IGBT/MOSFET switching times required for SSTC? / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

IGBT/MOSFET switching times required for SSTC?

Move Thread

LAN_403

Kiwihvguy

Wed Dec 07 2011, 09:59AM

Registered Member #3395 Joined: Thu Nov 04 2010, 08:42AM
Location: Christchurch, New Zealand
Posts: 193

I will be constructing Steve Ward's mini SSTC soon hopefully and was thinking of buying some powerful (but pricey) IGBTs and MOSFETs for the H-bridge because I want to input the TC primary with a voltage doubler to gain ~500vdc via variac. (like what Kizmo did)

I was thinking of buying either 48N50s (500v 48a) 44N50s (500v 44a) , 18N120BND (1200v 54a), 50N60 (600v 80a).
I think all of their switching times are fast enough, however what are the ideal switching times required for MOSFETs/IGBTs and how can you determine that in the semi-conductor pdf?

Thanks for your help, just educating myself in advance before I build it

Dr. Dark Current

Fri Dec 09 2011, 12:03PM

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

Basically you want the total switching time of the transistor (td(on)+tr+td(off)+tf) to be only a fraction of the cycle. Usually I won't go over 10% of a cycle for the transistor to spend during switching. So the "ballpark" maximum frequency you can use a given transistor at, based on my assumption would be: fmax=1/(10*tsw), where tsw is the total switching time.

PS: Use the switching times which are given in the datasheet for the operating temperature (usually 100-120 Â°C junction temperature (Tj)), the manufacturer usually also states switching times for a Tj of 25 Â°C, but these values are of no real use.

Goodchild

Fri Dec 09 2011, 07:56PM

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

I would say the 10% margin is mainly for soft switching with ZCS or ZVS. If your going to hard switch it can be slower than mud and probably still have about the same losses. This is because no matter how fast you switch you are still going to be switching with some current flowing in the switch.

Although I do agree with the 10% margin as a general guideline. If you keep it say <200KHz most of those switches should do fine in a SSTC configuration.

Kiwihvguy

Sat Dec 10 2011, 07:18AM

Registered Member #3395 Joined: Thu Nov 04 2010, 08:42AM
Location: Christchurch, New Zealand
Posts: 193

I will be building the mini SSTC as I mentioned before, so when you mention the "total switching time of the transistor to be only a fraction of the cycle", the cycle is as in the a cycle in the frequency that the SSTC resonates at?

It says on the schematic page that the 555 IC oscillates at about 200kHz or so. Is the 200kHz figure the figure I need to use to calculate whether the switching times are suitable?

This is my first TC project, so I'm kind of new to this. Your help is much appreciated.

Dr. Dark Current

Sat Dec 10 2011, 09:25AM

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

Are you looking at the final mini SSTC schematic, at the bottom of the page? I think it doesn't use a 555 as an oscillator, just an interrupter. The frequency is determined by the resonant frequency of your TC.

Kiwihvguy

Sat Dec 10 2011, 10:32AM

Registered Member #3395 Joined: Thu Nov 04 2010, 08:42AM
Location: Christchurch, New Zealand
Posts: 193

Yeah, sorry, excuse my mistake. I meant the 555 as the interrupter. Anyway, if the preferable switching time is to be determined by the resonant frequency, do you sort of have to blind-guess or estimate the best transistor ( if buy them before making the the coil and calculating its resonant frequency)? I know the the smaller the coil, the higher the resonant frequency but are you saying there aren't many ways of determining the best switching speeds required before making it?

I'm getting the idea that I should calculate the best switching speeds for a transistor after I make the secondary and calculate it's resonant frequency with a signal generator and a scope. Please correct me if need be. Otherwise, thanks for the help!

Dr. Dark Current

Sat Dec 10 2011, 02:30PM

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

Usually you first decide about the size of your coil and its resonant frequency, then based on the calculated frequency you select your transistors. You don't need to wind the actual coil to measure its resonance, JavaTC is usually very accurate in calculating the unloaded resonant frequency.

Kiwihvguy

Sun Dec 11 2011, 12:15AM

Registered Member #3395 Joined: Thu Nov 04 2010, 08:42AM
Location: Christchurch, New Zealand
Posts: 193

In the final schematic of the TC, it has fixed dimensions for the secondary size but can I construct a bigger secondary? I'm looking for decent output from a half-bridge.

I tried JavaTC but I gave because it was so complicated for me, I need somebody else to do it for me.

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