Asynchronous Rotary Spark Gap Tesla Coil

[profdc9]

Hello.  I made a video of my dual microwave oven transformer and put it on youtube of course.  You can see it here:

/>
The specs are:

Two microwave oven transformers with their grounds connected together to put them in series.  There is a voltage doubler on the transformers as well.  Each secondary is in series with a 0.95 uF microwave oven capacitor, and the diode is four microwave oven diodes in series in a Villard multiplier. 

An asynchronous rotary spark gap consisting of an angle grinder.  The electrode wheel is made of G10 phenolic, and there are 8 electrodes that are stainless steel bolt heads and nuts held in with lock washers.  The two points next to the electrode are 1/4"-20 stainless screws.  A lot of sparks fly off the ends of the screw heads when its running as the end of the screw heads get very hot and glow red.

The primary capacitor bank are capacitors from www.hvstuff.com.  I took a chance on these generic capacitors, which I think are Chinese made, and they seem to work well.   I ran the Tesla coil for 10 minutes at full power without problems.  I was careful to stay under the peak current limit of the capacitors, however.  A picture of the bank is attached to this message.  The capacitor bank is 25 nF in parallel with 13 nF.  The 25 nF is four 10 kVDC 100 nF capacitors in series, and the 13 nF is three 10 kVDC 40 nF capacitors in series.  The series combination was used to make sure the voltage rating was well above spec, and to keep the peak current within limits.

The primary coil is conical tapped at 5 turns made of 0.25" copper tubing.



The secondary is about 1000 turns of 26 awg wire on a 4" PVC pipe.  The topload is two large stainless salad bowls.

The power supply has a Terry filter on it.  The Terry filter has 10 1.8 kV MOVs in series, 2200 pF @ 18 kVDC filter ceramic filter capacitors, two 400 ohm resistors, and a dual safety spark gap.

I have been able to get about 1 meter sparks so far.    If anyone wants more details I'd be happy to share.

Enjoy the show!

Dan

[Hydron]

Nice work!

I have always been tempted to build a ASRSG spark-gap coil, mostly for that asynchronous rotary gap sound and the way the sparks move (especially without breakout point). I've still got a couple of matched MOTs in storage in case I find time one day...

I'm a little confused as to how you're doing the voltage doubling - do you have a DC output? (MSPAINT diagram could help!)

[profdc9]

I would be glad to oblige.  See the attached PDF for a full schematic.

It is a pulsed DC output at 60 Hz.  For 1/2 the cycle, the capacitors are charged.  For the other half, the power supply is in series with the capacitors, and it delivers double the voltage.  This probably contributes to the sparks being a little more wild.

Also, I have included a template for making the wheel for the angle grinder.  One of the hardest parts is balancing the wheel because it is rotating so fast.  This will get you close, then you put it on the angle grinder and grind off the edge until it is round using a file.

Watch the video at 0.25 X speed.  It looks cool that way.  Next time I'm going to take a video using a high speed camera.

Dan

[AndreiRS]

Nice. How many bps is that? The small grinders spin at 11000rpm, 8 electrodes... That is 1466 "hz" or bps. Usually people go for 120, 200, 400, if you could explain you choice on electrodes quantity. Look very good.

[Hydron]

Yeah I was wondering if it was pulsed DC or not - the pulsing action probably helps a lot with stopping the gap from just pulling a big arc directly from the power supply. The other way to do it is with DC resonant charging (see http://www.richieburnett.co.uk/dcreschg.html#resonant ), but that requires the extra choke and diodes.

[profdc9]

I actually put a triac-based speed controller on the angle grinder, so I can vary it between probably about 3000 RPM and 11000 RPM.  However, it seems to work best at the highest pulse rate.  The number of pulses is not really 1500 pulses per second as the voltage is only applied for half the cycle, and even during this half cycle the voltage can only bridge the spark gap for part of the cycle.

The number of electrodes is limited by the size of the angle grinder.  Originally I only used 4 electrodes but that was not enough, as I thought as you did that the pulse rate wouldn't be this high.   The electrodes can not be too small or they won't conduct the current and burn up.  But for lots of electrodes, you need a big wheel which needs to turn fast.  The angle grinder discs are 4.5" in diameter.  I used 5.4", but the phenolic disc is less dense than the typical corundum grit angle grinder disc.  So maybe I could have put 12 on there, but not many more.  The disc gets heavy too, and it is hard to balance, and things will start to vibrate strongly which damages the arbor.

I didn't want to have to build a full multiplier using two diodes and two capacitors, as both sets of capacitors and diodes would have to take the full voltage.  The Villard configuration I used is what's already used in a microwave oven, which is why it makes a magnetoconstrictive buzzing noise, and it seems to work ok.  I'm sure the power factor is horrible however.  Loneoceans tried this configuration in his dual MOT tesla  coil (http://www.loneoceans.com/labs/teslacoil2/) but he had problems with it because he had problems with the diodes getting blown by the high voltage RF spikes.  I didn't want to do what he had to do, which is use four transformers submerged in oil, because I have tried oil and it is very messy.  That's why I used four microwave oven diodes and highly overengineered the Terry filter and there doesn't seem to be a problem, and why I included those details it in the schematic.

Dan

[Mads Barnkob]

That is a good result for only using 2 MOTs, how much power are you putting into it? I was thinking about how well you fulfil the maximum spark length formula? http://deepfriedneon.com/tesla_f_calcspark.html

Your MMC is a little problematic, you have 25 nF in one string and 13 nF in the other, current sharing between the strings is determined by the difference in capacitance ratio, so the string with the highest capacitance will see the most current, this is a path for capacitor failure. Also the uneven wire length, different capacitors etc. is not optimal.

Huge thumbs up for always supplying great documentation to your posts! 

[profdc9]

I am aware that the current is shared unequally between the two chains of capacitors, but I have checked that the current for the larger chain is within limits.  I bought those capacitors when I didn't really know what I was doing, and put them in an arrangement that would be conservative with the voltage and current ratings.  Now that the website is back up, here is a link to the capacitors I bought.  They're very cheap!  I wasn't sure they would work at this price, but so far they have held up pretty well:

http://hvstuff.com/0-04uf-10kv-high-voltage-capacitor-hv-tesla-coil-ham
http://hvstuff.com/0-1uf-10kv-high-voltage-capacitor-hv-tesla-coil-ham

The problem with these is that the current rating is too low for the capacitance, especially the 0.1 uF capacitor.  If I was going to buy again, I would buy these perhaps which seem to a cheaper version of the 0.15 uF 2 kVDC size capacitors commonly used:

http://hvstuff.com/2kv-0-15uf-high-voltage-capacitor-hv-tesla-coil-ham

or perhaps just buy only the 40 nF capacitors alone.

I think I need more capacitance to increase the spark length, because 38 nF is too low, probably should be closer to 60 nF or even higher.  I have read about the formula you link to, but I think the pulse energy must be a factor as well as the total power, and so more capacitance is needed to realize the limit specified in the formula.  However, I would rather start work on a DRSSTC than spend more on capacitors for the MOT coil for now.

Dan