OLTC project / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

OLTC project

1 2 3 4 next last

Move Thread

LAN_403

Jrz126

Sun Jun 18 2006, 02:51PM

Registered Member #242 Joined: Thu Feb 23 2006, 11:37PM
Location: Erie PA
Posts: 210

So I have one of these IGBT's here and I want to make an OLTC.

Now I need to figure out how to drive it and whatnot... I've been reading up on it at Steve Conner's site. I was going to use 20V for my gate voltage (or should I over drive it?). So at Vge=20, the gate charge is 6500nc. Steve mentioned that you want a turn on time of 0.5uS, is this determined by the igbt or just a good rule of thumb to use?

oh yeah, and how do I determine the best coil size and whatnot?

I appologize for all of the noob-ish questions, but we all have to start somewhere...

Dr. Shark

Sun Jun 18 2006, 06:23PM

Registered Member #75 Joined: Thu Feb 09 2006, 09:30AM
Location: Montana, USA
Posts: 711

That's a big brick, you might want to gather some experience from "cheaper" silicon before you give that one a shot.
If you intend to drive the brick within spec, you will be fine with the 20V on the gate, it is only if you want to push higher currents that a higher gate voltage helps.

The 0.5us is basically a funcion of your resonance frequency. IF you build a tiny high frequency coil, you want 100ns, whereas a huge beast might get away with 1us.

Best Coil-size fol OLTCs traditionally is: The lower the resonant frequency, the better, so you want 2000 turns of fine wire, and make it as big as you like. The IGBT you got should be in for several kW of power.

Extreme Electronics

Sun Jun 18 2006, 08:41PM

Registered Member #74 Joined: Thu Feb 09 2006, 09:17AM
Location: Nottingham UK
Posts: 99

Hi,
Thats a really nice brick !! but its be a shame to blow it...

I really suggest you start with a smaller OLTC first. to get the "feel" of OLTC's

You can build a nice OLTC with a single small $10 IGBT that will give 1' sparks and wont cost you the earth when you kill it.. take a look at 4pf50's tha are well proven in OLTC service.

take a look at

the "on" time is dependant on your quenching time, which is dependant on your coils frequency and coupling. Basically the IGBT needs to be on for as long as it takes to transfer all of your primary power to your secondary, and no longer to prevent over heating.

You Will need to overdrive even a small IGBT in OLTC service, a small OLTC can easilly have 400A primary current. Fortunatly IGBT's can be overdriven to around 2x their pulse rating with 25V gate drive. so a single 4pf50 can take a pulse of 400A.

Bigger OLTCs get much more complecated, as Steves site will show you, but most of that electronics is to protect the brick from over voltage & over current.

for a guide to coil sizes etc take a look at my OLTC calculatior

I cant claim its accuracy, but its will give good ball-park figures.

Derek

Jrz126

Tue Jun 20 2006, 12:01PM

Registered Member #242 Joined: Thu Feb 23 2006, 11:37PM
Location: Erie PA
Posts: 210

I was thinking about building a smaller oltc, but I was thinking that it would be alot harder to blow this brick compared to a smaller TO-220 type igbt.
I've been reading up on how to use the IGBT bricks here Looks like there is quite a bit of good information there

Now for the secondary, do I want to keep the 4/1 H/W ratio? and whats a good resonant frequency to shoot for? I'm also planning on using 26 gauge (.404mm) wire, and probably an 8" cardboard concrete form.

Dr. Shark

Tue Jun 20 2006, 12:24PM

Registered Member #75 Joined: Thu Feb 09 2006, 09:30AM
Location: Montana, USA
Posts: 711

Blowing big IGBTs can be very easy. E.g. if the gate is not protected properly, and switching with too high dI/dt causes a voltage spike, it is dead. Get some TO247 devices to play with, they are a good compromise.

Short and fat would be a good idea for an OLTC, even something like 2:1 ratio. .4mm wire would give you almost 2000T, so it sounds good.

Steve Conner

Tue Jun 20 2006, 01:37PM

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

Meh. Calling a certain kind of Tesla coil obsolete doesn't really achieve anything. They're all variations on a 100 year old theme as far as I'm concerned, and hardly going to net you a Nobel Prize. Having said that, I admit the OLTC doesn't perform as well as the DRSSTC.

Jrz126

Tue Jun 20 2006, 05:56PM

Registered Member #242 Joined: Thu Feb 23 2006, 11:37PM
Location: Erie PA
Posts: 210

Well, I've only got one of these bricks, so it kinda limits my options.

