Desktop TC Mark II / Projects / Forums

Forums

4hv.org :: Forums :: Projects

« Previous topic | Next topic »

Desktop TC Mark II

Move Thread

LAN_403

Voltwad

Wed Jul 15 2009, 03:46AM

Registered Member #1829 Joined: Sun Nov 30 2008, 01:06AM
Location: Raleigh N.C.
Posts: 74

Hey everybuddy! I've decided to try my hand at another desktop Tesla coil. My first mini coil, which I posted on the third page of this thread was fun to build and I was impressed with it but it just wasn't powerful enough. Those little camera transformers and their transistors can't handle much juice. I decided to start from scratch and so far I think it's going pretty well.

1247625016 1829 FT72933 Thewholemesssmall

1247625016 1829 FT72933 Thewholemesssmall

I've started with winding my own flyback transformer from an old ATX PSU core. It's an EI core, about 1.25 x 1 x 0.5 inches. I wasn't counting but I estimate there's about 200 turns of 30 AWG in the secondary. It's six layers, each wrapped with one layer of kraft paper dipped in wax. The primary is eight turns of, I think it was, 22 AWG. Originally I planned on a single transistor topology, so I also wound a four turn feedback coil but I've since moved to a 555 driver. Later I think I'll wind another transformer without a feedback but for now this one will do. I think it puts out about 1.5 to 2 Kv peak but I need to make a resistor divider and scope it under full power to be sure.

The driver is the standard circuit. The big disc cap is a decoupling cap, there's an RC snubber across the primary and a zener across the gate and source of the IRFP250. The chip is an NE556n because they didn't have any singles at the Radio Shack I went to. So far I've already blown half of one of the chips I bought. That's why I'm using a separate nine volt battery for the driver power and the PSU you can see at the middle top of the first photo for the main supply. I can't seem to keep the 555 chip from overheating. Ultimately I'd like the supply to be no greater than nine volts because I want all the guts to fit into the box I made.

I also made some voltage multipliers. I made several but experimentation shows that two stages seem to produce the best output. They are two strings of three Newark part number 30C6172 capacitors and two strings of six Newark part number 76K2860 diodes. I mounted them on unclad perfboard and soldered the leads together on the backside. This yields a multiplier stage rated for 6Kv DC.

That's it for the supply. Now comes the coil and tank. I started with the secondary coil. I took two knockout plugs for two inch PVC pipe and wrapped five or six layers of kraft paper around them. I could've just used a piece of pipe but I figured the paper would have better performance characteristics. Also, the kraft paper former is thinner than 2" cellcore PVC so the plug forms a lip where it meets the tube allowing me to wind the wire right up to the ends without leaving exposed PVC. The result is, I think, a visually pleasing coil. I soaked the paper tube in polyurethane and let it dry. Then I wound around 400 turns of 30 AWG onto the former. My winding counter messed up so the turns are +/- 20. After several layers of varnish, I secured some 1/4-20 nylon nuts onto the centers of the plugs with epoxy and then soldered the end wires onto some washers that had gaps cut in them with a carborundum wheel to eliminate the shorted turn effect. The washers were then glued onto the ends to make electrical contacts.

Next I made the primary coil. JavaTC said I'd need about nine turns so I got about fifteen feet of 12 AWG solid core and stripped it. I cut and drilled some formers from the same red oak that the box is made of and then sort of screwed the coil into the holes in the formers. I attached the whole thing to the box lid and gave it a few coats of varnish.

Next came the topload. This is two discs of 1/2 inch blue styro insulation, chucked in a hand drill and sanded into a toroid shape. There is a disc of aluminum sheet on top and bottom and the edges are done with aluminum tape. It is secured to the top of the coil with an embedded 1/4-20 nylon screw.

