground (rather than the ground buss) per Merlin B's suggestion in his new(ish) grounding article. It helped as well, and it was easier to tell which buzz/hum was which after getting the humdinger installed.