small DC solid state Tesla coil - help needed / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

small DC solid state Tesla coil - help needed

Move Thread

LAN_403

aquilarubra

Sat May 12 2012, 02:57PM

Registered Member #2471 Joined: Wed Nov 18 2009, 07:23AM
Location:
Posts: 17

Hello, I was trying to build a 100W DC resistive solid state Tesla coil - a frequency controlled Violet Ray, to be precise (we could say a variant of OLTC). I tried to follow the explanations at Rickie Burnett site, keeping components at a minimum. However, it just does not work. The transformers, resistor and Mosfet get very hot, and that's it. I have a limited knowledge of electronics. So... could somebody guide me to troubleshoot it, or could somebody spot out if I did some mistakes?

Thank you!

Dr. Dark Current

Sat May 12 2012, 03:05PM

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

The impedance of the circuit is way too low for the mosfet, all that happens on turn-on is that it dissipates the energy in the tank cap, the Q of the circuit is close to zero. Look up on OLTC designs, they require completely different component specs.

See the Mini OLTC

aquilarubra

Sat May 12 2012, 03:21PM

Registered Member #2471 Joined: Wed Nov 18 2009, 07:23AM
Location:
Posts: 17

Thanks, but I'm not so expert to understand the meaning of impedance and Q. Does it mean I should put a resistor between the mosfet and the + pole?

The mini OLTC seems very similar. It only uses inductive charging, which is what I want to avoid. C1 is bigger, but I want to work with smaller energies. VRs usually use 0.1uF.

ScotchTapeLord

Sat May 12 2012, 03:44PM

Registered Member #1875 Joined: Sun Dec 21 2008, 06:36PM
Location:
Posts: 635

aquilarubra wrote ...

Thanks, but I'm not so expert to understand the meaning of impedance and Q. Does it mean I should put a resistor between the mosfet and the + pole?

The mini OLTC seems very similar. It only uses inductive charging, which is what I want to avoid. C1 is bigger, but I want to work with smaller energies. VRs usually use 0.1uF.

The only real difference here is that he's replaced your R1 with an inductor and uses two parallel switches, balanced by a split primary.

I think what you have to focus on are the values of your tank circuit. You need more capacitance. You also don't need such a high voltage capacitor in your tank circuit. Make sure your switch is capable of handling the peak current, which can be calculated using the surge impedance of whatever tank circuit you choose. (Hint: The surge impedance should be very low! Hence the parallel devices in the referenced mini-OLTC design)

Steve Conner

Sat May 12 2012, 03:53PM

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

My mini OLTC has a half-power switch that changes the charging from inductive to resistive. So, I can confirm it works with resistive charging too.

For good results, the surge impedance of your primary tank circuit (sqrt(L/C)) should be large compared to Rds(on) of the MOSFET and the DC resistance of the coil.

However, the surge impedance also needs to be low so that C can be big enough to store a decent amount of energy, without lowering the resonant frequency so much that an air-cored resonator is impractical. This is the central trade-off of OLTC design: the surge impedance has to be high but also low.

Increasing the charging voltage breaks the tradeoff, so I always used the highest charging voltage I thought I could get away with. Inductive charging helps by giving nearly twice the voltage of resistive charging. IGBTs also help because they have lower losses than high voltage MOSFETs. In my last OLTC I ended up running 1000V with 1200V IGBTs.

aquilarubra

Sat May 12 2012, 03:57PM

Registered Member #2471 Joined: Wed Nov 18 2009, 07:23AM
Location:
Posts: 17

Two parallel switches? I saw this schematics:

The second switch is only used to drive the main mosfet. I use a mosfet drive, instead. I know 3000V for C2 are excessive. I just used what I had on hand. Anyways, VRs all use .1uF, so I would not change that. VRs work at a resonance of 500kHz and diathermy machine around 1MHz. I would like to have around 1MHz. So I couldn't use a bigger tank cap.

To give an idea of the project:

Goal: make a trigger for a xenon lamp, which is frequency controlled.

The mosfet should be driven from 0 to 2 MHz and work at a resonant frequency of about 1 MHz.

aquilarubra

Sat May 12 2012, 04:00PM

Registered Member #2471 Joined: Wed Nov 18 2009, 07:23AM
Location:
Posts: 17

Note: I cannot use IGBTs, because it would not switch at the high frequencies I need. I should use a mosfet.

Steve Conner

Sat May 12 2012, 08:59PM

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

This is the schematic of my Mini OLTC, more or less as built.

It lacks the "Quench-o-meter" but has a switch to select resistive or inductive charging.

I've never tried building a MOSFET OLTC, but I wouldn't be too hopeful. Bear in mind though that the device doesn't actually switch at 1MHz. It switches at 100Hz or whatever the pulse repetition rate is, so it doesn't matter if the switching losses are high, because the number of switching events per second is low.

The device just needs to turn on fast enough that the current rise time is shorter than one half-cycle at 1MHz, and it doesn't need to turn off fast at all.

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