Solid State Disruptive Coil Spark Gaps / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

Solid State Disruptive Coil Spark Gaps

first previous 2 3 4 5 next last

Move Thread

LAN_403

Steve Ward

Sun May 14 2006, 05:42AM

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

Hey Conner McSteve,

a thought just popped into my head... how does your OLTC perform if you let it ring out for many notches? Is there a major hit in performance? Terry's "spark gap" is basically a simplified OLTC that doesnt have the ability to quench, so it would be easy to set one of your OLTCs like this and compare performance.

Terry,

We can see that your switch works just fine as a switch, but until you test it to make sparks, we will never know for sure how it works in a TC. Your test cant make assumptions about streamer formation, since we dont completely understand that part yet, so someone will just have to test your switch and see how it "really" performs.

I still prefer the "complex" DRSSTCs, much better use of the silicon i think.

HV Enthusiast

Sun May 14 2006, 12:41PM

Registered Member #15 Joined: Thu Feb 02 2006, 01:11PM
Location:
Posts: 3068

In regards to quenching, it may not be necessary with a high efficiency solid state switch. Remember, that in a conventional Tesla coil, quenching is necessary as energy transfers back and forth between primary and secondary and a considerable amount of the energy will be lost as heat due to losses within the spark gap. In a spark gap coil, we would like the spark gap to quench (usually within a few notches) to reduce these spark gap losses and keep most of the energy on the secondary side. However, since there is no spark gap or source for high losses with a solid state spark gap, quenching may not be necessary.

Of course, if the solid state spark gap turns off too soon or at a point where energy is at a maximum on the primary side, that could reduce performance.

But as was mentioned, lots of experimentation still needs to be performed to see exactly how this new circuit works, what effects no quenching has, and so far . . .

Steve Conner

Sun May 14 2006, 01:15PM

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

Hi Ward McSteve,

The OLTC really worked much the same if I let it ring out. The spark output was no different as far as I could tell. The primary circuit did get a bit hotter though due to more losses, and the charging voltage tried to kick up higher due to boost converter action, what with the switch being on longer.

Better use of the silicon is all very well, but "pretty is what works"

It seems to me that there's still an entry barrier to DRSSTCs due to them being hard to understand, although Steve Ward has done a lot to lower that with the simple feedback driver and ready-made PCBs. Terry's new thingy is "Just like a spark gap but solid state", and it comes with the trusted Fritz brand name

so folks will be comfortable with it.

I think it's kind of like the Chevy vs. riced-out Honda debate. The DRSSTC is the "Hot import" of Tesla coiling. You can make that Honda go faster than a Corvette, but it takes a lot of work with strange new technologies that can go wrong spectacularly. There may not be a clean answer, and there will always be plenty of champions on both sides.

I believe a DRSSTC can give about 4ft of streamers from four TO-247 IGBTs. I personally got 36", and trashed the IGBTs in the process, but I think others have done better since. I was running 600V devices on 400V DC, and I don't think anyone has ever tried TO-247s on a 600V DC bus, which would be the fair comparison since Terry's stack uses the 900V devices.

If you can't get 4ft from four TO-247s, there's always the OLTC record, 6'9" with 4.2kW :P If you can beat that (either directly, or scaled, by achieving bigger Freau and Watts numbers) we'll know the higher voltage tank circuit is more efficient than my 1kV system was. Break out the MOTs Terry!

HV Enthusiast

Sun May 14 2006, 02:41PM

Registered Member #15 Joined: Thu Feb 02 2006, 01:11PM
Location:
Posts: 3068

wrote ...

Better use of the silicon is all very well, but "pretty is what works" It seems to me that there's still an entry barrier to DRSSTCs due to them being hard to understand. . .

Other than some of the finer points and details regarding the DRSSTC, i don't believe the DRSSTC concept is hard to understand at all. The concept is quite simple and really not at all more difficult to understand than the conventional spark gap coil, although the implementation using electronic devices makes them a bit more complex.

wrote ...

I believe a DRSSTC can give about 4ft of streamers from four TO-247 IGBTs. I personally got 36", and trashed the IGBTs in the process, but I think others have done better since.

Correct. I have run my DRSSTC II system using 600V TO-247's as a drop in replacement for the 40N60's and have achieved output arcs exceeding 7 feet with these. There really is no difference between the TO-247s and ISOPAKs other than thermal properties (assuming identical dies). If you can cool a TO-247 properly, it should perform no better or worse than the same die in a SOT-227 package. I don't know why everyone believes these SOT-227 devices have some sort of magical property to work more profoundly than a TO-247. Its just thermal impedance, nothing more . . . If you're smart enough to develop a system to provide adequate cooling, that the TO-247 will work just as well.

Terry Fritz

Mon May 15 2006, 04:39AM

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

Hi,

There is a new paper here where I try to "tell all" in one single document:

Copy and past the link I guess... Or go here...

Going to take me a few years to figure this "forum" stuff out

This server does not have nuclear terrorist defenses so it should be easier to get through to.

It sounds like Dan might get the first "real" system going! I wired up a quickie with the two modules I had. It is not optimal at all but it did make sparks! I made a little 20 meg movie:

I need to get boards on order. and a few other parts... 20 IGBTs will be here Tuesday. I am still working on heatsinking.

how does your OLTC perform if you let it ring out for many notches? Is there a major hit in performance?

