CCPS (Capacitor Charging Power Supply) / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

CCPS (Capacitor Charging Power Supply)

first previous 9 10 11 12 next last

Move Thread

LAN_403

Marko

Sun Sept 23 2007, 05:03PM

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

I don't see any point (unless your point was to make others not to see the point). You need like a watt or two in overall at 15V for your gate supply, you could do it with an iron cored transformer smaller and lighter than that thing, or a royer oscillator driving a tiny ferrite core.

It's just a heavy Rube Goldberg machine this way

, much more things to get broken and to debug. Seriously, I would focus on perfecting the big one first

Finn Hammer

Sun Sept 23 2007, 10:26PM

Registered Member #205 Joined: Sat Feb 18 2006, 11:59AM
Location: SkÃ¸rping, Denmark
Posts: 741

Marko wrote ...

I don't see any point (unless your point was to make others not to see the point).

Marco, all

Daniel and I have a facination for industrial art, we see it as much as sculpture as anything else. Efficiency is not our main design target, although your points on thermal efficiency in the gate drive are well taken.
And then our arsenal of electronic solutions are limited.
For example, a royer oscillator. I know there is one in my scope, driving the flouresent tube that lights the display. because the first time I successfully repaired an electronic device was finding the defective cap in said supply.
But design one, I can`t.

Cheers, Finn Hammer

Finn Hammer

Tue Sept 25 2007, 08:36PM

Registered Member #205 Joined: Sat Feb 18 2006, 11:59AM
Location: SkÃ¸rping, Denmark
Posts: 741

Finn Hammer wrote ...

And then our arsenal of electronic solutions are limited.
For example, a royer oscillator. I know there is one in my scope, driving the flouresent tube that lights the display. because the first time I successfully repaired an electronic device was finding the defective cap in said supply.
But design one, I can`t.

Cheers, Finn Hammer

I just whipped a Mazilli driver together, and I don`t regret it.

The resonant frequency is determined by the primary L and Parallel C.

I tried it with 3 values of paralell caps. 10nF, 100nF and 1uF.
Resonant frequencies, 140kHz, 47kHz and 14.6kHz.
I measured the primary L to 76uH, so the resonant frequencies don`t match the expected frequency completely, but clearly a connection.
It delivers plenty of power, is small and neat.

We will be using such one supply for the Gate driver. Since it is isolated from the line, which probably is dirty when the coil is running, I will probably also run the voltage probe with one.

It is very satisfying to see a Mazilli oscillator in action for the first time. I just got aquainted with the term "household supply" and this sure is a nice one.

Cheers, Finn hammer

Steve Conner

Wed Sept 26 2007, 10:10AM

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

Yay, glad the Mazzilli oscillator worked for you. I used one to power my IGBT gate drivers too. IIRC, it runs at 60kHz and has a 68nF resonant cap.

It was only designed for DRSSTC duty, but it turned out to be able to drive a 300A brick CW at over 30kHz.

BTW, nice current probe, where did you get that? :)

Finn Hammer

Wed Sept 26 2007, 05:18PM

Registered Member #205 Joined: Sat Feb 18 2006, 11:59AM
Location: SkÃ¸rping, Denmark
Posts: 741

Steve Conner wrote ...

BTW, nice current probe, where did you get that? :)

EBay.

They show up every once in a while. I`ve got the 134 amplifier that goes with it, so with the scope at 50mA/div., it can do from 1A/Div. to 1mA/div. "up to 50mHz.
All I really need is a Rogowski coil based system, for those hard to get at places.
(I guess "need " is a relative word. Being the tool fetichist I am, perhaps "Would like to have" is better.
They never show up on EBay, though, and cost 800UKP, so that`s an investment for the future.

Cheers, Finn Hammer

edit

Steve Conner wrote ...

Yay, glad the Mazzilli oscillator worked for you. I used one to power my IGBT gate drivers too. IIRC, it runs at 60kHz and has a 68nF resonant cap.

