Glassman HV Supply / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

Glassman HV Supply

first previous 3 4 5 6 next last

Move Thread

LAN_403

Pinky's Brain

Wed Dec 01 2010, 12:54PM

Registered Member #2901 Joined: Thu Jun 03 2010, 01:25PM
Location:
Posts: 837

A series resonant converter has a LC tank circuit ... that's not inductive energy storage (or magnetic as Glassman put it).

Putting modulated rails on it or a QR flyback converter to avoid having to run at variable frequency is a nice thought experiment ... but seems unrealistic.

Finn Hammer

Wed Dec 01 2010, 04:31PM

Registered Member #205 Joined: Sat Feb 18 2006, 11:59AM
Location: SkÃ¸rping, Denmark
Posts: 741

Pinky's Brain wrote ...

Finn Hammer wrote ...

Now we know for a fact, that a good driver for a C-W voltage multiplier is a SLR Converter :

How do you come to the conclusion that they are using this? A regulated series resonant converter isn't fixed frequency is it?

Sounds like discontinuous flyback to me (zero current switching & fixed frequency).

CCPS has inherent short circuit driveability, it has zero voltage and current switching, and it operates at a fixed frequency.
If regulated, (the way that I would do it) it would be pulsed at full voltage, but still same frequency, just not CW.

Maby it is just wishfull thinking on my part since, as Steve has pointed out, The CCPS is one converter that I have a working knowledge of:

Cheers, Finn Hammer

Pinky's Brain

Wed Dec 01 2010, 05:44PM

Registered Member #2901 Joined: Thu Jun 03 2010, 01:25PM
Location:
Posts: 837

One big advantage of the flyback is that it doesn't give a shit about the fact that the power factor of a multiplier is atrocious.

Also I was wrong, you don't need two to generate the AC of course. Here is an example of how I think the Glassman power supply works (page 5&6).

Finn Hammer

Wed Dec 01 2010, 07:38PM

Registered Member #205 Joined: Sat Feb 18 2006, 11:59AM
Location: SkÃ¸rping, Denmark
Posts: 741

Pinky's Brain wrote ...

One big advantage of the flyback is that it doesn't give a shit about the fact that the power factor of a multiplier is atrocious.

Also I was wrong, you don't need two to generate the AC of course. Here is an example of how I think the Glassman power supply works (page 5&6).

A nice read, but donÂ´t you think the 20W is a bit on the low side? I am on the look for something that will deliver 1kW in, from the bottom of the stack, for 300kV out @ up to arounds 3mA. =:-)

Cheers, Finn Hammer

Pinky's Brain

Wed Dec 01 2010, 07:43PM

Registered Member #2901 Joined: Thu Jun 03 2010, 01:25PM
Location:
Posts: 837

More stages, more ferrite, more copper, more silicon.

A KWatt is not outside of the range of practical flyback converters.

PS. which is not to say I would know how to design one :) Just saying, other people have done it.

Steve Conner

Thu Dec 02 2010, 09:50AM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

Nobody makes flyback converters above a few hundred watts. The bigger you make them, the more snubbing they need and the more lossy they get. There is a fundamental limit to the performance, because the energy stored in the transformer's leakage inductance gets lost at every switching cycle, and that really hurts larger designs.

I think you can recycle the energy with active snubbers, but they use up ferrite, copper and silicon too. At 1kW everything is forward converters with half or full bridges.

When people here overdrive TV flyback transformers to 1kW using the "ZVS" driver, and draw huge arcs off them, it is actually running as a forward converter, not a flyback converter. If you use the transformer in the flyback mode it was designed for, you simply can't push that much power through it.

Pinky's Brain

Thu Dec 02 2010, 11:51AM

Registered Member #2901 Joined: Thu Jun 03 2010, 01:25PM
Location:
Posts: 837

Despite the downsides they have enough advantages people have used them at higher power. NASA for instance.

For resistive loads resonant converters are hard to beat, but rectifiers are an awfully annoying load which does hurt efficiency. There's also quite a few high power flyback designs which integrate isolation, AC->DC and PFC all in one nice package (the flyback can do it all in one pass). Also PFCs which use flybacks for that matter.

Hell, say you burn 100 Watt in the snubber, unacceptable for a commercial design ... but would you care?

PS. I didn't mean to suggest it could be done with a CRT flyback transformer.

Steve Conner

Thu Dec 02 2010, 12:03PM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

I don't get how that "Weinberg converter" in the NASA paper works, but I'm pretty sure it's a forward converter, not a flyback. Here

it says that it's related to the buck converter.

Anyway, NASA are pretty insulated from the real world of commercial SMPS design. They even had entirely new 500V, 100A, low Rds(on) rad-hardened FETs developed for that thing.

So are we, since we don't actually have to meet targets for efficiency and cost. Without those constraints you can get any power from any topology, like the old meme about strapping a jet engine to a Yugo.

Edit: I think I get it now, it's a buck-boost type thing. If you run the FETs at less than 50% duty cycle, it bucks. If you go above 50% and let them overlap, it acts as a boost converter, building up current in the DC bus inductor and boosting the output voltage above what a normal push-pull converter could manage.

Wolfram

Thu Dec 02 2010, 12:07PM

Registered Member #33 Joined: Sat Feb 04 2006, 01:31PM
Location: Norway
Posts: 971

Pinky's Brain wrote ...

Despite the downsides they have enough advantages people have used them at higher power.

That's not flyback topology, that's flyback current fed push pull, which is much more closely related with the current fed push pull topology.

Anders M.

Pinky's Brain

Thu Dec 02 2010, 12:14PM

Registered Member #2901 Joined: Thu Jun 03 2010, 01:25PM
Location:
Posts: 837

Flybacks are an isolated version of buck/boost.

The switches turn off current through the primaries abruptly, transferring the energy to the secondaries ... that's flyback. Saying it's more closely related to normal forward transformer action is silly. Yes it's push pull, but it's still flyback.

first previous 3 4 5 6 next last

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob