Correct hum mods?

[dobias]

I've been able to reduce the hum in my 2A3 Moondog amp to a minimum by:
-replacing the dc to the gain & driver tubes with ac & raised voltage.
-connecting the 2.5vac CTs with a 10uF cap to the humpot wiper.
The wiper has a 1K non-inductive resistor paralleled with a 47UF cap to ground.

Is this wiring OK or am I making a mistake?

I'm getting a stable 1.3mV of hum on both channels. When I short the meter leads it shows 0.2mV.... should that 0.2mV be deducted from the 1.3mV?

[gootee]

Yes the 0.2 mV gets deducted.

Do you have photos showing the wiring etc, and do you have a schematic?

[dobias]

gootee,
I built a custom version with the PS on one chassis & the two amp channels on the other chassis. I haven't gotten around to making a 'final' as-is schematic. The original schematic can be seen at: home.earthlink.net/~ivol/moondog_2A3. If that isn't sufficient, I'll knuckle down & draw one.
Thank you for responding.
Frank

[gootee]

Do you have photos of the wiring and layout of the version you're asking about?

[dobias]

Have you downloaded the original schematic at
"home.earthlink.net/~ivol/moondog_2a3" ?

Since I asked with this thread, I replaced the one turn humpot with a bypassed ten turn humpot & wound upwith a satisfying 0.6mV of hum. See the newer thread describing it.
I have a difficult time even seeing the humpot wiring through the dense wiring in the chassis. A photo wouldn't help.

[gootee]

Yes, I saw the schematic.

When you are measuring hum, is it with the inputs open, or inputs shorted, or with a source or resistor connected but with no signal? Have you tried all three ways? If not, can you?

I am assuming the hum is probably induced in conductor loops that have too much enclosed loop area, by picking up radiated electromagnetic fields, which are produced by other similar enclosed loop areas, or by components.

You typically try to stop both the reception and the radiation of the fields by keeping all natural pairs of conductors as close together as possible, everywhere, usually tightly twisted together if possible, to minimize their enclosed loop area, or by shielding (and not using the shield as a signal or signal return conductor and grounding it at one end only). Also, you try to keep the small-signal pairs away from everything else, etc.

By "natural pairs" I mean things like signal input/gnd pair, transformer primary pair, each transformer secondary pair, each heater pair, output pairs, any DC supply pairs, etc. Some are transmitters and some are receivers. How badly they do either is dependent on their enclosed loop area (see Faraday's Law, or Maxwell's Equations), and what types of currents they carry (amplitude and rate-of-change).

So when you see a single wire going across a layout, it's a potential source of hum because it's far from its return-path conductor, and is making a relatively large enclosed loop area with that return path.

In the layout on your web page, I saw that some pairs were twisted, but not tightly, and I saw some wires that maybe were not paired with their return-path conductors. So I wondered about your new layout's wiring. Do you have any hum in the unit pictured on your website?

[dobias]

gootee,
Thank you for the very concise explanation. In my latest rework of the wiring (in cramped quarters) I tried my best to use tightly twisted pairs of wires for anything over a few inches. The 6.3v heater & 2.5v filament transformers are in the amp chassis. At least the B+ is in a separate chassis.
The present 0.6mV hum results (with open inputs) are satisfying. If there are ways to improve it, I'll try. I'll also try shorted inputs & connected inputs today.
See my latest thread,"Hum Nirvana" for details so far.
Frank
Old too soon, smart too late.

[gootee]

With the inputs open, you can't tell if the input signal and signal return loops are having currents induced in them. Are your input grounds isolated from the chassis, and only connect to ground at the input resistors?

[dobias]

gootee,
Yes, the inputs are isolated from the chassis & have input resistors.
It doesn't make any difference to the hum level whether the inputs are open, grounded or have connections.
The right channel has risen from 0.6mV to 1.0mV since yesterday.
Frank

[gootee]

Originally Posted by dobias

gootee,
Yes, the inputs are isolated from the chassis & have input resistors.
It doesn't make any difference to the hum level whether the inputs are open, grounded or have connections.
The right channel has risen from 0.6mV to 1.0mV since yesterday.
Frank

Frank,

Sounds good! Input signal and ground are often problem areas. But yours must be well done.

Tom