DC Filaments/Heaters, help/advice...

[HTH]

I've installed an extra tube on a 2203 Marshall and it hums a bit more than I'd like, so I'm going to wire up the heaters for DC.

I've never done this before and am lifting the circuit straight from a Marshall TSL schematic.

It's just a bridge rectifier with a parallell pair of 330uF caps accross the DC side plus another series pair of 330uF caps accross the DC side with the junction of these two caps grounded.

Does that all sound ok? - seems pretty straightforward.

Where (ideally) would you locate the ground for the junction of the two 330uF caps?

Many Thanks as always.

[dai h.]

maybe I'm misremembering but I thought you weren't supposed to use a bridge with the center tap (heater)? There is a mesa schematic showing an added DC filament supply to V1 which consists of a bridge rect., 3300uF10V cap and two 39 ohm Rs. The 39 ohm Rs go from +6V and 0 to ground (respectively). There other heaters are referenced to ground through 100 ohm Rs. I think the 100 ohm R grounds would go to the noisy end as well as the 39 ohm Rs (Randall Aiken was saying noise couples through the filament winding because it's next to the other windings inside the PT IIRC.). Also a DC filament supply for a 12AX7 would be 300mA and since this is high it should be grouped with the noisy grounds like the B+ center tap, power tube cathodes, spk. return. I wonder if you really need it though.

[HTH]

maybe I'm misremembering but I thought you weren't supposed to use a bridge with the center tap (heater)? There is a mesa schematic showing an added DC filament supply to V1 which consists of a bridge rect., 3300uF10V cap and two 39 ohm Rs. The 39 ohm Rs go from +6V and 0 to ground (respectively). There other heaters are referenced to ground through 100 ohm Rs. I think the 100 ohm R grounds would go to the noisy end as well as the 39 ohm Rs (Randall Aiken was saying noise couples through the filament winding because it's next to the other windings inside the PT IIRC.). Also a DC filament supply for a 12AX7 would be 300mA and since this is high it should be grouped with the noisy grounds like the B+ center tap, power tube cathodes, spk. return. I wonder if you really need it though.

good point Dai, however (and I know I didn't mention this), the amp is a Zack Wylde 2203 - on the reissue type 2203s Marshall reference the heaters to ground through a pair of 100ohm resistors and a mini-pot on the PCB.

'GN11' on the PCB is the heater winding CT - it's just connected there and goes nowhere.

I might not need the DC heaters, you're right, but I do have a little more hum than I'd like so I want to try this out for comparision sake.

If not, I'll go back to AC heaters and elevate them above ground (around 30v). That killed ALL hum in a previous build.

[dai h.]

I haven't tried the adding a gain stage thing in a long time, but off the top of my head (thinking about how things are done in Soldano SLO100, CAE/Suhr modded high gain amps), I guess I would pay attention to the relative orientation of the heater wires vs. the other wires (try to keep at right angles), grounding, try to keep loops compact to help keep it more isolated(trying to reduce capacitive, inductive coupling to help maintain stability and keep noise low)--this would include use of shielded wires, grid stopper directly on grid, selected tube for input stage, maybe grounding elsewhere in the amp if it looks like it can be improved, etc.

also, i might give it its own filter

[Tubis]

The grounding in those old 2203 is very poor. I would do it properly before anything else.

[HTH]

I haven't tried the adding a gain stage thing in a long time, but off the top of my head (thinking about how things are done in Soldano SLO100, CAE/Suhr modded high gain amps), I guess I would pay attention to the relative orientation of the heater wires vs. the other wires (try to keep at right angles), grounding, try to keep loops compact to help keep it more isolated(trying to reduce capacitive, inductive coupling to help maintain stability and keep noise low)--this would include use of shielded wires, grid stopper directly on grid, selected tube for input stage, maybe grounding elsewhere in the amp if it looks like it can be improved, etc.

also, i might give it its own filter

I've added an extra 'node' on the HT line for this additional gain stage. I've just tapped off the same point where the existing V1 node comes from, then added a 10k resistor and 22uF cap. I figure this will keep things tighter and stop this extra stage from sagging down the others.

I've got shielded wire on all the grid runs and grid stoppers are direct on the tube bases.

I haven't got a tube specifically selected for the input stage, might try that as Watford Valves do offer that service.

Cheers.