Remember that the OLTC is a fairly high loss system. Low voltage and high current. The SISG thing is high voltage at lower current so the losses are far less like in a conventional coil. The primay "Q" is proportional to Lpri so this system can support far higher primary inductances for far higher "Q". However, I did run it into like 650nH/600nF here for high current testing which is similar to the OLTC case and the resistance is about 0.050 ohms:

Dan explains that now the gap losses are "gone" so running a coil with many notches should not be a problem. Do note that if you run the primary without the secondary (or out of tune), the caps might explode now since the firey spark gap can no longer be counted on to burn up the power!! Also, many notches drives up the primary RMS current a bit. System losses are sort of "all different" now.

I believe a DRSSTC can give about 4ft of streamers from four TO-247 IGBTs

One should check out this fire breathing IGBT in the SUPER TO-247 package!

Junction to case is 0.20 C/W. 1200V, 120/240 amps and 595W!! I have run them to 800 amps and they seem to "like" that!!

Rell.com has like 7000 leaded versions and is trying to sell them for like $8 each!! But a part broker might have already grabed them... They normall go for about $15. Mounting them to a heat sink is a challenge though. At 600W that would be water cooled copper! Fortunately in the SISG case, just a little heat sink is fine.

I still prefer the "complex" DRSSTCs, much better use of the silicon i think.

There is one big advantage the SISG might have. It should be very hard to break. It is "made" out of over voltage devices and everything is very well protected. I think the circuit can take a full streamer hit without damage. That alone might be its claim to fame in the commercial coiling world.

Of course, it might pave the way for future stuff too. Looke what Jimmy and all did with the OLTC!!

All the small testing is done now. Mark, Dan and me are ordering parts and boards for a full system, so we'll see how it goes!

Cheers,
Terry

Late news...

I have updated the master "tell all" document at:

http://drsstc.com/~sisg/SISG.pdf

I will try to keep it "updated" as time allows... A little slow now while I wait for parts...

I now have all the parts on order for a ~250W dual MOT coil. Should be going by next weekend... Mark or Dan will probably have their's going sooner

Pretty good for "idea - to final working coils" in three weeks!!

Peter mentioned "The reason I worry about the quenching is that with my SIDAC string in place of a spark gap, sparks were only 2 inches with 5kV input."

My 14.6 watt test at 1.8kV got to 5 inches with a 150nF primary cap

I think the SIDACs can take about 100 amps peak for "one" cycle and then they seem to conk out or go highly resistive. That was the problem with your test. Many nights I stared at your site wondering "why" it did not "work"... I think I know know... SIDACS just can't pass the primary current for any length of time... But they can "trigger"

If this all works out, as it now appears it will, it will be real cool

But what is even cooler is all the extensions and new stuff this could introduce!! People already "got plans" for sync triggering and quenching... Even talk of a "PIC" that could detect "situations" and act "appropiately"!! We can only hope that a giant thing may come of it like the DRSSTC did for solid state coils!!! Plans of even a Wysock model 13 class SISG gap

Some think that with all this modern stuff, that Tesla coils have gotten boring. But really is is just a 100 year old technoly "catching up" to the modern world!! It sure is hard!!! But it is coming very fast now

I look forward to arcing the next coil to all the SISG circuits to see if I "can" blow it up

There is a chance a negative streamer could mess up the gate voltage in the negative direction delaying or turning off the IGBT at a bad time... Probably should use "uni-directional" TVSs

) If it gets to the point where one "can't" blow it up, that is a real big step in this sport!!

Cheers again,

Terry

Dr. Drone

Tue May 16 2006, 09:52PM

Registered Member #290 Joined: Mon Mar 06 2006, 08:24PM
Location:
Posts: 1673

coaster_chris

Wed May 17 2006, 08:56AM

Registered Member #359 Joined: Sat Apr 01 2006, 09:27AM
Location: Utrecht, The Netherlands
Posts: 16

That's a job very nicely done Terry!!

The scope readings show good effecient results, and now seeing the movie of it al working...
SWEET!

!

Chris

Terry Fritz

Mon May 22 2006, 06:36AM

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

Hi,

Today I finally got everything working!!!

Firing at 3600V 150nF 120BPS 117W

It could arc to ground 23 inches!! 23 inches at 117W puts the Freau number at 2.13!! I ran out of time today to do any other tests but everything seems perfect!! It could probably arc further if I tried to tune the coil in and everythig

The IGBT sections seem to be working perfectly!! I wrote up the latest details in the ongoing PDF file here:

Mostly pictures of all the stuff. Here is a big picture of the final gap:

Here is an arc (15 inches):

There is a newer movie here done a few days ago (14 MB):

The MOT charging system was messed up so I was working on that the last two days.

I hope to get it going at the full 5400 volts tomorrow. Maybe I can get 35 inches out of it at 262 watts

Cheers,
Terry

ragnar

Mon May 22 2006, 11:22AM

Registered Member #63 Joined: Thu Feb 09 2006, 06:18AM
Location:
Posts: 1425

Terry! That's just... bloody brilliant!

HV Enthusiast

Mon May 22 2006, 11:29AM

Registered Member #15 Joined: Thu Feb 02 2006, 01:11PM
Location:
Posts: 3068

Awesome Terry. I haven't got around to getting my set-up running yet as i've been busy finishing up a few other things, but will get to it soon.

Dan

first previous 2 3 4 5 next last

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