High-voltage PowerPC-based inverter / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

High-voltage PowerPC-based inverter

1 2 next

Move Thread

LAN_403

Dr. Mario

Wed Aug 04 2010, 12:09AM

Registered Member #3059 Joined: Tue Aug 03 2010, 04:09AM
Location: My turf
Posts: 18

Hey all!

I'm wondering that if going with H-bridge is good enough for high-ampere high voltage, or should
I go with different SMPS topology? The whole affairs there in the metal PC box will be driven straight from
PWM generators found in Freescale MPC555 PowerPC microcontroller. (The reason I'm using PPC MCU is safety...)

I'm aiming to make a Helium-Ion laser capable of generating X-ray laser lights, so I'm investigating on the best way
to generate lots of Ampere of 20 - 75 kilovolts AC. (From there, it will be recifited to DC by the expensive, water-cooled
HV bridge recifiters, to charge some of special Mica or ceramic capacitor, with extremely low ESR rating.)

The reason I am asking you is because there are much of experienced peoples here on this forums who have gotten better results with tesla coils and / or HV inverters. And, yes, you can post your codes here if you want to show me how it's done by the softwares inside the MCU (I don't care what CPU architecture, I would study them in details to get better idea...)

And, Thanks for taking your time reading this post! (BTW, 4hv.org's a great website!)

Pinky's Brain

Wed Aug 04 2010, 06:25AM

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

Just curious, do you really mean that you want a continuous HV power supply in the >100 kW range?

HV Enthusiast

Wed Aug 04 2010, 12:19PM

Registered Member #15 Joined: Thu Feb 02 2006, 01:11PM
Location:
Posts: 3068

To be honest, if you have to ask about something as simple as topology for a big undertaking like that, you're probably not ready to undertake such a task.

Again, another thread asking for design hints without really specifying what you want.

If you want to get serious answers to your inquiry, be sure to list all the requirements for your design. Terms like "lots of Ampere" means absolutely nothing.

A better approach would be to say . . .

"Okay, i want to develop a high power, high voltage power supply. For input power, i have available 208VAC, 3 phase power, and for output, I would like to have regulated high voltage, adjustable from 50kVDC to 75kVD with a maximum continuous current of 100mA."

Dr. Mario

Wed Aug 04 2010, 10:18PM

Registered Member #3059 Joined: Tue Aug 03 2010, 04:09AM
Location: My turf
Posts: 18

Well, getting higher-ampere from high-voltage inverter transformer is not impossible at all. Consider a microwave oven inverter: 2.67kV .5 Amp (500 milliamps).

I have done few SMPS before. I was just curious if H-bridge is correct as assumed.

(Oh, and I already know about Eddy current problems: Going with a fat, solid wire for secondary is a bad idea, you could burn your hard works away in a matter of minutes, or maybe instantly. So, therefore, I'm going with a fat, multi-threaded wires(aka Litz wire), to prevent that from happening.)

Dr. Mario

Wed Aug 04 2010, 10:22PM

Registered Member #3059 Joined: Tue Aug 03 2010, 04:09AM
Location: My turf
Posts: 18

(To Pinky's Brain: Both yes and no. I, however, can adjust that power as much as I want to. Hence, the SMPS' totally tunable -
I just simply turn the optical-encoder knob to change AC waveform/frequency from the PPC MCU chip's PWM output)

Pinky's Brain

Thu Aug 05 2010, 07:41AM

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

I can't help but be sceptical ... that kind of setup is such a huge undertaking that worrying about software seems kind of silly.

Haven't ever heard of a x-ray laser powered through a continuous electrical discharge BTW. How exactly would that work?

Dr. Mario

Thu Aug 05 2010, 10:18PM

Registered Member #3059 Joined: Tue Aug 03 2010, 04:09AM
Location: My turf
Posts: 18

X-ray laser by discharge in gases occurs through Z-pinches: You just rip handful of electrons off the atoms of gases, such as Helium or Neon, so much that free-flying electrons will be slamming down to the "holes" of the atoms, that it would emit photons of ultra-short wavelength. Electrical and chirp-amplified IR lasers are two way to do that. Besides, it's not continuous, but instead, periods of dead-times (when AC wave crosses to zero-volt) - allowing Helium to return to "Ground State", cooling off.

Why software? Three reasons: To try and arrest the voltage spikes onto primary coil when fed at full power (12 Volts AC at 16.7 Amps) and, also safety... It should als should be able to monitor the brightness of special glass scintillator crystal being aligned near a solar cell connected to digital circuitry (in power meter setup) and send back the result to PowerPC microcontroller some informations to probably correct the frequency drift (eg. Optical output wattage drift). Again, I take safety of the others seriously (Yes, when the controller plug (D-15 D-sub plug) is to be disconnected by accident, the inverter will shut down instantly because X-ray is totally invisible and is pretty dangerous - a bad thing for the oblivious).

If I can't push ampere over 5 A limit at HF, then the only option left is to basically push voltage to the point it's high enough for Z-pinch to occur.

Pinky's Brain

Fri Aug 06 2010, 07:21AM

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

I can do thought experiments to try to unite your disparate statements to try to gauge what you want. (A couple 100 Watt of input power for several amperes of AC at near 100 kV does make it hard though.) The fact that you make people just guess at what exactly you want does give you the appearance of a troll you know.

Dr. Mario

Fri Aug 06 2010, 10:04PM

Registered Member #3059 Joined: Tue Aug 03 2010, 04:09AM
Location: My turf
Posts: 18

Yeah. I just only asked you guys few question before I would dirt my fingers with soldering pastes and fluxes.
I have done few inverters before, but this one is abit challenge, because I want to be able to ram the juices
down the Aluminum Oxide callipary tube filled with Helium or Argon, with enough brutal forces.

I think the best spot for HV without Litz wires exploding over the limit would be up to 180+ kilovolts before
any corona discharge starts to show up as a nasty short-circuit (after it's satisified with eating through insulation)
so, maybe 100 - 140 kilovolts at around 1 - 3 Amps, just like the inverter inside the Computed Tomography's X-ray head
(BTW, the X-ray load transformer inside it's HUGE, a bit bigger than the high-frequency microwave oven inverter transformer).

Dr. Mario

Fri Aug 06 2010, 10:08PM

Registered Member #3059 Joined: Tue Aug 03 2010, 04:09AM
Location: My turf
Posts: 18

And, also Pinky's Brain, I'm going to solder the chips that some people around here will lament is difficult to solder
without being PERFECTLY SYMMETRICAL, that PowerPC microcontroller's package option I picked is PBGA-388. Yea,
soldering the chips with ball solders are hard, but I have done coupla in the past.

1 2 next

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