Solid State coil / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

Solid State coil

Move Thread

LAN_403

Frosty90

Mon Oct 06 2008, 09:51AM

Registered Member #1617 Joined: Fri Aug 01 2008, 07:31AM
Location: Adelaide, South Australia
Posts: 139

Hello Everyone,

This is my first post, but i have been reading the forums now for a while, and I have some questions that hopefully you can help me with.

After building a SGTC i had decided i would try a small MOSFET coil, and after much experimentation and help from the pupman.com tesla coil mailing list, I finally got a small 3.5" coil working well with a full bridge of IRFP460's. The driver circuit consists of a CT on the secondary ground wire feeding into a 74hc14, then this signal is and-ed with an interuptor signal from a 555 and used to drive 4 tc4420 MOSFET drivers, which then drive 2 gate drive transformers and then the full bridge of mosfets. After lots of messing around and 'fun' i managed to get max 40cm sparks in interupted mode with 5 turns on the primary, and in "continuous" mode with hlfwave rectified mains and 15 or 10 turns on the primary i got about 20cm very hot sparks, with only mild warming of the mosfets. here is a couple of vids:

in interuped mode with 5 turns on the primary:

in continuous mod with 15 turns:

All was working reliably until i allowed the coil to make ground strikes repeatedly, then 2 of the fets died, (they didnt explode, i havnt had this happed yet at all thankfully, they simply went dead short and blew the fuse). A little examination revealed no apparent problems, but there was some evidence of arcing on the feedback CT, which i find odd but i think could have been due to the ground strikes.
After replacing all 4 mosfets and better insulating the CT, i could only gt the coil to give 10cm or so arcs before 2 more fets died again. This is where im at now, I havnt bothered to try again, as i dont want to use up all my MOSFETS! (I purchased 24 from "thai shop etc." on ebay at $1.60 each)
So what could have caused the failure? The heatsinks didnt even get warm, and the fets werent showing any signs of heating. I checked and re-checked the drive circuit and all seems well, and i replaced the tranzorbs across the gates just be sure as well, but as i said, i could only geta few seconds of feeble operation before it died again. So does anyone have any ideas as to what could have killed my MOSFETS?

Cheers
Jesse

Mates

Mon Oct 06 2008, 02:23PM

Registered Member #1025 Joined: Sun Sept 23 2007, 07:53PM
Location: Czech Rep.
Posts: 566

Frosty90 wrote ...

So what could have caused the failure? The heatsinks didnt even get warm, and the fets werent showing any signs of heating. I checked and re-checked the drive circuit and all seems well, and i replaced the tranzorbs across the gates just be sure as well, but as i said, i could only geta few seconds of feeble operation before it died again. So does anyone have any ideas as to what could have killed my MOSFETS?

Cheers
Jesse

I would start with placing some TVS across the FETs and zeners on the gates. IRF460s are bastards - they like the "cold death". In case your TVS rated let's say for 440V will be hot or even explode during the run than you know it's voltage spikes killing your transitors. If not than you can check for "cold death transitor" issue well discussed somewhere on this blog (maybe somebody can send the direct link)

Good luck man!

BTW: Ground strikes are shown very rearly in case of SSTCs and only the real SSTC design masters can afford to do it - guess why?

Steve Conner

Mon Oct 06 2008, 02:31PM

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

I've heard rumours that the FETs from thaishopetc. may be fakes.

If anyone wants to send me one or two unused ones, I'll test them out and post my findings.

Dr. Dark Current

Mon Oct 06 2008, 02:54PM

Registered Member #152 Joined: Sun Feb 12 2006, 03:36PM
Location: Czech Rep.
Posts: 3384

Ground arcs with secondary current feedback and no start-up oscillator are the ultimate MOSFET killers.

As you ground the secondary, it appears to have no inductance and capacitance, so nothing can resonate. This results in a long single-polarity pulse to the bridge, destroying it.

