[morzh]

Keep combating the AC hum.

1. The hum stops if the 75 tube is extracted.
2. The hum stops if the lower end of the volume regulator is disconnected.
3. Increasing capacitance of the cap decoupling the Chassis (10uF) to the negative reduces the hum, but does not get rid of it.
4. all GND rivets involved are checked and are fine.

I tried to complete the factory mod that was reflected in the "Changes" doc (discussed here on p.2) as it also requires disconnecting the lower end of the volume pot from C- (the midpoint between 260 and 70 Ohm resistors, connecting Negative to GND) and connecting it directly to GND.
This only made matters worse. The hum becomes just too loud.

I think I am dealing with a case of GND looping. Even though most GNDs come via rivets from the chassis locally and the rivets are fine and the chassis is low-inductance, low resistance path, somehow I think, coupled with long capacitors' leads (wires go all over the board, reaching good 5-6" in length) this will create some humming.

I think I will try to re-route the sensitive GND connection using wires to a single point where the capacitor connects to it.

[Mondial]

Yes it definitely seems there is an audio ground loop associated with the low end of the volume control. Probably the biggest problem is that there may be a couple of amps of AC filament current flowing through the chassis which is common grounded to the low level audio.

Are there two separate wires to the filament pins of each tube, or only a single wire and chassis ground to the remaining pin? On the more expensive sets, they ran twisted pairs of wire to the filament pins and grounded the center tap of the filament winding, so there was no filament current through the chassis.

You might try experimenting with where to ground the low end of the volume pot. Try to find a point on the chassis with minimum induced AC hum.

[morzh]

Yes there are chassis rivets that supply filaments. Ad as a matter of fact the connection point of the C- (two resistor) is nearby one of them.

I just wonder if this is a common 38/39 small radios' problem or am I the only one that lucky? I mean, if they had that noticeable a problem with the hum, they would correct it, wouldn't they?

[morzh]

Well, after spending another 3hrs with it...

It is in a way a GND loop, but little of it. I rerouted everything to the star GND-ing scheme and the result was just a bit better than before....and even that might've been some of wishful thinking.

I could swear that it started after I put it in the box and I started recalling events.
The recollection resulted in ""You used the variac and nnowe you just use the isolation transformer".
Once I plugged it in the Variac the hum went down about an order of magnitude. That is it is still there but it is quite bearable.

Than I measured the voltage in the Mains, 125V AC, and set the Variac to it. The hum immediately went up.
It goes up almost by a step about 117-120V AC. Bring it back under 115V and it is reasonably quiet.

Mistery solved for practical purposes.
If I were bold enough I'd rerouted the whole GND scheme, but...not worth it.

PS. Probably need a bucking transformer, eventually.