Battery powered ion trap / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

Battery powered ion trap

previous 1 2 3 next

Move Thread

LAN_403

Arcstarter

Wed Jun 30 2010, 03:49AM

Registered Member #1225 Joined: Sat Jan 12 2008, 01:24AM
Location: Beaumont, Texas, USA
Posts: 2253

Use the 'ZVS' royer driver and some iron core with the proper cap etc? The ZVS can do high freq fine, so why not low.

Then use something like a bug zapper transformer, wind a proper primary... Maybe that could work, easy to run off 12v with no inverter.

Pinky's Brain

Wed Jun 30 2010, 08:34AM

Registered Member #2901 Joined: Thu Jun 03 2010, 01:25PM
Location:
Posts: 837

The driver can do low frequency just fine, the problem is the transformer. If you just use an AC transformer you are essentially running the transformer at idle ... idle current is limited by the primary reactance, which at 50 Hz is quite small with small transformers.

You want a resonant circuit on the secondary and just give it the occasional kick, not put an AC voltage on the primary.

radiotech

Wed Jun 30 2010, 03:31PM

Registered Member #2463 Joined: Wed Nov 11 2009, 03:49AM
Location:
Posts: 1546

What is the definition of ion in this experiment? Dust particles that have acquired a charge dont seem to have the same electrochemical definition.

Pinky's Brain

Wed Jun 30 2010, 05:58PM

Registered Member #2901 Joined: Thu Jun 03 2010, 01:25PM
Location:
Posts: 837

Any way, something like this works in simulation at ~5 Watt (although that doesn't take into account losses in the flyback).

Hvsine

I'm suggesting you use the secondary of a transformer like a MOT or a car ignition coil as an inductor with bunch of HV capacitors to form a 100 Hz resonator at high voltage ... the primary is left essentially open, but you use the voltage on the primary to sync a flyback which gives the resonator a kick each time it's at peak positive voltage. You need to use a flyback with a substantially lower secondary inductance than the main inductor.

The 5 watt goes entirely in ohmic losses in the big inductor, but there is no way to avoid it really if you want something which is approximately a sine. A NST or a normal AC transformer is going to burn a lot more power than this.

radiotech

Wed Jun 30 2010, 10:03PM

Registered Member #2463 Joined: Wed Nov 11 2009, 03:49AM
Location:
Posts: 1546

C3 and L1 in the simulation circuit resonate at 1000 Hz.

Pinky's Brain

Wed Jun 30 2010, 10:48PM

Registered Member #2901 Joined: Thu Jun 03 2010, 01:25PM
Location:
Posts: 837

That's what the diode and the 5 kV DC supply are for ... L1 would be the secondary of the flyback transformer being used in true flyback mode (but I couldn't be bothered to simulate it to that extent). The flyback demagnetizing current only pulses C3, which resonates with the secondary of Tr1 the rest of the time.

If you do the regulation of the flyback primary pulse length in the non sophisticated way (trim it with a pot) you could drive the circuit into a >5 kV oscillation with some nasty effects though.

radiotech

Wed Jun 30 2010, 11:38PM

Registered Member #2463 Joined: Wed Nov 11 2009, 03:49AM
Location:
Posts: 1546

In TV service, especially in the later solid state sets , flyback pulse length was measured carefully after damaged parts were replaced, because if it was wrong the new parts would burn out again fairly soon. We looked for 12 to 17 uSec flyback time. Anything else indicated ringing, jitter or other issues . (NTSC)

zhenre

Thu Jul 01 2010, 02:33AM

Registered Member #2792 Joined: Fri Apr 09 2010, 03:57AM
Location:
Posts: 3

Pinky'sBrain: I guess I don't quite understand the simulation diagram. I understand the idea of using the secondary from a typical HV transformer (MOT/NST/etc.) and resonating it though. I can see why that would have lower power consumption than when the primary is used (though it would still be expensive and heavy). I'm not sure I understand how you are driving the oscillation though. And does this produce a square wave/sine wave or something else? Is the duty cycle 50%?

I also discussed with someone the possibility of using an IGBT and large pullup resistor. This might still work and I've ordered some IGBT's but I'm worried that with the resistor needed (>1GOhm) that the RC time might become a problem and perhaps even more so I'm not sure exactly how high resistance the igbt's are in the off state, are they really better than a 1GOhm?

radiotech: they are ions in the sense that they are charged particles. The reason I like flour for a demo is that its something that everyone is familiar with so you are levitating something that is "real." They are small enough that they look like particles yet large enough that its easy to scatter light off of them with a laser pointer.

Pinky's Brain

Thu Jul 01 2010, 08:05AM

Registered Member #2901 Joined: Thu Jun 03 2010, 01:25PM
Location:
Posts: 837

Car ignition coils are a little more compact ... neither MOTs nor ignition coils are expensive. 2.2 nF 10 kV ceramic capacitors can be had for 50 cents a piece, so that part of the resonant circuit won't break the bank either. -edit- scratch that, ignition coils have way too much resistance.

The circuit is in natural resonance, and is given a kick when it's at the top of the sine. So it's a sine with a small bit of distortion. Just run the simulation :) (It's for qucs, a free simulator ... but a bit temperamental, often it won't be able to get the simulation going and you have to tweak random things to get it to go.)

The simulation used a free running pulse source, but in simulation the resonance frequency can't drift. In reality it would, and since you are only adding a small bit of energy to the system on each cycle you can't afford to add it at the wrong time. So you differentiate the voltage on the primary of the resonance transformer, and pulse on the downward zero crossing.

As for how the pulse is generated. A flyback transformer with bottom of the secondary biased to -5 kV so it doesn't start conducting from ground on the bottom of the sine. L1 and D1 in the diagram would be the secondary and rectifier of the flyback (and R3 the winding resistance).

The selection of the flyback transformer is not entirely trivial. Might have to design/wind your own. Something like 20:1 winding ratio, 200 volt supply on the primary with a 1 kV IGBT or power MOSFET for pull down (when it switches off the bottom of the primary goes to ~700 volt).

Pinky's Brain

Thu Jul 01 2010, 11:44AM

Registered Member #2901 Joined: Thu Jun 03 2010, 01:25PM
Location:
Posts: 837

zhenre wrote ...
I also discussed with someone the possibility of using an IGBT and large pullup resistor. This might still work and I've ordered some IGBT's but I'm worried that with the resistor needed (>1GOhm) that the RC time might become a problem and perhaps even more so I'm not sure exactly how high resistance the igbt's are in the off state, are they really better than a 1GOhm?

Even if the resistance was much higher ... does breakdown voltage really go up that drastically with leakage current reduction that it can suddenly hold off 10 kV?

previous 1 2 3 next

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