Wireless powering SSTC / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

Wireless powering SSTC

1 2 3 4 next last

Move Thread

LAN_403

Marko

Fri Nov 16 2007, 11:59PM

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

Hi, all,

I've started this yet before summer, and actually done most of work over it, only to lose motivation to work on it after school started.

I'm going to post lots of pics, but I have too much questions I'd want to ask and so put this into TC board.

I would really want to make this thing work but I'm stuck... doing other projects just increased the already too big number of unfinished projects.

I had these larger 400kHz resonators sitting around unused. I thought it'd be cool if I could demonstratively power something wirelessly via capacitive coupling - and I decided to make that this coil's main use.

In order to pump larger amount of power, I needed beefier switches and heatsinking. I had no heatsinks even closely that big so I decided to construct one from aluminium sheet.
It was lot of work but looked good in the end...

I don't intend to filter or interrupt the coil yet. It'll simply run from mains, fullwave rectified.

I use 2 paralleled H bridges of IRFP450's. They were supposed to drive a bifilar primary.

I have really a ton of pics here and didn't know which to select - too many already.
They tell much more than words...

Aluminium I had cut up.

Resonators

Toroid. I add a breakout point to that screw by need, or supress it by a ball to run at low power without breakout.

Cutting the hole inside the Al sheet was problematic - I managed to do it by screwing all sheets together and using drill and electric saw. Sheet and it's sawn edges looked crappy and I decided to give it a full polish with steel wool - boring work but made it look really cool in the end.

1195257551 89 FT0 Sstc Construction 2 12

Bridge board layout. I stuck them deep inside but still didn't want to need to unsolder anything in orther to disassemble the thing. Probably owerwork though.

This is how I secure everything. Mosfets are oriented wierdly as you may notice, this is to give each mosfet his half of aluminium plane. THe fifth sheet has no mosfets on it, coold only the drivers and diode bridge.

This is putting everything together. I weakly glued the sil pads and put grease over them - overkill too, but meh. Two plexy boards up and down make the ''case'' - the primary and secondary are held by a single brass bolt also serving as a ground contact. Spacers are put between two middle planes instead of mosfets. Nuts at the edges are drilled through and act only as 5mm spacers too.

The primary form with the infamous speaker cable. :(

And secondary base holder with a brass nut soldered to a piece of PCB and secured. There goes secondary base.

Decoupling caps:

This is how the thing looks after assembly. I planned to build in a small SMPS for it, but I'll leave that for later. Connectors are to be built in.. power is separate for the bridge and controller section.

Now, questions:

- I intend to change the driver section by removing the 5V logic (Schmitt trigger) stage, use only 12V for drive. Drivers are UC 3710's. I wanted to use a center-tapped CT, clamped with zeners since that would give neat square wave alone; I have free inverted signal which I lead to input of non-inverting driver.

I start the thing up by discharging a small capacitor into the driver input via BS170 adn a diode, lasting few microseconds to initiate oscillation. Those would simply be done by flick of a switch, which also enables the dirvers - simply like a single-pulse startup circuit.

This gave a nice safety feature by making it impossible to turn the coil on by applying power to the bridge, and it looked elegant to me.

But now I wonder will this actually work? Other coils work smoothly at unfiltered fullwave rectified input... so secondary current isn't going to decay enough to stop the oscillation during that period, I hope?

Too bad I realize things like that only after I start building.

I should draw a full diagram now but it's too late and I can't think anymore...

- My primary now looks way too small to me. I looked at other coils using really low primary inductances like that and thought it should be OK... but I don't want to run interrupted only, and as it appears now I'll have 15A of magnetizing current!

I realized that after I actually bought the wire. This makes me too frustrated.

I'm going to post a lot more after I rest and get replies - I have too many pics here, and'll gladly take more, just ask for anything you want to see!

Marko

teslaguy

Sat Nov 17 2007, 04:14AM

Registered Member #617 Joined: Fri Mar 30 2007, 07:29PM
Location: Buffalo NY
Posts: 49

marko, you are a god.

Zum Beispiel

Sat Nov 17 2007, 01:51PM

Registered Member #514 Joined: Sun Feb 11 2007, 12:27AM
Location: Somewhere in Pirkanmaa, Finland
Posts: 295

Looking good!

I see you have one of those big industrial switches there... I just love those! I've got one of them on my Beer Cooler (tm) and I just love the feeling of power it gives: Ka-Thunk! I remember being fascinated by them as a kid when I went to work with my dad. Ah, memories...

As for running unfiltered fullwave rectified: If it fails to oscillate, you can always use 555 to kick it back to oscillation. My halfwave SSTC won't oscillate without a 555 (secondary base feedback), but maybe with fullwave it won't be a problem.

Marko

Sat Nov 17 2007, 02:40PM

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

Thanks guys

I must not use 555 as it generally is because it will make the circuit unstable.

The ''proper'' way or startup would be a NE555 oscillator which would get immediately disabled by the driver as the oscillation starts.
The problem is that I haven't planned to build in something like that and won't have enough board space.... so 'll have to think something up if I end unable to oscillate.

One other thing, I saw lots of people reporting trouble with secondary base feedback. What's really up with it? I'm just drawing the diagram.

I wanted to reduce the complexity and delays to minimum but am very unsure if everything will work as I planned.

Now.. please, I need opinions on this! I use 4 of these drivers:

I just saw no purpose in 74HC14 and wanted to get rid of it.

