Shielding a plate choke?

[mbrennwa]

I used some left over parts to build a nice tube amp. The input stage of this amp is a 6SN7 with a plate choke load. It works ok, but the choke picks up a bit of hum (most likely from the PSU transformer). My spectrum analyser shows a number of peaks at 50Hz and harmonics. Before I replace the choke with something else, I'd like to see how far I can take the choke design. Is it possible to shield the choke? Other suggestions on how to avoid the 50 Hz hum (and harmonics)?

 

[rayma]

the choke picks up a bit of hum

Try changing its placement first, maybe you can null out the pickup. Also twist the
choke's leads, if possible. Of course, other AC leads in the amp should also be twisted.

 

[trobbins]

Quite a few people have been able to reduce signal transfer between magnetic devices by changing orientation, as rayma suggests. If that is not easy to try, then you may be able to manually insert a plate of steel between the two devices. The thicker the plate the better, but also the more that the plate can divide or surround one or other devices, the better for assessment testing.

Do the transformer and choke have bell-ends, and is the chassis steel?

For some background, its worthwhile appreciating how the magnetic field splays out from a winding, and how that can more easily couple in to a similarly oriented winding just nearby - and then what would happen if you turned one winding 90 degrees vertically or horizontally.

 

[GoatGuy]

My first thought was, are we sure the choke load is picking up the hum + harmonics? or is it a case where the power supply itself is "hummy"?

One easy way to tell is to just use a "convenient resistor" instead of the choke. You know, something that drops B+ by 50 volts or so. If OP is using a cathode-lifting bias resistor, that's easy enough:

V_K = I_K • R_K …
I_K = V_K / R_K … so therefore if
50 = I_K (≡ I_A) • R_A then
R_A = 50 / I_K

Which in the schematic-less world might be something like

V_K = 4 volts
R_K = 820 Ω …
I_K = 4 ÷ 820 = 4.88 mA

R_A = 50 ÷ 4.88 ma = 50,000 ÷ 4.88
R_A = 10,245 Ω
R_A ≈ 10 kΩ

So … with those made up, but not exactly irresponsible numbers, I'd drop in a 10 kΩ resistor. Wattage? P = I E = 4.88 ma × 50 volts = 244 mW. A 1 watt resistor would work just fine. Or a couple of half watters in parallel. Or series. Whatever.

Just removing 1 lead of the plate load choke and soldering the test resistor in place then tests rather trivially the idea of whether the load choke is picking up hum magnetically, or whether the power supply is a hummingbird.

And the test can be done in what, 1 minute or less?
Just saying…
1 minute tests are worth a lot.

GoatGuy

 

[mbrennwa]

Try changing its placement first, maybe you can null out the pickup.

Moving the choke is tricky, because there's not much space. But I'll try.

...you may be able to manually insert a plate of steel between the two devices.

Do the transformer and choke have bell-ends, and is the chassis steel?

I couldn't find a steel plate, so I used some copper foil. That didn't do much.

The mains transformer is a toroidal. The choke does have bell ends. Chassis is aluminium base plate with wooden frame.

My first thought was, are we sure the choke load is picking up the hum + harmonics? or is it a case where the power supply itself is "hummy"?

One easy way to tell is to just use a "convenient resistor" instead of the choke.

Sure. That's what I did to identify the choke as the part where the hum is picked up.

 

[rayma]

Moving the choke is tricky, because there's not much space.

Any photos of the chassis?

 

[Trajan]

I second that you may have B+ hum and need more inductance. Sometimes it isn't too hard to splice in another choke in series to cure B+ hum if you have the room and can find a stealthy choke.

 

[trobbins]

I second that you may have B+ hum and need more inductance. Sometimes it isn't too hard to splice in another choke in series to cure B+ hum if you have the room and can find a stealthy choke.

plate choke load - not B+ filter choke

choke as the part where the hum is picked up.