If I use the 26 gauge wire, I'd need a height of 34" to get ~2000T. Should I make the coil diameter pretty wide? (they also had 12" concrete forms).
Is cardboard a good material to use?
In general, do I want to shoot for a lower or higher resonant freq?

Dr. Shark

Tue Jun 20 2006, 06:46PM

Registered Member #75 Joined: Thu Feb 09 2006, 09:30AM
Location: Montana, USA
Posts: 711

Don't let Dan discourage you, I think he is joking. With the recent interest in Terrys Solid State Spark Gap, I think OLTCs are better than ever. Of course you can buy longer sparks with a DRSSTC, but it is a hell of a lot more complicated!

Making the coil diameter wider is going to give you more inductance per foot of wire, so you want to make it pretty wide. The lower the frequency, the better, as it allows you to use a bigger primary capacitor for more bang energy.

Terry Fritz

Tue Jun 20 2006, 08:14PM

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

Hi,

So I have one of these IGBT's here and I want to make an OLTC.

Sweet!!! 20V should be fine on the gate unless you want to drive like 10,000 amps thorugh it, then go to 25 volts

0.5uS is fine. Mostly determined buy the resonant frequency of your coil and how fast the initial current will rise. You don't want to turn on "too fast" or the Ccg capacitance could over voltage the gate. But put a big 30V TVS across the gate to emitter anyway.

You want a big primary cap and small primary inductance. It is hard to get less than 1uH for the primary inductor so the biggest cap you can get. That usually drives down the secondary frequency making a secondary coil with lots of turns for high inductance. Be careful not to make the secondary coil too lossy though!! My 'first' OLTC suffered from high secondary losses since I got too carried away trying to make it 120BPS at 300V. You could go to like 900 to 1000V pretty much solving all that if you have the power supply to get that voltage. Or was your IGBT 2400V... if so you can really go to a higher voltage!

Everything begining with OLTC here:

Is about my 'first' one. But Steve's and latter ones or all modern with more of the details and problems figured out. But you can quickly click through the about just to check.

I don't think you "can" blow that brick. But start out running at lower voltage just to get everything tuned in before running at high power. Low break rates (like once every few seconds) also help setting up.

Now for the secondary, do I want to keep the 4/1 H/W ratio? and whats a good resonant frequency to shoot for? I'm also planning on using 26 gauge (.404mm) wire, and probably an 8" cardboard concrete form.

Sounds fine! Probably in the 50 - 100kHz range. See this for formulas and the JAVA on-line tool:

You wants lots of secondary peak voltage and lots of capacitor stored energy (1/2CV^2). Higher voltage really helps stored energy so you can use a smaller cap for higher resonant frequency and a more normal secondary.

I never thought about it till now, but the SISG drive circuit could run the single big high current brick OLTC too!!! In that case, you don't need nothin for a drive circuit

)

You may want to check this out:

I hate to recommend something "new and never tried" to a new coiler though. But I think you are an engineer so you know the usual electronic details. The SISG drive is probably far more reliable and easy than any conventional OLTC drive. We won't even mention the complexity of DRSSTC's

Newer DF-DRSSTC and SISG circuits will probably make them obsolete

And the OLTC will always be the "daddy"

But the DRSSTC is definitly the "thinking man's" machine!!! Once one has mastered the DRSSTC, everything else is "simple"...

If the SISG drive does not work, you are only out like $5 and 15 minutes of work

But it should work fine. I guess it would be a SISGOLTC

It would be the simplest "big" coil out there in that case with only "one" SISG section.

Be sure to use a minimum of copper too!! A "ten foot" length of 1/4 inch tubing is running $16 here!!!!

Cheers,

Terry

Marko

Tue Jun 20 2006, 08:17PM

Registered Member #89 Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145

Don't let Dan discourage you

OK you mean steve.. ? :)

Actually it seems that oltc's are indeed quite outperformed by DRSSTC's.

Looking at steve conner's OLTC 1 and 2 small one barely outsparks it's secondary length in the end.
If you are building an OLTC I guess you should make it huge. You want lowest possible Fres for biggest possible bang energy with single primary turn.

Conner's big one made 6 foot sparks(if I remember) with whole mess of toroids, from the pic it seems somewhat more than 2 secondary lenghts.

Ward's DRSSTC1 makes equal sparks that look much more impressive from many times smaller coil.

That brick will make a quite good OLTC since you already have it, but then you will figure out that people make longer sparks with DRSSTC's based on some 40N60's, even if you draw out 'maximum' of that IGBT and the coil itself.

I guess that's what Steve wanted to say...

2500V IGBT also adds room for bigger sparks, if you jump from 1000 to 2000V cap voltage you get four times more stored bang energy.

Good luck with your coil...

1 2 3 4 next last

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