Agentgates´s TPU setup with strange wavehill hump

[minde4000]

Hi,

I had a piece of pvc pipe and some wire so I made this A in less than an hour.

Primary AWG 22
Secondary AWG 15

As you can see I did not space my secondary yet (it wasn't spaced out in his first pics I suppose that was a working model?)

I ran it with my SG and Lawton 12VDC pwm. Still waiting for optimum pulser diagrams. I can see waves in my scope on primary... you know just a shape changing pulses as you adjust frequency... nothing special. But thats all I have done so far. 12VDC bulb connected as a load stays cold.

Also I would like to mention that PVC tube is 4.5" diam and 4" higth with 0.25" walls. And unfortunately it is not perfectly round and andgle apppears to be some 50-60 degrees. 

Regards Minde

[zerotensor]

The segments of wire on the outside of the cylinder are helical, while the inside segments are straight.  The attached plots depict the difference.  The outside wires hug the surface of the cylinder, whereas the inside wires conform to the surface of a hyperboloid of revolution.

[minde4000]

With 15 AWG 12 turn secondary.

M

[ramset]

Minde4000
If being neat and looking good makes them work that beauty should power your whole block
Beautifull !!
Chet

[agentgates]

Aside from your melted IGBT's, is your current coil producing more power out on the secondary than is put into the primary? What is the load on the output?

Yes, far more than I could supply with the PSU. Please see the last scope shot with the 240V/50W bulb on it. (125ohm DC resistance)

What is the input amps and volts, and what is the output amps and volts.

Input: 15V/ ~1A (ignore the relatively high amp as it is a single shot, FET fired by hand. It will significantly drop when the accelerator circuitry is done and will continously run)

Output: 126V with the mentioned load.

More info on the recent design

When I don't do the full circle with one shot (split the coil to 6 equal parts) you will see the oscillation behind the peak I showed:
http://www.samsonium.org/sites/default/files/P1060254.JPG

And after put them in series back to the original design (equal to 1 primary coil):
http://www.samsonium.org/sites/default/files/P1060255.JPG

So I increased the tube diameter on the next one (larger acrylic) to make more room between the primary windings for the 2 more primary coil. (BTW, acrylic materials arrived for Stefan's coil so I start to build it on his coil and finally send it to him ASAP to make everybody feel better )

@minde4000

Your secondary is not good with 3 turns. That never lighted bulb for me and will never do that. I presented that with the 2.2 ohm resistor on the power drop.

That coil is here and I used another similar design in 2 stages to boost the output:
http://www.samsonium.org/sites/default/files/P1070264.JPG

In other words there was 2 TPU driving each other and 1kV/3A bridge rectifier after every stages and resonant inductors  + foil cap after the last stage. It was driven by IGBT and was lasted 30 sec before the black plastic started to melt on the edge. (I didn't post that pic with the completed secondary, so sorry)

On the new design there is no inductor, foil cap, rectifier, nothing. Only the bulb and the scope.

So wind it as it was mentioned above. BTW that is a nice primary on the photo. (you really need the IGBT to fire that thick copper as it is not a current device, it is a voltage device!)

@zerotensor
WOW! Beautiful work. May I use them later on presentations?

@Mannix
Yes the junction problem will be solved on the next design (I hope I can buy a thinner copper) and I'm not looking to use IGBT or FET on the final design as the power consumption does not require. I'll solve it by even thinner primary (hairline) and use tiny transistors. It seems the thinner primary reduces the heat problem and more goes to the secondary. We'll see when the new 3CH design and the circuit is done.

More in a couple of hours.

UPDATE1

For calculating tube length to the 45 degrees:
http://www.samsonium.org/sites/default/files/formula.JPG

UPDATE2
@minde4000

Yes, that is nice now. Drive it and let us see a scopeshot.