It was only designed for DRSSTC duty, but it turned out to be able to drive a 300A brick CW at over 30kHz.

With full wave rectification, 100nF cap, 75kHz, and a healthy airgap in the core

It is doing 1.3Arms into 30Volts here:

which is enough to drive both bricks with 10ohm gate resistors CW to 300kHz as calculated with SEMIKRON`s own driver selection tool:

Cheers, Finn Hammer

Finn Hammer

Tue Oct 02 2007, 07:00PM

Registered Member #205 Joined: Sat Feb 18 2006, 11:59AM
Location: SkÃ¸rping, Denmark
Posts: 741

This is not Show Business, although some of my posts may have reminded you of a soap opera.....
However, I keep strickt adherence to the main rule for amateur entertainers: "Do not quit your day job"
So progress has been slow while I toil to keep the bills payd.

During the building of the Mazilli Driver life was a gravy train.

But when we hooked it up to the loaded driver, it became all too clear that Murphy has a major part in this soap.
The power into the driver passes trough 2 pcs. 1n5819 diodes which ought to be more than enough. But not untill we had burned up another set, and finally thrown in uf5408 instead, did the circuit finally show some ruggedness.
Things like this takes a lot of time to debug and set right, and I feel so stupid when it happens. On a day like that, it seems as everything that can go wrong, will go wrong. Just like Murphy said.

Murphy may be an entertainer, but I don`t like his act.

Cheers, Finn Hammer

Sulaiman

Tue Oct 02 2007, 07:19PM

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

Finn, that is a horrible waveform you've got there, I usually get a nice smooth sinewave (less emi/rfi problems)

I'm a little dubious of the very small series/dc inductor you've got there
just as a test try a small reel of wire as an air-cored inductor
if the waveform improves then you know what to do.

You may also want to try try adding some more parallel/resonant capacitance which will lower the resonant frequency
and minimize switching delay effects.

Marko

Tue Oct 02 2007, 09:46PM

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

Thou shall not use ferrite toroids for series inductor.

You need powdered iron core or largely gapped (rod) ferrite.

Why did you use multifilar windings for your secondaries? I can hardly imagine why would you need more than single wire for that power.

Finn Hammer

Wed Oct 03 2007, 05:55AM

Registered Member #205 Joined: Sat Feb 18 2006, 11:59AM
Location: SkÃ¸rping, Denmark
Posts: 741

Sulaiman wrote ...

Finn, that is a horrible waveform you've got there, I usually get a nice smooth sinewave (less emi/rfi problems)

This waveform is with the driver into a smoothing cap and a 30ohm resistor, to test it`s ability to deliver power. Without the resistor I get nice sine wave.

Sulaiman wrote ...

I'm a little dubious of the very small series/dc inductor you've got there
just as a test try a small reel of wire as an air-cored inductor
if the waveform improves then you know what to do.

Thanks for looking so closely. The incuctor below the circuit board is actually the one in service. Although the one on the board has the same inductance, it probably saturates, or somethng, at higher power levels.

Marko wrote ...

Thou shall not use ferrite toroids for series inductor.

You need powdered iron core or largely gapped (rod) ferrite.

I never found out quite what`s the difference. How can I tell the difference?

Marko wrote ...

Why did you use multifilar windings for your secondaries? I can hardly imagine why would you need more than single wire for that power.

I didn`t. It is a centertap commimg out. There is a + 24volt, -05volt supply to cater for.

Cheers, Finn Hammer

Steve Conner

Wed Oct 03 2007, 09:57AM

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

Hey Finn, just to say that Richie thinks you should lower your CCPS switching frequency to 20kHz. (ie, tune the resonant network to 40kHz) There is no single reason, except that it just "makes everything easier"

I also heard that spikes of twice the DC bus voltage are par for the course with IGBT bricks, and the "cure" is to use bricks rated at more than twice the DC bus voltage.

I used a choke-input filter with my Mazzilli PS, and I still get a nice sine wave under load.

first previous 9 10 11 12 next last

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