High Power HVPS / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

High Power HVPS

1 2 next

Move Thread

LAN_403

Masih

Fri Jan 04 2008, 09:50PM

Registered Member #1213 Joined: Fri Jan 04 2008, 09:40PM
Location:
Posts: 14

Hello friends
I am looking for different possible topologies of high power HVPS
actually i want to make a 16kv DC power supply(fixed voltage) with the output current of 300mA.
another question
can we use high voltage power supplies in series to achive a higher voltage?
are the rules same as low voltages?

:) Happy New Year to everyone

Kizmo

Fri Jan 04 2008, 11:43PM

Registered Member #599 Joined: Thu Mar 22 2007, 07:40PM
Location: Northern Finland, Rovaniemi
Posts: 624

wrote ...

Hello friends
I am looking for different possible topologies of high power HVPS
actually i want to make a 16kv DC power supply(fixed voltage) with the output current of 300mA.

16kV and 300mA RMS current? That makes 4800W of power! How accurate your voltage needs to be? Easiest thing would be to use HV transformer like potential transformer(s) with rectifier bridge and some capacitor if needed and beefy variac for fine tuning

wrote ...

can we use high voltage power supplies in series to achive a higher voltage?
are the rules same as low voltages?

It depends what kind of power supplies they are. But generally you can put at least two HV supplies in series (At least i have done that with success, but you do it at your own risk) if you ground one HV lead of both supplies and put your load between two remaining leads. Of course this doesn't work with already centertap grounded devices such as NST:s

Hazmatt_(The Underdog)

Sat Jan 05 2008, 03:45AM

Registered Member #135 Joined: Sat Feb 11 2006, 12:06AM
Location: Anywhere is fine
Posts: 1735

If you want it to be 16kvdc current regulated that's going to be a pretty sophisticated piece of hardware.
If you want to regulate it in any way you're going to need a small load across the terminals to draw a small current for feedback and control of the regulator, but at this kind of power that means a really big resistor, like several hundred watt or more.
To series two supplies you're going to have to design it with really high voltage isolation. To series two of them you will need each 16kv supply to have at least 32kv isolation built in, which is also going to be a challenge.

Masih

Sat Jan 05 2008, 05:43AM

Registered Member #1213 Joined: Fri Jan 04 2008, 09:40PM
Location:
Posts: 14

in this power supply the ouput ripplr should be less than 100v.
so chosing a suitable amount of capacitors is mandatory. as you know buying capacitors in Farad Ranges for HV is so costly. what do u thing about switching. u know the ripple formula V ripple = Idc/(2F*C)
if we increase frequency we can decrease capacitors.
But
what frequencies is suitable for this? I couldnt find useful web resources for designing such power supplies. I need tips and tricks in designing which are dominant in HV Design.it will be great if u introduce. thanks

Steve Conner

Sat Jan 05 2008, 01:36PM

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

What application is it for?

How efficient does it need to be?

What are the limits on size, weight and price?

Are you good enough at electronics to build a 4.8kW switching power supply? (If you're not, don't expect us to teach you everything you need to know)

The biggest SMPS I've seen built by an amateur was Ralph Hartwell's 1kW, 2000V power supply for a tube RF amp:

You should understand this design completely, see why it won't scale to 16kV, and have a good idea of how to do it differently so that it will. Hint: notice the problems he had with stray capacitance in the transformer secondary and output filter choke.

Masih

Sat Jan 05 2008, 07:28PM

Registered Member #1213 Joined: Fri Jan 04 2008, 09:40PM
Location:
Posts: 14

this power is going to be used in physics lab for laser,plasma researchs.
now I am evaluating different methods and topologies which are possible in this project. I dont expect teaching. I said before. only want to know topologies. can anyone count the possible topologies for this range of power supply? both linear and switching.

Steve Conner

Sun Jan 06 2008, 07:06PM

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

As far as switching power supplies go, one important choice you need to make is whether to go current-fed or voltage-fed. This article explains the difference.

Another important choice is whether to go for resonant or hard-switched.

A third choice is whether to go for stacked transformers, or a single transformer with internal diode splits like a TV flyback, or a single transformer with a Cockroft-Walton multiplier.

These choices should permute most of the possible topologies. Someone else can discuss linear supplies.

MOT_man

Sun Jan 06 2008, 07:16PM

Registered Member #1127 Joined: Mon Nov 19 2007, 12:08AM
Location:
Posts: 139

This is quite easy to build. I'd go with 2 large core 1000 watt MOTs from a microwave oven. Use 4 MO capacitors of value 1 uF 2500 V - 2 in series on each leg of your MOT output and 6 sets of HV microwave diodes. I'd expect the voltage to be at least 16 kV DC 300mA without much hassle. The MOTs will get fairly warm after a few minutes - an oil bath or high velocity fan used for cooling will help here!

I've seen some pretty insain MOT power supplies! I know a guy from university who wanted to build a multi kW 20kVDC 480mA unit. Yep, he used MOTs to do it! 4 MOTs in series with MO caps, diodes --- It was a beast! It was a home built laser PSU as far as I know.... CO2, I'm guessing. I don't have much experience with lasers....

mikeselectricstuff

Sun Jan 06 2008, 11:21PM

Registered Member #311 Joined: Sun Mar 12 2006, 08:28PM
Location:
Posts: 253

In a laser lab environment, chances are you have 3-phase mains available. One solution to reduce ripple in a conventional transformer arrangement might be to use 3 transformer/rectifier combinations, one on each phase.
However 100v ripple on 16KV is going to be a challenge. Do you really need this or is it an arbitary figure decided withour regard to how easy or otherwise it woudl eb to achieve? Unless your time is very cheap you would probably be better off looking for a (used) commercial supply from people who know how to make this stuff.
Running at anything significantly higher than line frequency poses increasing problems with capacitance loading the secondary. I believe this is why TV flybacks often have diodes in series with seperate winding sections.

Masih

Mon Jan 07 2008, 07:54PM

Registered Member #1213 Joined: Fri Jan 04 2008, 09:40PM
Location:
Posts: 14

the output voltage is not arbitrary and calculated by experts. actualy i can provide a transformer for producing HV from mains voltage. first I decided to chose the staright forward method of full bridge diodes and capacitor to achieve this High voltage. but when I calculated the capacity of capacitor for 100v of ripple I got disappointed of this way.
the capacitance which yields is 30uF which is huge in size and cost. so I decided to work on switch mode.
but about MOTs: Not bad as an Idea.but there are some questions. Does it able to provide the current for more than an hour.Mots are small and will get warm during normal application. I have not worked with them. so dont know how they respond during my cited range of appication? tell me

1 2 next

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