In waaay over my head. (sourcing IGBTs) / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

In waaay over my head. (sourcing IGBTs)

1 2 next

Move Thread

LAN_403

sniglet

Thu Oct 21 2010, 02:38AM

Registered Member #3339 Joined: Wed Oct 20 2010, 03:12PM
Location:
Posts: 6

First, I REALLY appreciate your patience, there are holes in my knowledge, I suspect they're bigger than I'd like)

So I bought aautomation's sstc kit. It states it's NOT a beginner's kit. He's not kidding. I've got some rudimenatry background (automotive electrical, home wiring, wiring up a VFD for the lathe, Burning proms for the Vette, Basic Stamp stuff) I put it together, but it needs an oscilloscope, so I build one of those first ( http://www.sparkfun.com/commerce/product_info.php?products_id=9484 ) In ignoring some of the recommendations, I burn out the IGBT's and a fuse. (after getting three discharges)

Replace them with spares he kindly included, built a signal generator ( this one: http://www.canakit.com/200khz-sine-1mhz-square-function-generator-kit-ck307b-uk307b-ck307-uk307.html ) to send 4 VAC to the AC side (rather than the 12 VAC from the power supply) and learn about current. (Voltage supplied dropped from 4 VAC to .3 VAC when the power was turned on.)

Built a 7 VAC power supply with a radioshack transformer that could drive the circuit. Got about 5 minutes use this time, producing approximately 2 cm long sparks. I was homing in on tuning the primary coil when the sparks went away. So, improvement, but the results are still terminal, and honestly I am I'll equipped to speak about this stuff.

The IGBT's are a) out of stock, b) Half a world away, c) $18 a piece, d) and will ship in 4 to six weeks, in e)lots of 1000 pieces. A little research has shown they're pricey as they are out of production.

So can I replace FGL60N100BNTD IGBT's (Collector voltage 1000V, 60 Amp Collector current at 25C, 42 Amp Collector current at 100C) with IGBT's within 20% of those values? There's no shortage of 600V IGBTs ($3.67 60Amp/30 Amp), or 1200V IGBTs ($5.53, 60 Amp at 25 C, 30 Amp at 100C) but for electrical stuff, I'm kind of plug and play and don't understand the theory beyond a really high level.

FWIW, my research here leads me to think part of the problem is heat rejection in the latter attempt, and too much voltage on the initial attempt. At least I'm not making the same problems each time.

brtaman

Thu Oct 21 2010, 06:01AM

Registered Member #2161 Joined: Fri Jun 05 2009, 03:36PM
Location:
Posts: 247

There is nothing special about the IGBT's you want to replace.

However, I am not familiar with the kit you purchased. I would be happy to help with your choice, if you provide more detail.

How does the SSTC function? Is it a DRSSTC with a tuned primary circuit, or a regular SSTC that is interrupted or perhaps CW? What is the Operating Frequency of your secondary coil?

Some pictures of your setup would also be good, if you are blowing IGBT's I would address potential problems in the build first, before blowing up more silicon.

The IGBTs aren't exactly lightning fast and there is quite a few possible replacements out there. If you provide links to the IGBT's you are considering purchasing , a recommendation can be made.

Best Regards,
brtaman

sniglet

Thu Oct 21 2010, 12:11PM

Registered Member #3339 Joined: Wed Oct 20 2010, 03:12PM
Location:
Posts: 6

Photos of the build are here:

the schematic is here:

The last three photos of the board are how it exists today. I have to apologize at the condition of it. I used a TON of solder on the first pair of IGBTS, the scorching you see is from trying to use a micro-torch and desoldering braid to remove the solder. (I've since moved to a vacuum bulb with much better results, I've also taken to cutting the dead chip out, then removing the wire and solder as it gave better, quicker, cooler access to the area for removal. It IS electrically sound.

It _IS_ a DRSSTC, but it only lists two IGBT's, not '4 IGBTs in an H-drive configuration. I don't know the operating frequency of the secondary as the kit claims it's 'control circuits handle the part of finding the secondary coil's resonant frequency qutomatically'.

The original IGBTS are Fairchild FGL60N100BNTD (

)

Possible, but not exact replacements:

International Rectifier IRG4PC60UPBF link ( 600v, 40 A @ 100C, 75 A @ 25C, $4.83)

Fairchild FGA20N120FTD link (1200V 80 A @ 25C, 40A @ 100C, $8.26)

I've verified the TTL Side is producing the proper waveforms. I did NOT replace the electrolytics as they appear to be operating, at least partially, in the last test with the 2 cm sparks.

