triac control of DC output
 

Any comments or schematics about using a triac to vary the duty cycle of AC power being fed to a transformer primary with its control voltage coming, as negative feedback, from the DC output of the rectifiers on the transformer? In other words. I'd like to have a linear DC supply being regulated by using a triac like a glorified light dimmer. I need 50V at up to 50 amps with fair(+/- 5-10%) regulation.
I've seen how SMPS are made and I'm not sure I will ever find a 50KHz transformer that will do the job, and I will never make one that big. I'm not concerned with the size of the transformer.
It also seems to me, perhaps I'm wrong, that my idea would be simpler than any of the SMPS I've seen.

There is a SMPS in the 2006 ARRL handbook, it requires building from scratch at least one transformer, and it's a BIG one! Its a very involved project and its 13.8Vdc not 50V.

Part two: Toroid corp of Maryland sells a transformer "kit", it has a pre-wound primary and you wind your own secondary to meet the voltage you need. They have cores from around 30VA to 1400VA. Three of the 1400VA cores, each with about 55 turns of wire will meet my power needs. But, I'd still like to use the triac control idea to keep the rail close to 50 Vdc.

At this power level the triac control approach is not very practical. The duty cycles are small, all the output energy for a whole half cycle has to be delivered during maybe 20% of the time. A 2500VA transformer won't be able to handle full continuous output, 10KVA is a more reasonable figure, this is due to the low duty cycle. Output capacitance requirements are also big. A SMPS is a much more sensible solution.

All things considered, I think this is an impractical project. It was to be the power supply for a 1000+Watt solid state RF power amp. I think I will use tubes instead. They are more efficent, cheaper, more reliable. And the power supply is smaller,cheaper, lighter and easier to build. I bet most people would never think that transistors are inferior to tubes in the ways I described. Thanks for your input.

50V 50A sounds like something you can find ready-made. A power supply for telecom applications should be trimmable to give that voltage. (look for 48V supplies)

I'll keep my eyes open. but like I said in my last post I think it is better to build a 3000V 500 mA supply for use with a tube instead.