6AH4 preamp

[TubeDriver]

I'll second Roscoe's idea, experiment with layout. My 1626 preamp with a LCLC (also DC on filaments) is very quiet with my horns, it would be dead quiet on your Proacs.

[tomp]

Nothing works better to deal with radiated magnetic fields than distance. With sensitive circuits I often put the transformer in a separate enclosure.

[Pelliott321]

I only see two chokes in the power supply. Maybe just an early in the build picture
The second cap in the pi can be larger, maybe 100uf.
Choke input filter would require a higher voltage B+

[Roscoe Primrose]

FWIW, I vaguely remember Roscoe fighting to get the ripple eliminated on one of his preamps, with 5-6 sections of C-L-C filtering....it never got to inaudible IIRC.

That was a filament supply for 26s….

That said, I’ve built several tube line stages with just a CLC supply that were more than quiet enough, so I’m not convinced it’s a power supply problem…. Grover, can you put a ‘scope on the power supply and see what you actually have as far as ripple?

Roscoe

[Grover Gardner]

I only see two chokes in the power supply. Maybe just an early in the build picture
The second cap in the pi can be larger, maybe 100uf.
Choke input filter would require a higher voltage B+

Yes, I added the 15H choke later.

Thanks for the suggestions, everyone. No change in the layout offers relief. And the more filtering I add the worse it sounds. Choke input would require a different power tranny, I might try that. OTOH I feel like a light-weight reg'd supply might be a better answer. I might have something here I could try.

[Grover Gardner]

FWIW, I vaguely remember Roscoe fighting to get the ripple eliminated on one of his preamps, with 5-6 sections of C-L-C filtering....it never got to inaudible IIRC.

That was a filament supply for 26s….

That said, I’ve built several tube line stages with just a CLC supply that were more than quiet enough, so I’m not convinced it’s a power supply problem…. Grover, can you put a ‘scope on the power supply and see what you actually have as far as ripple?

Roscoe

I could, if I knew how. I'd need a cap to the probe, right? Some on Audiokarma modelled the original CLCLC in PSUD and got 3mV ripple.

[tomp]

Most scopes have the ability internally to select a DC , AC, or ground mode.

[Roscoe Primrose]

OTOH I feel like a light-weight reg'd supply might be a better answer. I might have something here I could try.

Gas tubes are easy

Roscoe

[Grover Gardner]

Most scopes have the ability internally to select a DC , AC, or ground mode.

This is a USB scope. I can look it up.

[SoundMods]

I hate to bring up the obvious -- but it is an open bread-board without shielding.

Contained within a steel or aluminum enclosure (aka. Faraday Shield) the circuit is protected from airborne issues like hum and RFI/EMI.