After I push the button for turn-on I enable the drivers (active low input is pulled low) and PNP transstor is driven into conducting state. 5nF cap has been before charged to the supply voltage and is now suddenly dumped into the driver input. It's time constant with 5k resistor is 5us which is enough to get the oscillation going. After it's discharged insignifican current flows through 100k resistor.

I thought this would make neat safety feature because it makes impossible to turn the coil ON by any way except remote controller/modulator. (I don't plan using modulator yet).

If power to bridge is suddenly lost the button needs to be cycled to restart.

I use two paralelled drivers, each driving it's own primary on a single GDT to assure current sharing. 8 IRFP450's are driven in total.

Dr. Dark Current

Sat Nov 17 2007, 03:05PM

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

Marko wrote ...

One other thing, I saw lots of people reporting trouble with secondary base feedback. What's really up with it? I'm just drawing the diagram.

The problem with secondary base feedback for SSTC arises when you try to draw any arcs to ground. Two things can happen- either it just stops oscillating or worse- your bridge explodes. I've been experimenting quite a lot with sec. CT feedback on my SSTC but I just could not get it right for ground arcs.

I find the base current feedback too much unreliable and unstable. Antenna feedback fixes this.

Marko

Sat Nov 17 2007, 03:28PM

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

I find the base current feedback too much unreliable and unstable. Antenna feedback fixes this.

Wasn't it told that ''antenna feedback is unreliable and unstable'', and that ''base current feedback fixes that''?

I used CT feedback with PLL and it worked fine... but I'm not too experienced with all these things.

Two things can happen- either it just stops oscillating or worse- your bridge explodes. I've been experimenting quite a lot with sec. CT feedback on my SSTC but I just could not get it right for ground arcs.

Why would the bridge explode?

Dr. Dark Current

Sat Nov 17 2007, 03:50PM

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

wrote ...

I find the base current feedback too much unreliable and unstable. Antenna feedback fixes this.

Wasn't it told that ''antenna feedback is unreliable and unstable'', and that ''base current feedback fixes that''?

Maybe but I had exactly the opposite experience

wrote ...

Why would the bridge explode?

I don't know, mine did

Zum Beispiel

Sat Nov 17 2007, 06:12PM

Registered Member #514 Joined: Sun Feb 11 2007, 12:27AM
Location: Somewhere in Pirkanmaa, Finland
Posts: 295

I've been using base feedback for a while now with no failures...

With an antenna you have the problem that it can't be too close to the resonator. This is a problem when you want to box up the coil nicely. Sure, it may work, but seeing tiny arcs of corona shoot from an antenna, that is connected to a 5V chip, was too much for me

Getting the coil to oscillate with CT feedback was a fiddly process, but once I got it working, it continued to work (if you understand what I'm saying) Haven't tried pulling ground arcs, but from all I've seen secondary base feedback is the way to go.

Also, I'd like to add, that I've had no problems with using a 555 starter... I've set it up for Â½ the operating frequency and the signal is greatly weakened before going into the feedback system. Zero problems. This is with around 300kHz operating frequency and the classic Steve W.-style feedback driver.

Just my 2cents

Marko

Wed Nov 21 2007, 08:01PM

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

Well, all this I posted appears to be vain... interest in forum is just so low that no projects are worth posting anyway. Even much-more-famous guy's projects get ignored.

Avalanche

Wed Nov 21 2007, 10:25PM

Registered Member #103 Joined: Thu Feb 09 2006, 08:16PM
Location: Derby, UK
Posts: 845

Well I find your project threads interesting to read, and I'm sure others do even though there aren't always a lot of replies. With a lot of things like this, I think it's just a case of 'try it and see', then report your findings. Your theory looks sound, and whether or not the coil will lose oscillation between cycles will depend on a number of characteristics of your setup. I'll add some of my thoughts, beware this might not be much use, and/or completely wrong, and that's the reason I don't usually write replies

You're using this for wireless power transfer, so I'm guessing you won't want to go creating streamers, therefore it does have a high chance of maintaining oscillation between cycles - due to the high q-factor which you should try to maintain as much as possible. I'm a bit worried that it will lose it with that CT on the secondary loaded with a 1k resistor but I could be wrong!

I actually tried a similar approach about a year and a half ago with a center tapped CT, but eventually I ended up doing something completely different : I used an ordinary CT, with a potential divider on it made with a couple of resistors. I grounded the center point, and this resulted in 2 sine waves 180 out of phase. The whole idea was to not load the CT down, so I played around with the number of turns on the CT and the value of the resistors until I had 2 sine waves with a peak-peak no greater than 24v. These two sine waves went into 2 zero-crossing detectors (made with fast comparators) and the resulting signals were buffered and drove the MOSFETs through a transformer.

It worked - but the phase delay was too much and there was some hard switching. It lost oscillation when drawing arcs, but could have run forever no breakout. With breakout it was a bit unreliable, but I reckon it could have worked. I think I used smooting caps on the supply as well!

Hmm, all that was about reducing load on the CT by eliminating clamping, not really much help is it. I don't really have any more useful input other than 'give it a try'

Power up slowly with a large light bulb in the line! Is their any reason you can't add smoothing caps to the supply?

As for feedback - secondary base should be fine if the coil isn't even breaking out. Might get a more 'even' signal from the primary current though especially with that high coupling.

Looks really good anyway, excellent piece of engineering so far!

1 2 3 4 next last

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