Finally, I did the tank cap. Here you can see the cap in the box I made and some extra plates sitting to the right. It's just cut aluminum sheet and layers of overhead transparency sheet. Tomorrow I'm going to get some waxed paper and put sheets between the plates and transparency to add some protection for the dielectric. I tuned the coil by adjusting the cap value. I used a 2n3055 driven by a ~2 or 3 % duty cycle square wave to simulate the gap and 5 volts for the supply and adjusted the cap to get a maximum voltage on the topload. The system seemed to peak at about 5 nF.

So, left to do; Make a heatsink for the FET. this'll probably have to be mounted on the outside of the box on the side for thermal and spatial requirements. Wind a new flyback. Get a single 555 chip and make a pcb for the driver. Put waxed paper in the cap. mount all this junk in the box.

So far I've been able to get 2.5 inch sparks to a grounded counterpoise but very little in the way of streamers. I figured if I could get sparks like that I should see a little streamer action. Perhaps the light has just been too bright. Anyway, I hope to have some pics of the coil in operation tomorrow.

Here's a more detailed pic of the whole setup. And here's the kitty.

UPDATE

Operating pics. Here's a closeup. The counterpoise is about 1.75 inches from the tip of the breakout point. This is the best I can do so far with 9 volts input.

1247670815 1829 FT72933 Sparks2close

Here's a larger pic with sparks about 1.5 inches long. Here's the larger version of the closeup. In it you can almost see that at 9 volts the supply draws almost 4 amps. Finally, here's a youtube video of the coil in operation.

UPDATE

Here's a picture of the driver board I just completed. This board incorporates a 7805 regulator for the timer supply. I also forgot to use an electrolytic on the breadboard but I included one on this board so there's no longer any heating of the 555 chip although the 'lytic gets a little warm. Also on this board is a snubber for the flyback, a zener diode for the FET, and screw terminals for power input and out to the flyback. I believe this board will work well up to 35 volts input should I choose to up the power.

And here's a picture of the brush discharge with a counterpoise. I still can't get any discharge without one though.

AWESOME UPDATE!

1247798788 1829 FT72933 Streamerssmall

Streamers!

Although this is using my NST and a rather large RQ style gap. Here is a larger pic. To the very left you can see some of the gap made from 1/2" copper pipe end pieces sitting on a piece of tile. To the right of that is the same cap that I have in previous photos although the capacitance has been reduced to compensate for the reduced secondary capacitance. Notice the pretty blue glow around the plates

I think it's around 1.5 ish nF now. Then the coil with the new topload which is actually the topload from my first desktop TC. And finally a variac to power the NST. Here is also a youtube video. So the smaller topload seems crucial to achieving streamers without a counterpoise but this is also with a 1/2 rectified NST. 180 watts. My flyback can push maybe 60 for a few seconds. I'll try the flyback with the new topload in a bit and see how it works.

UPDATE again.

Ok, after this I'm going to bed I promise. I've hooked up the coil with the new gap and topload to my flyback supply and here's the results. One of the prettiest photos so far I think. Now, if I can fit an RQ gap into the box then it'll all fit except for the FET and heatsink. And the FET needs to be outside the box anyway 'cause it gets hot. Like second degree burn hot. Even with a large sink on it. This photo was taken at 13.5 volts with the circuit drawing around 4 amps so about 48-60 watts. The flyback gets warm enough for the wax to start running out of it but I think if I make a small container for it and vacuum pot in wax then it'll get soft but stay in place so that's fine. Also the electrolytic on the driver board is getting disturbingly warm. I think I'll have to find a larger cap. While I'm at it I'll also see if there's a better FET than an IRFP250. Run for more than a minute and I think this thing could start a fire. Finally, I think it's interesting, the voltage should be roughly the same as the NST, only the current is less. And yet, instead of one streamer discharge from the point, there are brushes from all 'round the topload. I would have imagined the streamers would be the same only more intermittent. Perhaps the new supply and the wiring connecting it changed the capacitance slightly and detuned the primary a bit. Tomorrow, more experimenting with the tank cap.

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