I had hoped to have it running for Halloween (it's supposed to be the focalpoint of Frankenstein's lab, last year that was a Jacob's Ladder) But i suspect it's just going to be static window dressing.

Maurizio Valvo

Fri Oct 22 2010, 10:49PM

Registered Member #3219 Joined: Mon Sept 20 2010, 09:06PM
Location: LeinÃ¬ (Turin), Italy
Posts: 23

I have used for my full bridge DRSSTC (running from 230Vac rectified mains) IXGR 60N60C2D1, 600V, 75A, 300A pulsed (IXYS HiPerFAST series).

Ken M.

Sat Oct 23 2010, 12:01AM

Registered Member #618 Joined: Sat Mar 31 2007, 04:15AM
Location: Us-Great Lakes
Posts: 628

I may have a missunderstanding about DRSSTC but I'm pretty sure thats an interrupted SSTC.

I this because if you look at most of the DRSSTC's built on this forum most have CT's on the primary side to monitor primary current and voltage since a DRSSTC has both the primary and secondary coils resonating, creating high current pulses as well as high voltage spikes...

Try putting some ZVS diodes (think thats what they're called) across the C and E of the igbt's the internal anti parallel diode in most fets and IGBts recover too slowly and receive huge current spikes causing the igbt or fet to die.

dmg

Sat Oct 23 2010, 12:21AM

Registered Member #2628 Joined: Fri Jan 15 2010, 12:23AM
Location:
Posts: 627

They are called TVS diodes (transient voltage suppression), the 1.5KE series are quite good for these.
also, be sure to get the bidirectional ones (suffix of part # is CA for bi), instead of the unidirectional TVS

mounting the TVS as close to the IGBT as possible is needed for them to actually be of any use for stoping those transients.

Sulaiman

Sat Oct 23 2010, 06:23AM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

What are you feeding the "Logic Input" with?
It should be short (maybe 50 to 250 us) pulses, you need an 'interrupter'
if the logic input is active for too long it will blow your transistors,
start with a low repetition frequency, 1 to 10 Hz with a short ON time, 50 us
and slowly increase the ON time for more spark power,
too much and you'll need new transistors!
Maybe sparkfun have an Interrupter kit?

Personally I'd connect the center-tap of T2 secondary to the junction of C1 & C2 to protect the eletrolytics from over-volting.

Goodchild

Sat Oct 23 2010, 06:45AM

Registered Member #2292 Joined: Fri Aug 14 2009, 05:33PM
Location: The Wild West AKA Arizona
Posts: 795

Ken M. wrote ...

I may have a missunderstanding about DRSSTC but I'm pretty sure thats an interrupted SSTC.

I this because if you look at most of the DRSSTC's built on this forum most have CT's on the primary side to monitor primary current and voltage since a DRSSTC has both the primary and secondary coils resonating, creating high current pulses as well as high voltage spikes...

Were the feedback comes from has nothing to do with what type of coil it is. What defines a DRSSTC is it having a resonant tank circuit.

Ken M.

Sat Oct 23 2010, 10:10AM

Registered Member #618 Joined: Sat Mar 31 2007, 04:15AM
Location: Us-Great Lakes
Posts: 628

That's sort of what I figured Goodchild, originally I thought I would say "ALL DRSSTC's on this forum" but I had shot myself in the foot saying things like that too many times, so I looked at some of the DRSSTC's and sure enough STeve ward has a secondary Base feedback DRSSTC, albeit it was his .5 DRSSTC, but none the less.

So I guess that means one could make a DRSSTC that gets the frequency off of the secondary but still implements a OCD on the primary.

Back to the original topic, Looking at the schematic again and Noticed that it has a 120:12V step down that then feeds into a 110,110:12V,12V wired as a step up, and the schematic says 400v, unless I've totally lost base with my basic AC knowledge wouldn't that only produce 220v?

Lethal Shot

Sat Oct 23 2010, 01:13PM

Registered Member #1623 Joined: Tue Aug 05 2008, 03:31PM
Location: The Netherlands
Posts: 39

Don't forget that the AC waveform is rectified meaning the DC voltage is sqrt(2) * VAC = 310V.
Where the figure 400VDC comes from is a mistery to myself too, it is probably chosen this way to be on the safe side when selecting components.

The thing that strikes me as odd is the low voltage of the MMC (only 2kV), isn't the impendance of the capacitor going to drive the voltage across it much higher in high current DRSSTC operation?

1 2 next

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