I have a theorie on this.martin manning wrote: ↑Mon Apr 26, 2021 10:13 pm What doesn't quite make sense to me is 550V plates. To get down to the advertised output power of 150W would require 100V of sag, and then the screens would really like to drop another 100V, down to 350 or so.
In the schematic drawn by Talbany after the rectifier there are two 100uF caps in series. Then next comes a standby switch and behind that another pair of 100uF caps in series. I don't think this is correct. Using the standby switch will ware out these capacitors fast. Using the switch is much like shorting the first two caps over the ESR of the second two caps in series, and the two caps behind the standby get hit by a 550V DC supply with low internal resistance created by the ESR of the first two caps in series. There is a short but very high loading current. So to me it would seem best not to use the standby switch in this given schematic at all.
Placing a choke after the standby switch will fix that problem because a rapid change will result in a high impedance of the choke, since the load curve of a choke is opposite to the load curve of a capacitor. So using the standby switch now, the caps will no longer get 'shorted' and have to deal with a large current spike.
Looking at the schematic of the Sunn 1200s, which is rated at 120 watts with tube rectifiers, there are also two chokes in parallel after the standby switch. When switching to a solid state rectifier, the chokes are still there. Both chokes are mounted inside the chassis, and are pretty small. There is a 40 volt voltage drop over them leaving 500V on the plates.
I think the Dumbleland 150 would have a similar powersupply with bigger filtercaps than the Sunn amps and one 400 or maybe 500mA choke with something like 100 or 120 ohm DC resistance.