First SSTC's first light! What tweaks/refinements are possible? / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

First SSTC's first light! What tweaks/refinements are possible?

1 2 3 next

Move Thread

LAN_403

aonomus

Sun Sept 28 2008, 03:50AM

Registered Member #1497 Joined: Thu May 22 2008, 05:24AM
Location: Toronto, Ontario, Canada
Posts: 801

For the last 3 weeks I've been on and off building Steve Ward's Mini SSTC5

Right, so after a frustrating day of accidentally blowing my halfbridge mosfets, I fixed the problem (one loose wire from the GDT = avalanching power through one mosfet that fails as a short, avalanching through the other mosfet), and ran my SSTC from 30VDC off of a benchtop power supply.

Anyways, my questions are how to give this project its finishing touches, more specifically a good breakout point ontop (apparently galvanized screws and solder do not work), and also if there is a way to prevent corrona from forming on the feedback antenna? Any other finishing touches that would either improve performance or just make it look cooler? I'll take more photos tomorrow when there is decent lighting, but the body of the SSTC is made from MDF and dimensional lumber.

From the pictures below, you can see that:
1. I do get breakout, however I am also loosing a crapton of energy through the little bend where the wire goes onto the plexi, I will varnish that several times to fix that problem.
2. You can see just *slightly* some corrona forming on the feedback antenna, its some scrap enameled magnet wire from a unwound torroid, will extra varnish/insulation fix this? What happens if I get a full fledged strike to the antenna?

(Click the photo, you won't be able to see the feedback antenna's corrona at this size)

MRacerxdl

Sun Sept 28 2008, 05:37PM

Registered Member #989 Joined: Sat Sept 08 2007, 02:15AM
Location: SÃ£o Paulo, Brazil
Posts: 476

If you use the Clamp diodes, its not a problem the corona at Antenna, once, when I was testing it opened an arc between the top of load and the antenna, nothing happens, only the wire gets burned.

The better thing to stop corona from other points is getting a topload with a breakpoint, not direct a breakpoint. Also topload increase sparks.

uzzors2k

Sun Sept 28 2008, 07:05PM

Registered Member #95 Joined: Thu Feb 09 2006, 04:57PM
Location: Norway
Posts: 1308

Making a little loop on the end of the antenna and moving it further away should decrease the amount of corona. For your next coil bring the secondary wire in through the side of the former, and cover the bend with lots of epoxy.

aonomus

Mon Sept 29 2008, 12:00PM

Registered Member #1497 Joined: Thu May 22 2008, 05:24AM
Location: Toronto, Ontario, Canada
Posts: 801

Topload and breakout point blew another set of MOSFETs + fuse. I think I'll have to adjust the primary to have a longer off time and shorter on time....

MRacerxdl

Mon Sept 29 2008, 10:50PM

Registered Member #989 Joined: Sat Sept 08 2007, 02:15AM
Location: SÃ£o Paulo, Brazil
Posts: 476

That can be much current consuming, how much turns have your primary?
I forgot to mention, that topload also increases the current.

uzzors2k

Tue Sept 30 2008, 12:25PM

Registered Member #95 Joined: Thu Feb 09 2006, 04:57PM
Location: Norway
Posts: 1308

aonomus wrote ...

I think I'll have to adjust the primary to have a longer off time and shorter on time....

? Do you mean the interrupter or actual primary? By tuning the primary you can set the coil to draw pretty much any current at any voltage.

aonomus

Tue Sept 30 2008, 08:07PM

Registered Member #1497 Joined: Thu May 22 2008, 05:24AM
Location: Toronto, Ontario, Canada
Posts: 801

Following some discussion on IRC about what things I could do to reduce the coronna on the wire leading up to the breakout point, Steve McConner suggested I use a open ring of copper pipe (ie: a circle with a small gap) and place it just above the offending point where more breakout likes to occur. Anyone have any thoughts on this? My only fear is that adding the copper pipe (in theory the copper pipe is smooth enough to prevent breakout, so all the energy continues onto the breakout point at the top) will act like a topload itself, and then cause another halfbridge failure.

Also, a more powerful example of the broken insulation is seen here:

Pushing the SSTC to higher levels with a breakout point works, I only need to be conservative as to how high a voltage I can apply to the primary (before MOSFET breakdown). What I've noticed is that running off a benchtop power supply that can only produce 30VDC, the energy gets recycled back into the smoothing cap where I've seen it run up to 40VDC, and theoretically with my variac's range of 135V max, I might exceed the 200VDC rating of the MOSFETs (which might have been the cause of some failure early on).

CT2

Tue Sept 30 2008, 09:24PM

Registered Member #180 Joined: Thu Feb 16 2006, 02:12AM
Location: Ontario, Canada
Posts: 187

I would suggest the copper ring, I did the exact same thing for my coil and it worked perfectly. It will have some "topload" like behavior but it shouldn't destroy your bridge. If you are getting too much current try adding more turns to your primary, this will lower the current, as you increase the voltage you will most likely HAVE to do this. I would recomend not running it without that copper ring steve suggested, the breakout at the point where the wire bends is hot stuff, it will either burn your wire or melt your coil form... bad stuff.

EDIT
This is how I did my ring

. The first one is just 14 AWG wire bent into a ring, the second one uses copper pipe.

MRacerxdl

Tue Sept 30 2008, 10:21PM

Registered Member #989 Joined: Sat Sept 08 2007, 02:15AM
Location: SÃ£o Paulo, Brazil
Posts: 476

If your mosfets are for 200V, you can run perfectly a 127 Mains retified, I was runned IRFP250 direct from mains, at Full Wave Smoothed Supply (150-170VDC) it doesnt make anything to the mosfets.

aonomus

Wed Oct 01 2008, 05:34AM

Registered Member #1497 Joined: Thu May 22 2008, 05:24AM
Location: Toronto, Ontario, Canada
Posts: 801

So after blowing a total of 10 (!!!) IRFP250N's, I tried replacing them with the MTW32N20E (more current, slightly slower) and changing the 555 to give a 10 uS on time, 20 uS off time pulse. With the slower pulse and slightly better MOSFET I've gotten some more power out of the SSTC (read: I can run a topload with enough power to cause the TV nearby to turn itself on) without blowing anything up (yet).

The one major recommendation I can give to anyone making their first SSTC: use panelmount breakers! (not the pushbutton type, actual toggle type which have both the bimetallic strip for slight overcurrent, and magnetic trip for spike/short overcurrent). While I am doing bad things by bypassing the fuse in my variac (fast blow is too fast for the breaker to trip), the time delay between fuse and breaker is negligible so long as I don't trip it too often.

Major variations from Steve Ward's design:
Secondary: 4" diam (instead of 4.5")
MOSFETs: MTW32N20E currently
Tank caps: 2x 1uF 1000V caps ($2 electronics goldmine) instead of 0.68uF

Edit: Note that most of the spark photos are taken at low power (20-30%) because it was about 1AM and I'm assuming people mind the noise, as well as the fact that TV's turn themseves on in this things presence...
Edit 2: I totally used seconds instead of microseconds.... a 10second pulse would fry any mosfets and blow the breaker yet again.

Photos:

Burning steel wool

Photoflash tube from a camera, overheats somewhat quickly so not recommended for prolonged usage

Aluminum foil wrapped over the breakout point, as it melts the point becomes rounded and breakout becomes harder. This eventually burned out (yet another) set of MOSFETs

1 2 3 next

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