would it be pausible to modulate a drsstc with a brick / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

would it be pausible to modulate a drsstc with a brick

1 2 next

Move Thread

LAN_403

teravolt

Mon Oct 02 2006, 03:49AM

Registered Member #195 Joined: Fri Feb 17 2006, 08:27PM
Location: Berkeley, ca.
Posts: 1111

on Richie Burnett's web site he is able to modulate his tesla by changing the voltage at the power supply to the bridge. Now I know a brick such as a CM600HA-24 is designed to work in a on or off state but could it be used for this purpas if the gate is changed say around 3-5v. I spose it would have to be heat sunk well because it would be in a sum what linear regon. Has any body tried this? N.B.

Steve Ward

Mon Oct 02 2006, 05:00AM

Registered Member #146 Joined: Sun Feb 12 2006, 04:21AM
Location: Austin Tx
Posts: 1055

It would probably work, but you're right, the dissipation could be enourmous for a high powered coil. You might want to look at how switch-mode audio amps work. Thats basically the idea, except your output has a 1/2Vin DC offset (so you sweep between 0V and Vin, generally). But building a switcher to work with the non-linear load of a SSTC corona could be pretty challanging i would think. FYI, SSTCs present mostly a constant current sink after a certain length of corona has been created. This implies that the load impedance is not constant.

Steve Conner

Mon Oct 02 2006, 09:12AM

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

Most IGBT brick datasheets I've seen say "Not for linear use". I think it's because the dies aren't guaranteed to share current in the linear region, only in the saturated region.

teravolt

Mon Oct 02 2006, 07:36PM

Registered Member #195 Joined: Fri Feb 17 2006, 08:27PM
Location: Berkeley, ca.
Posts: 1111

So Steve your saying that those bricks are a composet of multiple wafers and not one peace of silicon. there must be a way of modulating a brick maby with some other components to generat a audio signal. maby pulse width modulation with a filter before it goes into the bridge. N.B.

Steve Conner

Mon Oct 02 2006, 10:17PM

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

Correct, they have several small pieces of silicon inside. Here's a picture of the guts of one we destroyed at a UK Teslathon a while back:

(picture from this page:

PWM would probably be the most "correct" way to do it. But, I remember Jason Rollette using an IGBT brick in the linear mode as a current limiter for his railgun charger, and he never killed it.

teravolt

Tue Oct 03 2006, 04:38AM

Registered Member #195 Joined: Fri Feb 17 2006, 08:27PM
Location: Berkeley, ca.
Posts: 1111

I guess I will just have try something out. I would think that each segment is probly cut from the same wafer to enshure uniform current sharing. I almoast think that a 1000w car amp could used to supply the current and voltage all though I don't think that they put out more than 100v. I guess that I am looking at building a 600v car amp. It would be ineresting to modulate a tesla with real audio. Correct me if I am wrong but a CW tesla does not make as much noise as a modulated tesla and a audio based tesla might be clear enough to pass voice. thanks for the input. N.B.

Finn Hammer

Tue Oct 03 2006, 07:12PM

Registered Member #205 Joined: Sat Feb 18 2006, 11:59AM
Location: SkÃ¸rping, Denmark
Posts: 741

A serial regulator element would have to stand both current and power of the coil it regulates. Current is no problem, but power could be, depending on the coil size.
Since you are making a class A circuit, half of the full power is dissipated in the regulating device.

Bricks like the CM600, beng able to dissipate 4100W, will handle up to a large sstc, with sufficient cooling.

As suggested by Steve, on the day I blew that CM600,

(in this coil

)I have studied Source followers, simulated them, and even breadboarded one, and it appears to work fine. But I have never put it to the test on a SSTC, I hope you will beat me to it.

Cheers, Finn Hammer

Reaching

Tue Oct 03 2006, 07:20PM

Registered Member #76 Joined: Thu Feb 09 2006, 10:04AM
Location: Hemer, Germany
Posts: 458

yeah, and steve is right, a igbt brick is made out of several wafers consisting several igbts connected in parallel. heres a pic of an toshiba igbt brick i destroyed a year ago

here you can see 1 igbt of the halfbridge consisting 6 single igbts and 3 antiparallel diodes in a seperate "cluster" the right miniature silicon rectangle is the gate, the collector connection is made out of a copper like material underneath the silicon, on the left you can see the emitter and the smaller wafer is the antiparallel diode made of 3 seperate diodes connected in parallel. yeah interesing

teravolt

Wed Oct 04 2006, 05:00AM

Registered Member #195 Joined: Fri Feb 17 2006, 08:27PM
Location: Berkeley, ca.
Posts: 1111

I'm thinking that you would need a controlable current source good up to about 15khz. in this way it could controle current in to the tesla through the bridge. modulating 500amps is still a tall order not to mension what it would do to zero crossing and other things.
How about this idea what if the dead time on the bridge was width modulated. They say that dead time should be no grater than 5% but how would it afect the primary tank if the dead time was only say 85 % or 50%. Could this be done to reduce power tranfered to the primary sucsessfully? this would be a lot easier to do by modifing the controle electronics

Steve Conner

Wed Oct 04 2006, 09:49AM

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

Teravolt, several people have already built SSTCs and flyback drivers etc. that produce audio by PWM. So we know it works. (I never tried it myself though: I've only ever used FM modulation that gets converted to AM by the resonator.)

As for the deadtime being no greater than 5%: I accidentally fired up my DRSSTC a while back with the deadtime set to around 40%. The PLL tuning control didn't seem to have much effect, and the sparks were small and weedy- about half the usual size funnily enough- but nothing blew up.

1 2 next

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