High frequency power supply design.

This forum is for specialized infomation important to the construction and safe operation of the high voltage electrical supplies and related circuitry needed for fusor operation.
Post Reply
Martin Shahi
Posts: 20
Joined: Thu Aug 29, 2013 1:06 pm
Real name: Martin Shahi

High frequency power supply design.

Post by Martin Shahi » Sat Nov 16, 2013 7:32 pm

Hello everybody. So I'm about to order the electronic components so that I can build my own high frequency ferrite power supply. However before I do that I would be very grateful if the electronics wizards out there could double check my design and spot any flaws if there are any. This is meant to power a full fusor, so would this be able to provide enough power to it?
Essentially it is a H-bridge of NPN transistors controlled by my raspberry pi to invert the rectified DC back into very high frequency AC (minimum 10 kHz).
Thanks very much

Martin
High frequency psu.tiff
High frequency psu.tiff (138.18 KiB) Viewed 3196 times
Edit: If you're having trouble viewing the image like I am, right click the icon and open in new tab.
Edit 2: Assume I'm using 10 AWG copper wire.

Ross Moffett
Posts: 89
Joined: Tue Aug 27, 2013 2:24 am
Real name: Ross Moffett

Re: High frequency power supply design.

Post by Ross Moffett » Sat Nov 16, 2013 7:45 pm

I'm planning something similar myself.

USE FUSES!! Parts of your circuit are exposed to UN-isolated mains voltage, you'll burn your house down and/or electrocute yourself.
Your "safety switches" isolate only the signal, while leaving the whole thing energized. How safe is that?
You aren't going to be able to drive FETs at any significant power level using the signal directly from an MCU. You'll need to isolate with MOSFET or IGBT driver chips, depending on which you're planning to use there.

What you've got here is only the most basic interpolation of what this kind of power supply has in the circuit. All kinds of supporting circuitry, transient voltage suppression, output voltage & current monitoring and other safety circuitry will need to be added. Strongly suggest you buy some good books on high voltage/SMPS design/transformer design, internet research can only carry you so far.

Martin Shahi
Posts: 20
Joined: Thu Aug 29, 2013 1:06 pm
Real name: Martin Shahi

Re: High frequency power supply design.

Post by Martin Shahi » Sat Nov 16, 2013 7:59 pm

Ross Moffett wrote: FETs at any significant power level using the signal directly from an MCU. Strongly suggest you buy some good books on high voltage/SMPS design/transformer design, internet research can only carry you so far.
Thanks for the advice! Could you please clarify though, what does FET and MCU stand for? Also could you recommend any good books for this sort of thing?

Ross Moffett
Posts: 89
Joined: Tue Aug 27, 2013 2:24 am
Real name: Ross Moffett

Re: High frequency power supply design.

Post by Ross Moffett » Sat Nov 16, 2013 10:07 pm

I'm sorta winging it because I studied this stuff in college, but when I buy books I trust those with lots of positive reviews on amazon or look for online courses book requirements. MIT open courseware, for example.

Google with "define: FET" and it will change your life.

Field effect transistor
Metal oxide semiconducting field effect transistor
Insulated gate bipolar junction transistor
Micro-controller unit (generic term for all microcomputers being used for control)

User avatar
Chris Bradley
Posts: 2930
Joined: Fri May 02, 2008 11:05 am
Real name:

Re: High frequency power supply design.

Post by Chris Bradley » Sat Nov 16, 2013 10:40 pm

It is quite a mystery as to what your circuit is supposed to be showing.

You have a very steep learning curve ahead of you if you want to go down this route of building an H-bridge. The 'quickest' way might be to go pull data sheets for H-bridge drivers - manufacturers of these put in some good information into their data sheets, obviously aiming to be as practically helpful as possible to sell their parts. If these datasheets fail to inform you because they are beyond your current level of learning, then it's difficult to make a suggestion as to how you are best placed to get to that point.

Some drivers make it so simple it is almost as easy as it looks! IR2453 [ http://www.farnell.com/datasheets/48102.pdf ] has in built oscillator, bootstrapping and dead-time. (If you do not understand bootstrapping and dead-time, and the difference between inverting and non-inverting drivers, then it is still better you find out before using a driver like this.)

A problem you do not yet appear to know you have is that without filtering on your mains rectifier stage you will have an output which follows the mains frequency.

Jerry Biehler
Posts: 900
Joined: Tue Nov 24, 2009 8:08 am
Real name:
Location: Beaverton, OR

Re: High frequency power supply design.

Post by Jerry Biehler » Mon Nov 18, 2013 8:29 am

You might want to find an existing design and use that. Your circuit, well..., its bad, very bad.

And dont use a raspi for this. That is just silly. This can be ran from a 555 or one of the many PWM chips out there or even a SMPS IC.

Ross Moffett
Posts: 89
Joined: Tue Aug 27, 2013 2:24 am
Real name: Ross Moffett

Re: High frequency power supply design.

Post by Ross Moffett » Mon Nov 18, 2013 3:37 pm

I'm planning to use a micro-controller for mine.. while the SMPS drive chips are available, they don't seem to accommodate feedback from a high voltage line very well. Anybody got a recommendation of a purpose-built SMPS chip which incorporates both current and voltage limiting for the DC output supply?

Post Reply