Author Topic: Designing a 500w 13.56 Mhz RF Generator  (Read 423 times)

0 Members and 1 Guest are viewing this topic.

Online thinksnow247

  • Newbie
  • Posts: 3
  • Country: us
Designing a 500w 13.56 Mhz RF Generator
« on: December 07, 2017, 09:03:29 AM »
So I am a bit out of my element here as I come from a mechanical engineering background and have experience doing automation design and integration. I can get by soldering surface mount components and sort of know my way around a scope for debugging, but I have huge random holes in my knowledge so I am looking for some guidance.

RF power generators design is a whole other level beyond what I currently do. I am looking to build an industrial generator to drive a load that has been tuned close to 50ohms. The loads is a coil with parallel capacitance heating a small metal part. Fixed setup, so not a lot of dynamic changes in the load.

Microsemi has a design that looks like it might work and uses one of their integrated RF drive/mosfet modules that I am looking to use as a starting point: https://www.microsemi.com/sites/default/files/micnotes/DRF1200_06102008.pdf

The design seems fairly simple and compact because of the module. Keeping the size of this system small is very important.

My first question is very basic. What is the best way to supply the 220V dc that is needed to produce around 500 watts output? I can use a high voltage bench supply for testing, but this thing needs to be packaged up into as small of a box as possible. Standard available ac-dc power supplies top off at around 50V until you get to stuff in the 360-400v range. Can I feed this thing through a step up transformer so I can use something like a 600w 50v power supply. What other approach is needed?

I have a bunch more questions, but let's start with powering this thing in a rational way.
 

Offline Kire Pûdsje

  • Regular Contributor
  • *
  • Posts: 53
  • Country: nl
Re: Designing a 500w 13.56 Mhz RF Generator
« Reply #1 on: December 07, 2017, 11:06:28 AM »
Define small. Also try searching for adjustable CNC supply on e.g. ebay.

Also since this is a class E amplifier, I would not play around too much with the load impedance by using transformers, etc
« Last Edit: December 07, 2017, 11:10:09 AM by Kire Pûdsje »
 

Offline jgalak

  • Contributor
  • Posts: 49
  • Country: us
  • KQ2Z
Re: Designing a 500w 13.56 Mhz RF Generator
« Reply #2 on: December 07, 2017, 03:25:54 PM »
Do you have 220V AC available where this is going to be used?  How pure does the DC need to be for this application?

Can you get away with just a bridge rectifier from the mains and some capacitors to smooth out the ripple?  (just make sure it's all rated for high enough voltage)
 

Online thinksnow247

  • Newbie
  • Posts: 3
  • Country: us
Re: Designing a 500w 13.56 Mhz RF Generator
« Reply #3 on: December 08, 2017, 06:28:57 AM »
Define small. Also try searching for adjustable CNC supply on e.g. ebay.

Also since this is a class E amplifier, I would not play around too much with the load impedance by using transformers, etc

Thanks for the ebay suggestion, that gives me some supplies that would probably work for a prototype. Is there a reason I don't see this sort of supply from the major brand name manufactures? Is it a compliance issue or are they just custom order items?

As far as size, I would love to find a conduction cooled brick style supply model as I can easily integrate the heat sinks as part of the chassis structure:
https://www.artesyn.com/power/assets/ait02zpfc_datasheet_rev01_1192030301.pdf
http://www.meanwellusa.com/webapp/product/search.aspx?prod=HEP-480

There are some high voltage/power LED power supplies for constant current, not sure if those can be workable in this situation.
 

Online mikeselectricstuff

  • Super Contributor
  • ***
  • Posts: 10197
  • Country: gb
    • Mike's Electric Stuff
Re: Designing a 500w 13.56 Mhz RF Generator
« Reply #4 on: December 08, 2017, 08:41:28 AM »
Define small. Also try searching for adjustable CNC supply on e.g. ebay.

Also since this is a class E amplifier, I would not play around too much with the load impedance by using transformers, etc

Thanks for the ebay suggestion, that gives me some supplies that would probably work for a prototype. Is there a reason I don't see this sort of supply from the major brand name manufactures? Is it a compliance issue or are they just custom order items?

As far as size, I would love to find a conduction cooled brick style supply model as I can easily integrate the heat sinks as part of the chassis structure:
https://www.artesyn.com/power/assets/ait02zpfc_datasheet_rev01_1192030301.pdf
http://www.meanwellusa.com/webapp/product/search.aspx?prod=HEP-480

There are some high voltage/power LED power supplies for constant current, not sure if those can be workable in this situation.
There aren't many applications for high voltage DC supplies. For 200V, four 48v PSUs in series might be a practical solution.
As well as the HEP unit above, look at Meanwell CLG and HLG series. There are very few PSUs designed for baseplate/conduction cooling. The Meanwell ones are by far the cheapest and most readily available.
Bear in mind that SMPUs may well freak out if they see significant amounts of 13MHz being kicked back into their output.
Youtube channel:Taking wierd stuff apart. Very apart.
Mike's Electric Stuff: High voltage, vintage electronics etc.
Day Job: Mostly LEDs
 

Offline CJay

  • Super Contributor
  • ***
  • Posts: 2048
  • Country: gb
  • 2E0EOA
Re: Designing a 500w 13.56 Mhz RF Generator
« Reply #5 on: December 08, 2017, 09:20:46 AM »
A mains isolation transformer would be my preference but it won't be small or light, maybe a toroid?
2E0EOA
 

Online thinksnow247

  • Newbie
  • Posts: 3
  • Country: us
Re: Designing a 500w 13.56 Mhz RF Generator
« Reply #6 on: December 09, 2017, 07:22:59 AM »
Thanks for the input so far. I have a TON of learning to do. My understanding of this type of high voltage design is that they are supposed to achieve high efficiency by reducing current conduction losses vs operating at 24V/48V. Stringing together a bunch of power supplies to get to high voltage is probably a bit self defeating.

I think the proper approach to this is to choose a power supply and design the load network to accommodate and produce the desired power output.

So maybe start with something like 360V supply: http://www.tdk-lambda.com/products/sps/catalog/eng/pf-a.pdf

Or start at 48V.

And redo the load network to make it work. Am I headed in the right direction here?

There is also the compliance/safety issues I will need to consider. Is it not practical to build a product with high voltage DC? I am planning on a grounded metal case with no openings to the high voltage/RF side of things and trying to conduction cool everything to limit emissions and for safety. There are other amplifier designs in this power range that use 24v and 48v supplies, but in theory a high voltage system should be more efficient? :-//

 

Offline CopperCone

  • Frequent Contributor
  • **
  • Posts: 373
  • Country: us
Re: Designing a 500w 13.56 Mhz RF Generator
« Reply #7 on: December 10, 2017, 05:51:26 AM »
I think power sequencing might be an issue for series supplies. I'm not sure how they should be powered up.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf