CWDRSSTC / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

CWDRSSTC

previous 1 2 3 4 next last

Move Thread

LAN_403

Goodchild

Sat Nov 12 2011, 06:16PM

Registered Member #2292 Joined: Fri Aug 14 2009, 05:33PM
Location: The Wild West AKA Arizona
Posts: 795

You know one of the things that makes the QCW so wonderful is the fact that it's tank voltage is rather low. A result of the lower current.

The tank Z is not that high, usually it resides in the 25 to 30 ohm range although this seems high, when you are only running 150A to 200A this still only equates to 6Kv peak at the worst case! The tank in my QCW usually doesn't see more than 4.5Kv under normal operation.

I have also run my QCW in CW mode and the results are less that exciting. Other than a big bush of plasma and a lot of power draw and melted tungsten breakout point (yes tungsten

) I have conclude that CW is not all that great. I think the highish duty cycles say in the 30% to 60% range are the butter zone.

Running a coil like the one you are talking about building using half rec mains would be a very similar operation and run at about 50% duty. This should make for very nice operation. The current should stay under 100A in most cases keeping your tank cap happy.

The QCW is a great type coil because it's mode of operation is very controlled. lots of control means longer sparks and less power draw.

Hope this helps,
Eric

Marko

Sat Nov 12 2011, 06:44PM

Registered Member #89 Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145

Hey Goodchild

Well, as a result of accumulated brain damage I have a special fetish for true CW sparks... being so hot and powerful, as well as the sheer excitement of 10's of kW going in with nothing blowing up makes all the pulsed sparks look 'faux'... but yeah I'd definitely love to see what would QCW sparks look with 10kW input as well!

I don't think using a buck converter is necessary at all, just having a comparator fed with a ramp and it's output acting to control the current should do well enough in my opinion.

Marko

Dr. Dark Current

Sat Nov 12 2011, 06:45PM

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

Nice work, Marko.
Over the years, I came up with various start-up oscillators for direct amplification feedback. I'll attach a circuit I'm going to use for the coil. Without feedback, it works as a normal hysteresis oscillator, but after the tank current rings up a bit, the oscillation is blocked and the inverter starts amplifying the feedback signal. The diode thyngus is just a bipolar voltage clamp. Note that this circuit will need a real-world testing, but I'm convinced it will work.

Goodchild, did you observe what effect the tank capacitance had on the behavior of the coil?

Marko

Sat Nov 12 2011, 07:46PM

Registered Member #89 Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145

Hm, as far as I remember this is similar to what kim ladha used for his "OMG induction heater" from a while ago, and as far as I know it tended to blow up too.
I'm not sure what kind of instabilities might be happening there (perhaps someone with knowledge on nonlinear systems could help) but from the noises I could hear from the inverter itself I think they were pretty bad. I once had a fullbridge of 47N50 mosfets blow instantly upon switching the power on with no explanation at all.

From what I know, your oscillator has to have some sort of hysteresis regarding the signal that disables it too, otherwise it will fight it. So once the feedback signal appears it won't start back up until it has well fallen to zero. I've actually made a driver that does all this, but it turned overly complex and for some reason (probably poor thermal design) that coil was a mosfet eater so I abandoned the approach.

Finally, one simple advice that can save you countless IGBT lives - use overcurrent protection! No matter if you even use DR or go hard switched, a simple CT on the bridge output, a LM311 and a NE555 will prevent virtually any damaging overcurrent fault (apart from shoot through, but this will not be a problem if you use GDT's).

Steve Ward's original SSTC's seemed to work excellently with FWR and HWR inputs and were quite simple designs, yet hardly anyone makes reference to them today.

Marko

Dr. Dark Current

Sat Nov 12 2011, 07:55PM

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

Marko, I'm going to use this circuit for primary current feedback (with two cascaded CTs). Once the primary current rings up, the signal should be strong and nothing (eg. ground arcs) can disturb it. (note a too short arc can cause the primary current to ring up to a dangerous value)
I would never use a direct amplification feedback for secondary base or antenna input, as the signal is lost during ground arcs and the feedback loop can start to act erratic.

Marko

Sat Nov 12 2011, 08:17PM

Registered Member #89 Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145

Dr. Kilovolt wrote ...

Marko, I'm going to use this circuit for primary current feedback (with two cascaded CTs). Once the primary current rings up, the signal should be strong and nothing (eg. ground arcs) can disturb it. (note a too short arc can cause the primary current to ring up to a dangerous value)
I would never use a direct amplification feedback for secondary base or antenna input, as the signal is lost during ground arcs and the feedback loop can start to act erratic.

Well, what I meant was more like a comparator with hysteresis (or at least another HC14 gate used as one) to independently sense the amplitude of primary current and turn off the oscillator, until the current drops too low again. If you let the primary current "fight" the oscillator for gain in linear region the current may never ring up to fully override it and cause assorted problems (IIRC, this circuit was particularly fiddly in induction heaters ad and had to be tuned finely to resonant frequency of the tank). It's not that I want to pester you, but I just had a lot of bad experience with these circuits.

BTW, I can upload my version of this thing if you're interested (and If I can find it!).

I still think that using a sensitive fast comparator (for 100kHz range, LM119 and like will be more than good enough) instead of going through all this trouble.

Or heck, why not just use PLL (Copy Steve Ward's circuit, add overcurrent protection and better heatsinking to the devices!).

Marko

Dr. Dark Current

Sat Nov 12 2011, 08:29PM

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

Marko wrote ...

Well, what I meant was more like a comparator with hysteresis (or at least another HC14 gate used as one) to independently sense the amplitude of primary current and turn off the oscillator, until the current drops too low again.

Of course the circuit I posted is for oscillator/resonant frequency feedback, not overcurrent protection. If I was going to use an overcurrent protection, I would put a shunt resistor between the two CTs (to eliminate the need for another CT) and use this to control a small SCR which would shut down the gate drivers through their enable pins.

Sure you can upload the circuit if you want, I'll take a look.

Marko

Sat Nov 12 2011, 09:32PM

Registered Member #89 Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145

Dr. Kilovolt wrote ...

Nah, this would be a separate comparator, nothing to do with overcurrent protection, just turning off the oscillator properly.

Here is the pic of what I used, it's a bit weirder than I expected but has a point... the 74HC14 gates provide enough hysteresis to decouple the oscillator from the feedback signal well. Though I'm not sure if all this is necessary when a comparator could simply be used that might not even require any oscillators to start.

Dr. Dark Current

Sat Nov 12 2011, 09:52PM

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

Yes now I see what you're up to with the circuit, sorry I thought you were referring to the overcurrent comparator.
The circuit I posted should lock at very low bus voltage, and it should be stable once it locks. There *might* be a problem with turning the TC on at the positive peak of mains voltage, but I should get around this problem somehow if it happens (like a zero cross detecting SSR or something like that). But I believe this will not be necessary.
I have used similar design of start up oscillator for an electronic feedback for "zvs" driver and "zvs" Tesla coil, and there were no problems with it, though the feedback was voltage not current.

Marko

Sun Nov 13 2011, 01:26AM

Registered Member #89 Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145

"zvs" Tesla coil

Hm, even though it's a bit offtopic (or not, as I guess it categorizes as CWDRSSTC too) - but, OMG pix/schemz/vidz?

Marko

previous 1 2 3 4 next last

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