A small DRSSTC with a Coke can sized secondary and an Fres of only ~260kHz ! / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

A small DRSSTC with a Coke can sized secondary and an Fres of only ~260kHz !

previous 1 2 3 4 next

Move Thread

LAN_403

Dr. Dark Current

Wed Aug 15 2012, 10:43AM

Registered Member #152 Joined: Sun Feb 12 2006, 03:36PM
Location: Czech Rep.
Posts: 3384

I'm surprised the caps work well, those small leaded HV ceramics (Z5U/Y5U type usually) have a dissipation factor as high as 5% which is enormous... Maybe the increase in arc length you see is because of the caps heating up quickly?

plazmatron

Wed Aug 15 2012, 12:04PM

Registered Member #1134 Joined: Tue Nov 20 2007, 04:39PM
Location: Bonnie Scotland
Posts: 351

Killa-X wrote ...

So how are the caps connected exactly? Your adding capacitance to it, correct? I'm guessing in series with the topload and top wire output?

.....I mean..is there anything wrong with leaving them as-is and just putting them in wax / mineral oil?

The MMC is connected in parallel with the secondary. i.e one end is connected to the top-load end, and the other is connected to ground.

This coil uses standard wire ended caps, soldered in series, and potted in paraffin wax.
The discussion about stripping off the epoxy and making a stack, is a thought experiment geared towards building really small DRSSTC's, which would necessitate to building of a more compact MMC, with an enormous capacitance.

Dr. Dark Current wrote ...

I'm surprised the caps work well, those small leaded HV ceramics (Z5U/Y5U type usually) have a dissipation factor as high as 5% which is enormous... Maybe the increase in arc length you see is because of the caps heating up quickly?

Me too! Little disc ceramics do tend to be a poor choice for Tesla coils of any description. Having said that, the string is of 14 devices, and there is a lot of metal and epoxy in there contributing to the heated volume, and the average power is very low. But I have to say when these tiny things turned up from China, I was very apprehensive about their performance, there really is very little voume to deal with any heating.

I have not measured the temperature of the caps after a run, however I do know the heating is not enough to soften the wax, since the top of the MMC is visible when you remove the toroid.

Re: the change in arc length, I'm beginning to think it may simply be a feedback or interference issue to which the large Csec contributes. The coil was built upon a previously used DRSSTC test-bed which is in an unshielded plastic case which just isn't a good idea anyway, and I had to increase coupling in the secondary base feedback transformer to get it to oscillate at all, and this could probably do with being increased some more.

I am just waiting on a tiny 12v mains transformer, and the electronics are going to be shoehorned into a small metal case, so we shall see if performance improves any.

Incidentally, for those who wanted to know, I put a couple of bare discs in concentrated Nitric acid over night, which cleaned off all the remaining silver. It never touched the ceramic at all.

Killa-X

Wed Aug 15 2012, 02:13PM

Registered Member #1643 Joined: Mon Aug 18 2008, 06:10PM
Location:
Posts: 1039

Is there a point of limit where you can have too much cap? I mean, you can take a 1" x 4" coil for example, something that maybe up in the 4000mhz range, and make that 200khz with enough cap (given it fits inside). But, im sure theres a point where it doesnt work like that?

But 30awg for that is stupid as you would need 430pF of cap if im not mistaken, to get down to 250khz .. horrible xD about 100pF if you went with 36awg, etc.

But if you have some skill, and get 42awg wire, you can use your 100KV 22pF MMC and get about 270/250khz. But by then id be afraid of melting the wire...

plazmatron

Wed Aug 15 2012, 06:19PM

Registered Member #1134 Joined: Tue Nov 20 2007, 04:39PM
Location: Bonnie Scotland
Posts: 351

Killa-X wrote ...

Is there a point of limit where you can have too much cap? I mean, you can take a 1" x 4" coil for example, something that maybe up in the 4000mhz range, and make that 200khz with enough cap (given it fits inside). But, im sure theres a point where it doesnt work like that?

But 30awg for that is stupid as you would need 430pF of cap if im not mistaken, to get down to 250khz .. horrible xD about 100pF if you went with 36awg, etc.

But if you have some skill, and get 42awg wire, you can use your 100KV 22pF MMC and get about 270/250khz. But by then id be afraid of melting the wire...

I'm sure there is a point where you will have just too much cap, but for small coils, even pulling it down a couple of hundred kHz, will relieve some stress off the IGBT's if nothing else.

I was plugging in numbers on a tesla coil calculator, to see what results I would get with 0.125mm wire on a 1.5" by 3" form. This gives a fres of around 1.8 MHz. It would take 100pF of Csec to drop it to 765kHz, and 240pF to drop it to under 500kHz.