Frosty90

Tue Oct 07 2008, 04:24AM

Registered Member #1617 Joined: Fri Aug 01 2008, 07:31AM
Location: Adelaide, South Australia
Posts: 139

Hi,

Thanks very much for the replies,

Mates, i did have tranzorbs (instead of zeners) across the gate (18v), but not across the drain/source
Dr Kilovolt: I didnt mention it but i did have a startup oscillator, but i cant remember if i had it connected when the thing blew.
Steve; thats relly interesting about the fakes, i did wonder why they could b so cheap! I have plenty so if you want i can send some, justlet me know

Ive decided to start again and build a whole new coil, as i have most of the parts and have thought alot more carefully about the design, but i would really like to continue to use CT feedback, so what would be the safest (for my h-bridge) way to do that? I dont like to use an antenna, because my first attempt resulted in more streamers from the antenna than the secondary!, plus it wasnt as reliable (for me) as the CT.
Also, if i want to just use 1 GDT, to drive 4 fets, should the core be larger? At the moment the cores i have are about 3cm diametre toroids with about 0.5cm square cross section.

Cheers
Jesse

Dylan

Tue Oct 07 2008, 09:37AM

Registered Member #1024 Joined: Sun Sept 23 2007, 10:56AM
Location: Northern NSW, Australia
Posts: 95

Hi Jesse,
Firstly congratulations on your coil, the video is great.
You mention you would preffer to continue using CT feedback, as opposed to an antenna, and thats fine, I agree, and all my secondary feedback coils use a CT.
The issue is secondary feedback vs primary current feedback and the overcurrent protection options available.

---------------------
Basically if you are using secondary feedback, which if fine for most small to medium coils, and if you do not have any form of primary current limiting ( either active or by nature of your primary circuits impedance ) , then you should probably avoid direct ground strikes. Keep the streamers in the air :).
---------------------

There are a few options for saving your bridge in the event of loosing secondary feedback:
-An oscilator to keep the bridge moving and rely on the primary inductive reactance.
-A dc blocking cap in your primary circuit (implicit in a half bridge) just like in a DRSSTC tank , just not tuned.
Or both :)
-Active primary current limiting in conjunction with one or both of the above and using a CT, or on its own using a DC reactive current to voltage converter (a resistor

).
If you add primary current limiting ground strikes should definately be more survivable.
Normally in an SSTC current limiting is achieved simply by a high impedence (lots of turns) primary, but you could in theory implement active primary current limiting in an SSTC.
However if you are adding active primary current feedback using a primary CT you may as well implement primary feedback at the same time. Primary feedback will give much more robust performance under varying secondary tank conditions, and will improve ground strike survivability, however primary feedback requires a tuned primary tank, so implies a DRSSTC

.

You can see how all roads lead to DRSSTC !

So now I come full circle and we end up where (for simplicities sake) most of my smaller coils are, secondary base feedback via a CT, no current limiting, no oscilator, a high impedence primary, a half bridge drive (with implicit DC blocking) and fingers crossed during direct ground strikes!
So many options!
I have not yet built a SSTC with active primary current limiting and secondary feedback, but its on the list!
The thing I really love about secondary feedback it no tuning, oh the joy!
Anyway all the best, and may your silicon retain its magic smoke.

Frosty90

Tue Oct 07 2008, 10:19AM

Registered Member #1617 Joined: Fri Aug 01 2008, 07:31AM
Location: Adelaide, South Australia
Posts: 139

Hi,

Thanks for the great advice Dylan! Infact when the coil died I did actually have all the things you mentioned for 'safety' (I had a 10A lamp ballast ahead of the bridge to try and limit fault current, I also had a DC blocking cap in series with the primary, a 0.1 uF cornel-dubelier cap), but strangely it was two fets on the same side of the bridge that died, and the other two were fine, I would have expected all 4 to die if one went short from over current, but I was pushing it severly, deliberately alowing ground strikes repeatedly, so it probably would have been alright otherwise, but I have learnt my lesson now!!

Once again, thankyou verymuch for all the help!

Cheers,
Jesse

Dr. Dark Current

Tue Oct 07 2008, 12:45PM

Registered Member #152 Joined: Sun Feb 12 2006, 03:36PM
Location: Czech Rep.
Posts: 3384

Frosty90 wrote ...

...I also had a DC blocking cap in series with the primary, a 0.1 uF cornel-dubelier cap

Oops! That could have been close to resonating with your primary and as you shorted the secondary, primary inductance dropped possibly resonating with your cap and causing huge currents across the switching devices.

Next time use at least 1uF to be safe.

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