SSTC questions. / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

SSTC questions.

1 2 next

Move Thread

LAN_403

M.A.D.

Tue Oct 18 2011, 01:17AM

Registered Member #4052 Joined: Thu Aug 11 2011, 04:43AM
Location: IN ,USA
Posts: 69

I am building my first SSTC, and am using steve ward's mini design. However I am, at least for now, using a much smaller secondary. According to Javatc it is resonant at 353.98KHz. The secondary in the design is resonant at about 160KHz.

So my question is this going to cause a problem.

Also what is considered too long for the wires going to the GDT.

Polonium210

Tue Oct 18 2011, 10:30PM

Registered Member #3450 Joined: Sun Nov 28 2010, 05:01AM
Location:
Posts: 61

I use a smaller secondary for that coil and got better output. 3"x7" of #30.
Keep your GDT wires as short as possible and twist each pair of wires together tightly and you should have no problems.

genious 7

Wed Oct 19 2011, 01:19AM

Registered Member #2887 Joined: Sat May 29 2010, 11:10PM
Location: Panama City, Panama
Posts: 107

As said, be careful with the gate transformer. Leakage inductance is bad in itself, but more so if the frequency is higher, as in your case. Twist the wires tightly and you should have no issues.

M.A.D.

Wed Oct 19 2011, 03:05AM

Registered Member #4052 Joined: Thu Aug 11 2011, 04:43AM
Location: IN ,USA
Posts: 69

Thank you, but I am still worried about the GDT.

My design has the circuit on 3 boards, and when I created the layout I did not know about this evil leakage inductance. So when I ran out of room on the mosfet board, I put the driver on its own board. The traces from the GDT to the mosfets are short, but the leads going to the driver are about 5'' each, including the traces on the board. Is this too long?

Also, I used 20 awg solid wire to wind the GDT, because I had some laying around. you can barely see the toroid but it is all very tight. Is this wire going to be a problem?

genious 7

Wed Oct 19 2011, 05:06PM

Registered Member #2887 Joined: Sat May 29 2010, 11:10PM
Location: Panama City, Panama
Posts: 107

Can't tell without a photo, but the windings must be twisted for a good performance. Please seen this link, it may help you.

Nevertheless, a more experienced member may guide you better.

Polonium210

Thu Oct 20 2011, 01:02AM

Registered Member #3450 Joined: Sun Nov 28 2010, 05:01AM
Location:
Posts: 61

Here is a picture where you can see how your GDT should look.

I connect the GDT wires directly to the MOSFET's or IGBT's and the wire from the GDT to the driver board is about 5". Hope this helps.

M.A.D.

Thu Oct 20 2011, 04:43AM

Registered Member #4052 Joined: Thu Aug 11 2011, 04:43AM
Location: IN ,USA
Posts: 69

All right, thank you for the help.

My GDT does not look as clean because, as I have already said, I used wire thicker that the 24awg specified in the schematic, but it is still very tight.

I will probably be fine with how it is. However, if this does go horrifically wrong, what size of fuse would be good to prevent the enormous mains filter cap from turning the entire board into a bon fire. I am not actualy sure how much amperage the bridge should pull so I went overkill on the mains stuff, but how much power should I expect it to draw?

M.A.D.

Fri Nov 11 2011, 11:24PM

Registered Member #4052 Joined: Thu Aug 11 2011, 04:43AM
Location: IN ,USA
Posts: 69

All right, so I finnished this coil and fired it up. After a few minor problems I had it running. This is where the major problems started. The MOSFETs ran hot even with the interupter set on a low power. Also the MOSFETs heated unevenly and after after a while the fuses blew and the MOSFETs tested as blown. They did not explode or catch fire, they just died.

I am planning on redesigning the bridge section. after doing many hours of research I have decided to use all of the optional diodes for protection, and use twisted wires instead of board traces to go between the GDT and the gates.

I had used wirewound resistors on the gates, not even thinking about their potential inductance. what would be a better type of resistor to use?

Also any other sugestions would be greatly appreciated.

Daniel Kramnik

Mon Nov 14 2011, 12:22AM

Registered Member #3885 Joined: Sun May 15 2011, 12:47AM
Location: Newton, Massachusetts, United States
Posts: 94

Normal SSTCs shouldn't be that sensitive to leakage inductance. Did you scope the gates of your MOSFETs to see what the GDT was putting out? I would go back and make sure that leakage inductance is your problem before redesigning your entire coil.

The more likely alternative is that you used the wrong number of primary turns - did you adjust the number of turns as you were experimenting, or did you just pick the number that Steve used? The general procedure for determining the right number of primary turns for an SSTC is to start with too many and then remove turns until just before your MOSFETs start heating too much.

Pictures would help.

Good luck!

M.A.D.

Mon Nov 14 2011, 08:19PM

Registered Member #4052 Joined: Thu Aug 11 2011, 04:43AM
Location: IN ,USA
Posts: 69

I unfortunatly do not have a scope, so I am completely in the dark as to the wave forms. So since it is impossible for me to specifecally identify the cause, I thought it would be best to just eliminate everything potentialy problematic.

My coil is a lot smaller that Steve's and the primary is inside of the secondary, so to make up for the small diameter (2'') I used 18 turns, this is equal to about 6 turns on steve's coil. I had decided to start with a little higher inductance than the original design for saftey. However it is possible that this was not enough so I will try more turns.

I do believe however that it is still necessary to redesign the board that the bridge is on. Between the traces going to the gates and the wirewound resistors, I probably have a lot of stray inductance. Also the gate on one of the MOSFETs had failled so I think adding a lot of protective diodes is worth the effort.

I am srry for no having more information about my problem, however my equipment is mostly abstract (nonexistant).

1 2 next

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