Isolation Trans Hum

[drewl]

Hey all,
Got a Magnetone Starlet model 107, widow maker.

I installed an isolation trans, grounded power cord and fuse, but am getting what sounds like filament hum.

I tried isolating the end of the filament string from the rest of the circuit, but it needs to be connected or grounded for the 12AX7 to draw heater current.

Is there a way to isolate the filament from the rest of the amp circuit?

[Jon Snell]

It is an AC/DC amplifier. Connect the ground to the -ve side of the 50u capacitor and earth the switched point where the 117v comes in, (see added line). As you have isolated it, that will be OK, if you mean mains hum. Or feed it with 117v DC from your isolation transformer.

[Helmholtz]

I assume there was no hum without the isolation transformer?

If you mounted the iso on the chassis you might have some hum crosstalk (magnetic field coupling) between the iso and the OT. I would try different mounting positions/orientations of the iso.

[drewl]

I'm not following, connection the "circuit" grounds to the negative end of the filter caps, the AC low to the chassis?

[drewl]

Because if I don't ground the negative side of the filter caps to the end of the filament string, I don't get high voltage DC.

[J M Fahey]

I´d simply short across the 270k//.05uF "life saving" network to bring preamp /jacks/power amp/main supply griounds together, since it´s not longer needed.

Which jon snell is also saying, using different words.

As of hum, all widow makers do

[Enzo]

Not only is it a single ended amp running off the first node - which hum by their nature, it also is a half wave rectified power supply, feeding the power tube. SO that is a double whammy of hum.

Regarding single ended amps, Fender once sent out a bulletin to warranty centers telling us "DO not try to repair the hum in Champ amps, it is normal".

[drewl]

Great, thanks guys.

Just making sure it was something stupid I did.

Have another amp that needs an iso.

I need to study some transformer theory for another project I'll be bugging you guys about.
They sure didn't teach it by the time I was in school.

[J M Fahey]

I need to study some transformer theory for another project I'll be bugging you guys about.
They sure didn't teach it by the time I was in school.

Please do, everything adds up and helps, but here your problem was basically grounding.

Since not having a power transformer meant the input jack ground is connected *straight* to guitar strings, so straight at your chest and you had 50% of plugging mains plug the wrong way (and even if it were 1% ), they "minimized" damage by limiting current to a less than deadly value.
Courtesy of a 150k to 270k resistor, similar to what you see inside a neon screwdriver which also sends current through your chest

Problem is it destroys your grounding
They sort of compensate, by paralleling it with a .05uF cap (which becomes a Death Cap and must be treated as such, if it shorts you're dead) so at most Audio frequencies grounding is somewhat improved ... but not at hum frequencies, it´s either one or the other.

I suggested shorting that protection network ONLY because now you will use an isolation transformer.

[nickb]

Great, thanks guys.

Just making sure it was something stupid I did.

Have another amp that needs an iso.

I need to study some transformer theory for another project I'll be bugging you guys about.
They sure didn't teach it by the time I was in school.

There are another couple of things I'd suggest. First, you really should add a slow-blow fuse, suggest 500mA, in series with the supply to the amp. Second, a trick to lower the hum due to power supply ripple, is to convert the supply to from 60Hz to 120Hz by adding a 1A 1000V bridge rectifier ( you could use 4 x 1N4007 rectifiers) after the iso transformer.

[Helmholtz]

Second, a trick to lower the hum due to power supply ripple, is to convert the supply to from 60Hz to 120Hz by adding a 1A 1000V bridge rectifier ( you could use 4 x 1N4007 rectifiers) after the iso transformer.

Very good proposal. This will also relieve the isolation transformer from the asymmetrical DC load with half-wave rectification, which increases transformer losses, mechanical vibration and radiated hum.
As B+ would increase somewhat, I might also consider adding another supply node (maybe using a choke) for improved filtering - if required.

[drewl]

What, get rid of the rectifier tube or feed it DC?

[Helmholtz]

What, get rid of the rectifier tube or feed it DC?

I meant getting rid of the rectifier tube. There is no sound benefit from a tube rectifier in a single ended class A amplifier.

[nickb]

What, get rid of the rectifier tube or feed it DC?

Nowhere did i say to remove it.

You should to leave it in place as It is a required part of the heater supply. If you remove it the HT will rise significantly and you need to change the heater dropper resistor to compensate for missing rectifier. It just turns a quick easy tweak into a major job if you do.

[Helmholtz]

Without the 35Z5 rectifier tube the heater series resistor should be increased to 390R/25W wirewound.
The increased DCV output with a diode bridge can be used to improve hum filtering by an additional supply node as mentioned above.