Micro SSTC Tuning / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

Micro SSTC Tuning

Move Thread

LAN_403

alan sailer

Fri Nov 25 2016, 10:59PM

Registered Member #59110 Joined: Mon Apr 11 2016, 04:35PM
Location: Camarillo, California
Posts: 74

Hello there.

I spent the last few days working up a small solid state coil after Steve Ward's single MOSFET coil> The coill is together and working as well as I could expect ( ~1 inch arcs at 19.5volts) but I'm curious about several aspects of the unit.

First I'm wondering about the role of the small 0.001 uF capacitor across the MOSFET. I'm guessing it is part of the tuning for the unit. If so is it a worthwhile effort to try various values both higher and lower?

Second, I tried messing with the coupling of the primary to the secondary. Raising/lowering the primary and ended up setting it at a point that gave highest voltage. This also seemed to be an area of lower current draw from the primary power supply. The current waveform also changes drastically during this procedure. The current waveform I will admit, makes no sense to me whatsoever. Is there an optimal current/voltage waveform for this style of coil?

Third, I started out withe the called for MOSFET IFRP460 but after blowing up two I ran out of them and changed to an IRFP250. The coil did no seem to be radically different. I also added a varistor for over-voltage protection. What are some of the characteristics to look for in a MOSFET in this circuit?

Finally I am curious about the grounding of the unit. Right now it has no dedicated ground and seems to work fine. My first coil, a SGTC had a real by-god ground rod since many of the sources I read were adamant that a real ground was essential. My second coil, made from LoneOceans board is a medium sized SSTC running at 120AC and has no ground other than the mains.

As a side note, the character of the coil changed a lot as I went from my first light to final set-up. At some points the thing was quite angry, with a larger corona brush and drawing over four amps at 20 volts. As I neatened up the layout and went from two supplies for the unit (12 volts for the driver and 20 volts for the primary) to a circuit nearly identical to Steve Ward's the corona settle down to a quiet discharge drawing about two amps.

Cheers.

Sulaiman

Sat Nov 26 2016, 04:06PM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

few of us keep entire circuit diagrams in our heads,
of those that do not,
most will not take the effort to look for the diagram and post a link if you yourself do not.

consider yourself chastised :)

E.TexasTesla

Wed Dec 07 2016, 04:41AM

Registered Member #4362 Joined: Sat Jan 21 2012, 03:44AM
Location: Texas
Posts: 98

Hello Alan, welcome to the forum.

To answer you first question you might want to visit this link

It has a good description of why the tank cap is needed.

Second Question is also explained in the link but basically every coil has a sweet spot when it comes to primary coupling.
I have some coils that work best with the primary one half inch above the base of the secondary.

Third question you should look at the datasheets for both fets. Compare switching times and voltage ratings.
Current ratings are not critical in a tesla circuit due to the short on times and most tesla circuits are pushing them way over max specs. Over voltage = sudden death. ;(

Fourth question is somewhat complex. Without some form of ground plane the coil will not work.

alan sailer

Thu Dec 08 2016, 12:52AM

Registered Member #59110 Joined: Mon Apr 11 2016, 04:35PM
Location: Camarillo, California
Posts: 74

E.TexasTesla,

Thanks for taking the time to reply.

I will check out the link and see if it helps me to better understand.

As far as device rating it is good to know overvoltage is most common for FET death in tesla coils. I notice that people use different methods of protecting the FET. Back to back zeners on the gate seems pretty common. But methods seem to differ on the output side. Some use snubber caps, some use resistor/capacitor snubbers and some use varistors (MOV).

Is there any reason for selecting one of these over the other?

I guess I'll have to keep wondering about grounding. I have a background in microwave circuit and ground is always very important and very explicit ie ground planes. This tesla stuff seems like witchcraft by comparison.

Incidentally my interest in tesla coils had one purpose; I want to get a coil powerful enough to light up some home brew plasma tubes/globes. I'd like a coil that does a good job of doing this but still be relatively safe.

Cheers.

Dr. Slack

Fri Dec 09 2016, 03:12PM

Registered Member #72 Joined: Thu Feb 09 2006, 08:29AM
Location: UK St. Albans
Posts: 1659

Hi Alan,

If you have a background in microwave, then you already know all you need to know about grounding. The thing you have to bear in mind to understand what you read about it in Tesla forums is that most coilers do not understand it, and treat it as witchcraft. Remember that at common sizes and frequencies, a room width is << lambda/10, so you can to first order ignore the inductance of a connecting wire.

Take two ideal situations, one operating a TC in a Faraday cage, the other in the middle of a large flat field, in each case with the bottom of the secondary wired straight down to the cage or ground. The capacitance of the top load in each case will be the same order of magnitude, as it goes as log(dimension ratios). The electric field leaves the topload, and terminates on ground everywhere. That capacitance will be in the order of 10pF.

Now if we take a less ideal case, operating in a house, or the 10th floor of a high-rise, where do the electric field lines terminate? The answer is the conducting structure, the plumbing, and the mains wiring. If the coiler has run a wire from his secondary 0v to a remote ground, then the TC will be exciting the circuit between the house wiring and that remote ground. However it will still work quite happily, though possibly trashing RF reception anywhere that shares the imediate wiring.

What he should do of course is to connect the secondary 0v to the local mains ground, and use high voltage low value caps to short the mains supply together at RF to avoid exciting some mains conductors with respect to the others, and forget about the remote ground. One is it doesn't really matter too much, and the other is that the mains supply goes down to the ground anyway.

Many coilers will use a counterpoise, which is a sheet of alli foil, piece of chicken wire, or a star of conductors just underneath the coil, to make a local ground plane. Ideally that's also connected to wiring ground locally.

alan sailer

Sat Dec 10 2016, 01:18AM

Registered Member #59110 Joined: Mon Apr 11 2016, 04:35PM
Location: Camarillo, California
Posts: 74

Dr, Slack,

That is a very interesting way of putting things and will give me some ideas to chew over.

I've built four coils at this point and the secondary frequencies are all over the map (from 200kHz to 5MHz). Even at 5MHz the wavelengths (as you said) are still pretty huge. So ground should be a much more distributed concept (to state it badly).

The first one is the only coil (SGTC) that I paid attention to the ground explicitly. I was worried about high frequency getting onto the line and destroying household electronics. The main ground was an earth rod, way overdone and in addition I had a lot of mains connected filtering, the very same HV capacitors hooked from one line to the next.

Your topload capacitance description also makes lots of sense. It's completely the same situation of microstrip vs stripline characteristics. the stripline is completely shielded and the microstrip only half, but to first order both have very similar fringing capacitance.

Thanks for taking the time to explain in more familiar terms. One of these days I will just have to take an afternoon and play with various grounding schemes and see what I see.

Cheers.

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