Adjustable B+ Bench supply for tube amp experiments

I put together this hi voltage bench supply from mostly parts I had hanging around. It turns out to be very useful for other purposes like safely/gradually powering up a direct coupled amplifier or reforming large power supply caps. The only purchased parts were the variac and some high voltage banana jacks.
I think it came out pretty nice. Kinda looks steampunk on my bench. With that in mind I plan to add a DPDT knife switch on the front panel to control whether the high V is turned on in one position or the filter caps get quickly discharged in the other. To make this touch safe, (exposed conductors) I'll need a relay, perhaps solid state.
It has:
1. A high Vdc section, up to 550vdc at 100ma, that outputs to a stepped RC filter arrangement like you would see in a typical tube amp plus an array of output jacks for distribution.
2. Raw adjustable AC from the variac available at a pair of Jacks.
3. Two tube filament 6.3VAC at 1.7amp supplies distributed with a pair of jacks.
4. A large meter movement that displays the main B+ DC value.
5. A knob to control the main B+ DC amplitude.
6. Mains is fused on the back. I may replace this with a circuit breaker.

The design uses a variac connected to the primary of a large 500VAC transformer.

How about those slick graphics? - printouts taped over with shipping tape.
 

Attachments

  • IMG_0042.jpg
    IMG_0042.jpg
    933.9 KB · Views: 279
  • IMG_0046.jpg
    IMG_0046.jpg
    952.3 KB · Views: 268
  • IMG_0041.jpg
    IMG_0041.jpg
    995.8 KB · Views: 267
Nice. One thing i would add is a current meter so you know what the load is drawing, will be very useful in the long run.

I use a Heathkit IP-2717 high voltage supply which has a standby switch to disconnect the HV, however i always always always unplugged the + cable before going into the device under test. Breaker switches and all that are technically great, but physically unplugging is the best insurance against user error.
 
Last edited:
Nice execution Duncan2. I especially like your "single-point terminals on PCB construction" method.
Those terminals are apparently insulted from the mounting surface?
The clad copper PCB resembles a GND plane.
Both you and mcandmar highlighted the use of a current meter which I will add to my unit.....and the regraduate the dispay method - cool!
I'll have to look that up.
 
Nifty build, a HV PSU is invaluable.

Mains is fused on the back. I may replace this with a circuit breaker
I'd keep the mains fuse and fuse the HT with a small fast fuse. Circuit breakers in my experience don't act fast enough and aren't available in small enough current ratings.

That said I do go through a lot of fuses,it can get expensive (luckily I have about 3000 Russian NOS fuses), so a circuit breaker makes sense. Best thing to do is test said CB and see how it performs.

Andy.