oxbow_lake wrote: ↑Fri Jun 20, 2025 4:34 pm
[...] I want to push the 12L6s past their maximum specs, which are 200V plate and 125V respectively.
[...]Because these tubes do not have 6.3V heaters, I do not want to use a power transformer with both HT and heater windings. The most economical option seems to be a 115:230CT PT at ~$20 and a separate 12V filament transformer (~$15).
However, with SS rectification this is going to give me about 310V HT.
If you're open to more than one transformer, use three: a 120:120 @ 100ma, a 120:12 @ 60-100ma for the heater, and a 120:34v@ 100ma for boosting. The 120V output and 34V output can be stacked to get 150Vac, which full-wave-bridge rectifies with SS to 154*1.414 = 216Vdc, minus ripple sag and/or transformer winding losses. Transformers in this power range are available in PCB mount, so you can put all three on a single hand wired or printed board. They cost in the range of $10-$15 in this range, so the cost can be similar, and you get the transformer voltages you want.
These alternatives typically don't give you the visual of the bell-enclosed transformer above the chassis, though.
You can make your screen voltage from 200V with a resistive divider and a MOSFET follower for the screens. With a resistive divider the screens will not be hard-regulated and will sag with the main B+. The12L6 data sheet says it wants 6ma on the screens, so a MOSFET dropping 100V at 6ma is dissipating 0.6W; very do-able.
I thought about using a 115:115V isolation transformer with a voltage doubler, which in my experience would produce an HT closer to 270V, but Merlin says using a doubler to provide disparate HT voltages is a bad idea because it will cause DC in the transformer.
That's a common misconception. But transformers simply can not move DC across the windings, so whatever comes out of the secondary is AC. To a first approximation, you can't saturate a transformer from the secondary.
To a second approximation, you can cause a mild DC offset on the primary because of the I-R drop in the transformer primary windings with a half-wave circuit. It is very unusual for this to cause any issues on a non-toroidal core. It's possible to contrive a setup to do it, but it's ignored in general in the industry. The world is (or was, before switching power supplies) full of low power designs with half wave rectifiers, which is what a half-wave doubler is. Then there are full-wave doubler circuits, which remove the possibility for the second-approximation issue.
Using a dropping resistor for 50V right off the bad is less than optimal.
Yeah, it's not perfect, but then optimal is a special wound transformer, so you're kind of past that already. But you do have lots of options to make it work.
"It's not what we don't know that gets us in trouble. It's what we know for sure that just ain't so"
Mark Twain
I was thinking something like 230-260V HT, split in half for the screens. 
