2kW DRSSTC Converter / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

2kW DRSSTC Converter

1 2 3 next

Move Thread

LAN_403

HV Enthusiast

Wed Aug 02 2006, 01:40PM

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

We are just about finished of our new commercial DRSSTC prototype. The design utilizes a standard commercial brick-size converter and can provide power output up to 2kW (1kW derated). We plan on using these eventually for our commercial DRSSTC design which will be in a much smaller form factor that the one previously shown.

So far, the converter has provided power levels of up to 2kW (tested) with arc lengths exceeding 50 inches.

Everything except for the primary capacitor and DC power supply are integrated into the converter including feedback transformers, drive electronics, fault circuitry, etc...

Attached is a conceptual image of the converters. I'll post actual images once I receive the converter labels from the printhouse.

1154526012 15 FT0 Drsstc Converter Concept01

ragnar

Wed Aug 02 2006, 01:56PM

Registered Member #63 Joined: Thu Feb 09 2006, 06:18AM
Location:
Posts: 1425

Heh, love the 3D models.

Is the brick-sized converter supposed to be the drop-in electronics for anybody who could care to make a resonator, or do you sell these with complete systems for e.g. universities?

And may I ask how many is "we"?

Steve Conner

Wed Aug 02 2006, 02:03PM

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

What exactly is that box? a PFC front end that would work with your own inverter? or some kind of inverter that can drive the DRSSTC tank circuit directly?

It looks like a Vicor PFC front end to me.

HV Enthusiast

Wed Aug 02 2006, 02:16PM

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

The box is a complete DRSSTC converter. Simply feed it with say 400VDC (500VDC maximum), hook up your resonator, and the way you go.
It operates up to 1MHz (yep, 1MHz), although there is a derating curve vs. frequency.

Steve Conner

Wed Aug 02 2006, 03:02PM

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

That's pretty impressive. How much does it cost?

HV Enthusiast

Wed Aug 02 2006, 03:30PM

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

Steve Conner wrote ...

That's pretty impressive. How much does it cost?

A lot!

These are the types of devices I've been using (surface mount) and have had very good results with them.
The second device does much better at higher frequencies (1MHz i.e.), but has higher conductive losses. The
devices are inexpensive also, about $4.00 per pop is small quantities.

http://www.ixys.com/99109.pdf

http://www.ixys.com/99041.pdf

Also, the converter is made up of two boards (one on top of the other), and the bottom board is a special IMS aluminum substrate board.

Steve Conner

Wed Aug 02 2006, 04:00PM

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

So you've bought a commercial resonant converter driver and changed out the MOSFETs for IGBTs? Good going!

As for the price, I'm interested to see how it compares to the PLL electronics package handmade by my bad self. I've sold about a half dozen of them so far.

Terry Fritz

Wed Aug 02 2006, 05:23PM

Registered Member #393 Joined: Tue Apr 18 2006, 12:30AM
Location:
Posts: 297

Cool Dan!!!

You must be pretty sure of the electronics to pack it! No changing a resitor value now

Having played in the design and prodcution of such things in the past a few minor suggestions (even if it is too late and you prolly all ready know all this

)

These packages tend to "bow" a bit. Your transistors may be small enough not to want to bend things as they cool though and your power level is probably resonable if there is some spacing to the heatsink. A solution is to put two more screws in the middle. We also put the power stuff near the screws and the low power guts in the middle. One often needs 1/8 inch thick minimum bases.

Be careful of heat sink greasing. Best to brush it on in a smooth even layer and pull it back off to check for voids. If one puts on too much in hopes the screws will squeeze it out, you will just bow the package from the enourmous pressure of the grease. Very soft thin thermal pads are a simple option too.

SMT caps are real bad at flexing, they just break. Best to go to through hole there... That is what kills most (almost all!) Vicors...

You may want "locking" pin connectors (for the control pins) with locking tabs on each side. The cables always want to fall off over time. There are high force connectors though that won't fall off easily. Or, mount another thing directly above the connector so it can't come off.

If you use potting compounds at home, be sure to store them an discard them when the date expires. That is really important! The stuff goes bad easily.

You should thermal cycle the package about 100 times. If you can't get access to real HALT equipment, just go from the oven to the freezer with three fo them. That will detect any screwed up bondinging and expansion issues. I am not sure how you are bonding to the IMS, but that is always a bad spot...

If the base is meant to be at ground, you will probably have to officially hi-pot test it if there is any chance at all for real voltage making it to the mains ground.

That's pretty impressive. How much does it cost?

A quick calculation shows that it will cost Dan a whole lot less than it will cost us

Very Cool!!!

Cheers,

Terry

robert

Wed Aug 02 2006, 06:39PM

Registered Member #188 Joined: Thu Feb 16 2006, 05:18PM
Location:
Posts: 67

Yes, thermal grease is dangerous on these modules.

I have mounted several (Astec AMPSS II and Vicor) dc/dc modules at my last job and i was happy to finally recieve the ordered phase-change film.
Gerase works but one needs to heat the Heatsink and converter to 80Â°C or above to soften it enough so it flows properly and can be spreads out by carefully pressing on the module on different places (wearing thick heat resistant gloves because its darn hot).Takes a long time and really sucks.
When it was cooled, the extremely thin grease film hardened again and it was impossible to pull the modules off because of the strong adhesion.
The mounting screws need to be tightened with a torque-indicating driver and need spring washers of some special expensive type too.

Of course everything needs to be _real_ clean (using analytical grade solvents and cleanroom-proof wipes) and the assembly should be best done in a cleanroom.Gloves are necessary too because they didnt want to see any fingerprints anywhere.

At home im having a 260w AMPSS module supplying 13,8v to some battersy chargers and other stuff.
Nice things, small and real efficient.
Didnt bother with PFC, just rectified 230v and a 1mF 500v filter cap (and 3-stage noise-filtering, inrush current limiting and some more emi filtering).

Using the film is simple, just stick it on, screw the module to the heatsink, heat it up to 80Â° in a drying oven, after cooling down re-tighten it and thats it.
Never had any failiures, but i had failiures on vicor modules that were mounted by some amateur, just a blob of grease and heavily overtightened bolts.

Terry Fritz

Thu Aug 03 2006, 06:35AM

Registered Member #393 Joined: Tue Apr 18 2006, 12:30AM
Location:
Posts: 297

Sounds like Robert's place was running modules as hard as we used to.

I never played much with the Phase Change waxy stuff but it sure looks cool! We could not do the massive reliability verification for such a "new thing". We did use some diamond stuff very happily(!!) but it sure was expensive!!

Once the silicone grease runs off the standard stuff, it is "dust"... The magic washers are "bellville" and they are an "extremely"(!!) nice thing for this stuff!!! MCmaster Carr has them... That $400 torque drive is nice... but just compress the bellville washer 50% and be happy

)

Of course, the trick is not to run things so hard that they need exotic thermal transfer layers... Big copper tungsten packages to 0.5 mil is "stupid" even though we made thousands...

But beware that thermal grease "sucks" compared to metal and these big bricks can have many thousanths of an inch of gap. Big bases help spread the heat, but there is a real limit. Omega sells little thermal stickers to test things that can be tested with worst case spaced thermal grease layers.

But if the heat per in^2 is not to high, it gets real easy

That is where you want to be.

I had the nightmare of testing the very first IMS boards...

I assume it is "all fixed" by now though....

These modules have a fews issues, but if you play them right, they are sweet!!

I don't mean to scare one, but these module can go "bad" and you need to be a bit carefull... but then they are "nice"

Cheers,

Terry

1 2 3 next

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