Building of my first tesla coil

[Skylex86]

Hello everyone

I'm back after a few months of absence
I have made some progress on my DRSSTC project, I received all the components for the management cards.
I finished putting them together this weekend.



I also found a large capacitor



the support of the primary coil has been painted
I got a metal box to store the control cards, in addition to the fence that will be put all around it 

I still have some measurements to make, in particular the inductance of the primary coil to be able to calculate the value of the resonant tank capacitor and also soldered the interrupter (It's the same as the one onetesla TS).

a quick question:
can we test the control board without the high power part ?

Alexis

[Mads Barnkob]

Good to see your progressing well with your coil and I love your home-etched boards, looks just like mine

A word of advise on the zipties on the primary, if your primary coil heats up zipties are very easily loosing up and the little tap will let go, perhaps shorting out a few turns of your primary coil. So be sure to watch the temperature of it or you could fasten the turns with stiff copper wire, you have enough space to avoid short circuits from that.

You do not really need to do measurements of your primary inductance, JavaTC is so close to reality that you can just plot your dimensions into that: http://www.classictesla.com/java/javatc/javatc.html

You can test the UD without power on the bridge, but you need to "fake" the feedback. Use a signal generator to feed a sine wave into the rectifier diodes of the feedback input. You can use a power supply for the OCD voltage level, just inject it after the diodes, or use a 2nd signal generator if you have one.

[Skylex86]

Hello

Well, it's not moving very fast for the moment but it's going to change !
I'm finally on holidays !
I receive my function generator next Monday or Tuesday.
today I made the rolling base of the tesla :



tomorrow she's going to paint.

I have one more question about IGBTs
I found a 5SNS 0300U120100, not very expensive and not far from home
what do you think of this IGBT module ?

Here are the Datasheets :
https://datasheetspdf.com//pdf-down/5/S/N/5SNS0300U120100-ABB.pdf


Thank you in advance for your answers.
Alexis

[Mads Barnkob]

Do not worry about time flying, some of my coils took years to finish

Those ABB 3-phase IGBT bricks would have to be used as half-bridge bricks where you drive en 3 half-bridges in parallel and combine the output terminals, both for making a half-brigde and two of them for a full-bridge. Their switching speed is fair for maximum 150kHz (I did not check heat dissipation! Just looked at the times), it has a good high pulse current rating and short-circuit rating, but BUT, there is always a but. The gate charge is really large and it will require much more from your driver, so you have to check your 24VDC rail on the driver and see if its sagging, you might have to add a more powerful external gate drive voltage power supply instead of a 24V 1A regulator. http://kaizerpowerelectronics.dk/calculators/igbt-gate-drive-calculator/

Your control box is looking good, what kind of safety features did you put in it? I see emergency stop and what does the switches and lamps do?

[Skylex86]

Hello Mads

If the only problem is that they consume a little more than the others, I make them eat differently
unless the rev 1.3 card is not able to support the required intensity

What do you mean by:I did not check heat dissipation ?

according to the datasheets I would have 960w to dissipate 
Total power dissipation Ptot Th = 25 °C, per switch (IGBT) 960 W

[Mads Barnkob]

With 6 gates in parallel, for a full-bridge, you would need atleast 10Apeak driving power and 0,144A average driving power, you risk having sagging 24V from that heavy load, so beef the 24V rail up with larger lytics, but I also think you need something larger than a 24V 1A linear regulator.

All datasheet parameters are given for hardswitching in a regular inverter type of application, like a motor drive.

A DRSSTC utilizes softswitching/resonant switching to handle the much higher peak currents we need for spark generation in a very short time.

You need to calculate the Fmax2 value and you can follow my guide here on how to do it: http://kaizerpowerelectronics.dk/tesla-coils/drsstc-design-guide/igbts/

Just remember there is a lot of assumptions and it is worst case scenarios, so you can tweak some of the values of derate them even  further, it all just comes at the cost of lessened life time of your components

[Skylex86]

Your control box is looking good, what kind of safety features did you put in it? I see emergency stop and what does the switches and lamps do?

The lamps are used to know which safety is activated (When there is just the power supply connected the green light is on, if the circuit-breaker is passing through then the orange light comes on and if the electromechanical contactor is also passing through the red light comes on).
Moreover it is impossible to turn it on without a key

The rest is used by interrupter (the screen and the red buttons).
I'm programming the atmega of the interrupter


I calculated gate charge :

but I couldn't find gate resistor int

I also tried to calculate Fmax2,but I couldn't find RθJC, Pcond

But i compared the CM600DU-24NF with the 5SNS
I compared the CM600DU-24NF with the 5SNS
The 5SNS takes less time to activate and deactivate, Total power dissipation is much higher for the CM600

I also measured the inductance of my primary coil, I am at 7µH. Then I calculated it with JAVATC and it shows me 7.28µH
So I will start on 7µH and take access to the capacitor to be able to tune it between 6.5 and 7.5µ


otherwise the trolley has been painted, it already looks better 

[Skylex86]

Hello everyone
It's been a while since I've been here. The tesla coil hasn't progressed much since the last time.
I had other important projects to take care of and with the studies, I had even less time to devote to them. 
But having a little more time at the moment, I decide to get back to it.

I recently received 5 igbt SKM 400GB 17E4.
I think they will be able to go well beyond the 500A spike I was considering.

I'm going to resume the calculations to finish the power part of the tesla.
I will keep you informed of my progress.

Have a nice day
Alexis