Obviously we need a wee sanity check on the design here, and personally I would be inclined to try 100pF, and be happy to get the thing under 1MHz. However this is a new thing in the Tesla coil world, and experimenting with these things is most of the fun!

Once I get some more spare cash, I intend to try and pull a 1.8MHz coil to under 500kHz, and it will either work, or it wont, but nobody will know for sure unless it is tried out. You really have to gamble with these things to see progress.

Les

Steve Conner

Wed Aug 15 2012, 08:37PM

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

If you use too much capacitance, the loaded Q will be excessive. From Terry Fritz's streamer models, I estimated that a Tesla resonator should have a characteristic impedance (sqrt(L/C)) of about 20 to 50k ohms, which gives a loaded Q of about 10 with ordinary SGTC or DRSSTC streamers. (The VTTC and QCW are somewhat different.)

If you try to make a tiny coil with low Fo for easy drive by IGBTs, you end up with too much inductance and not enough capacitance. The characteristic impedance ends up too high. The capacitor bank helps to remedy this, but you don't want to use so much that Zo drops below 20k.

plazmatron

Wed Aug 15 2012, 09:01PM

Registered Member #1134 Joined: Tue Nov 20 2007, 04:39PM
Location: Bonnie Scotland
Posts: 351

Thanks Steve, that gives me something to go on!

In that case, I can't use 200 odd pF with the 1.5/3" resonator described, since my Zo would drop to around 4.7k however, it would suggest that such a form wound with finer wire, and a more conservative Csec would be perfectly do-able.

It sure would be cool to have a 3 inch high coil producing 6"+ arcs!

brandon3055

Thu Aug 16 2012, 02:45AM

Registered Member #4548 Joined: Mon Apr 23 2012, 03:52AM
Location: tasmania
Posts: 271

Wow that is so cool I love it
BTW with a Dr that small (or smaller) would it be possible to use mosfets instead of igbt's?

Ben Solon

Thu Aug 16 2012, 03:44AM

Registered Member #3900 Joined: Thu May 19 2011, 08:28PM
Location:
Posts: 600

Destroyer of mosfets wrote ...

Wow that is so cool I love it
BTW with a Dr that small (or smaller) would it be possible to use mosfets instead of igbt's?

It all depends on the pulse current. Personany I think that the 3" secondary with a ~fres of 750khz could be driven with mosfets to avoid the igbt speed down. And for a decent spam length to secondary length ratio could be easily archived with minimal pulse current for a coil of this size.

plazmatron

Thu Aug 16 2012, 03:45AM

Registered Member #1134 Joined: Tue Nov 20 2007, 04:39PM
Location: Bonnie Scotland
Posts: 351

Destroyer of mosfets wrote ...

Wow that is so cool I love it
BTW with a Dr that small (or smaller) would it be possible to use mosfets instead of igbt's?

It has been done before, but why?
TO-247 IGBT's are cheap enough, and since it is possible to drop the fres with a secondary MMC to something IGBTs will happily handle, you might as well stick with IGBT's

Even in this small coil, I expect the primary current would be just too high for MOSFET's to handle well anyway, but I lack the equipment to measure it properly (must get a new scope!)

Uspring

Thu Aug 16 2012, 12:39PM

Registered Member #3988 Joined: Thu Jul 07 2011, 03:25PM
Location:
Posts: 711

plazmatron wrote:

In that case, I can't use 200 odd pF with the 1.5/3" resonator described, since my Zo would drop to around 4.7k however, it would suggest that such a form wound with finer wire, and a more conservative Csec would be perfectly do-able.

With a spark load resistance of a ballpark 500k that would amount to a secondary Q of about 100. That would be very difficult to get into tune.

I estimate your secondary Z to be about 30k now, which results in a Q of 17. That is ok but still on the high side. From my experience there is no problem of having a lot lower Q down to maybe 3. That makes tuning much easier.

Essentially that means increasing your secondary inductance. This will have an effect on your primary currents, though. A low Q coil behaves much like a standard transformer. A streamer resistance on the secondary will show up as a resistance in the primary given by:

Rprimary=Rstreamer * k^2 * Lprim/Lsec

From the data you gave I estimate Lsec to be about 18mH, Lprim about 1.5uH and I guess k to be about 0.25. For a streamer R of say 500k, Rprimary would then be about 2.5 ohms. This value should match the bus voltage and primary current you want to work with.

If you now, e.g. double secondary inductance, Rprimary will halve. That essentially means doubling primary current if the bus voltage is kept constant. Whether this is good or bad depends on your bridge and power supply.

previous 1 2 3 4 next

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