Electronic ignition thread / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

Electronic ignition thread

first previous 2 3 4 5 next last

Move Thread

LAN_403

Ash Small

Mon May 20 2013, 01:13PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

Proud Mary wrote ...

Perhaps it would help if you drew up an equivalent circuit for the transformer, the HT cable, and the plug, with all the different inductances, capacitances, and resistances, just to make sure it could really fire 150 pulses in a millisecond.

That's easier said than done. The flyback in a 150kHz monitor fires at this rate, but I probably want more volts and amps than these flybacks produce.

My thinking is tha I can reduce secondary windings (and associated capacitance) by dumping huge amounts of power into the primary (some of the other systems I've read about put 300V into the primary). This 'might' enable me to use a 'basket weave' secondary, thereby reducing capacitance even further, and assisting in cooling. (see other recent threads on HV transformers/coil winding machines).

I imagine the ferrite will be pushed well into saturation, but at these duty cycles that shouldn't be a problem. As I understand it, the energy is stored in the airgap in a flyback transformer, so getting the airgap right will be the biggest problem. (I think)

I can reduce the frequency if I have to, but 150kHz for 1mS seems like a reasonable figure to aim for initially.

I may be able to push more power through a slower ferrite, which will get hot at these frequencies, but with a duty cycle of only a few % it might work.

I guess I need to do some more reading on flyback topology and reluctance, etc.

EDIT: I guess a capacitor accross the spark plug could be of huge benefit in a multi-spark system. Voltage would build up until breakdown was achieved, after which a lower voltage discharge would occur.

Proud Mary

Mon May 20 2013, 02:05PM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

Here's something that might help you, Mr Ash:

Clarke, John, A high-energy capacitor discharge ignition system, Silicon Chips, September 1997.

The whole thing fits inside what looks to be an Eddystone diecast metal box, the author provides a circuit diagram and parts list, and claims it produces "beyond 1000 sparks per second", which he seems to think very fast indeed.

As this paper is fifteen years old, I expect you could adapt it to run faster using more recent silicon technology. It would be a place to start.

Ash Small

Mon May 20 2013, 02:38PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

Yes, that says ~10 sparks in 5mS, but I think I read elsewhere that modern systems produce ~10 sparks in 0.5mS. Also, I think he's using a conventional automotive coil, which isn't fast enough to work efficiently at those discharge rates. I imagine the first spark will be much bigger than subsequent sparks. (The modern systems don't have a separate ignition coil, the HT lead comes straight from the diecast box)

I'm pretty certain that what's needed is something more like the flyback from a 150kHz monitor, which is capable of firing at that rate.

That circuit does suggest putting a 1uF capacitor in series with the primary, though. I may be able to 'borrow' a few ideas from it.

Thanks for posting it, though. I couln't access the 'high res' version.

Proud Mary

Mon May 20 2013, 03:04PM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

Ash Small wrote ...

I'm pretty certain that what's needed is something more like the flyback from a 150kHz monitor, which is capable of firing at that rate.

Won't a 150 kHz monitor LOPT have diode-split construction, so that the output pulse train emerges as DC?

Ash Small

Mon May 20 2013, 03:29PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

Proud Mary wrote ...

Won't a 150 kHz monitor LOPT have diode-split construction, so that the output pulse train emerges as DC?

Yes, I'd assume so, and I'm guessing I'll require ~3 times the voltage, so I couldn't 'just use a 150kHz monitor flyback'.

DC wouldn't be a problem, though, as long as the voltage, current and speed ratings are suitable.

I've only seen a capacitor in series with the primary on push-pull topology before (full and half bridges). I can't see it working with flyback topology, as flyback topology requires that the pulse is switched off at maximum current. A series capacitor would 'snub' any subsequent voltage peak, or does it protect the MOSFET totem pole? Would a series capacitor work with a flyback primary?

Steve Conner

Mon May 20 2013, 04:01PM

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

wrote ...
In my opinion the 'knock sensor' method is not ideal because a better ignition system would allow more advance, and therefore more power (more reliable ignition at more advance)

I have an old book by Sir Harry Ricardo, where he rejects the idea that some ignition systems are "better". He basically says that the mixture either ignites and burns completely or it doesn't. If it doesn't, the result is an obvious misfire.

In other words, if the engine is already running without misfiring, improvements to the ignition system will make no difference to the performance.

Proud Mary

Mon May 20 2013, 04:01PM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

Don't you think the half-wave rectified DC coming from your 150 kHz LOPT module might view all the stray capacitances and inductances in the ignition circuit as opportunities for a bit of unofficiial filtering to get away from the raw DC look that make it socially unacceptable to most respectable circuit elements? Then where would all your seperate little sparks be - plink, plink, plink, in ze dustbin mitt dem?

Ash Small

Mon May 20 2013, 04:21PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

Steve Conner wrote ...

wrote ...
In my opinion the 'knock sensor' method is not ideal because a better ignition system would allow more advance, and therefore more power (more reliable ignition at more advance)

Sir Harry Ricardo was writing back in the days of magneto ignition. I've run magnetos before and they are brilliant, the spark gets bigger as the RPM increases. I'd require two magnetos for this engine, due to it being a 180 degree parallel twin.

Also it's been proven that bigger sparks/multiple sparks do have benefits. Racers wouldn't bother with them otherwise.

Mass produced engines are built down to a price. All kinds of improvements are possible. This is why racers modify engines and get greatly improved performance from them.

And to get back to the original point, a bigger spark/multiple spark will ignite the mixture earlier, if lots of advance without knock is what's required.

The downside is accelerated electrode erosion, which is something that vehicle manufacturers try to keep to a minimum.

Proud Mary wrote ...

Don't you think the half-wave rectified DC coming from your 150 kHz LOPT module might view all the stray capacitances and inductances in the ignition circuit as opportunities for a bit of unofficiial filtering to get away from the raw DC look that make it socially unacceptable to most respectable circuit elements? Then where would all your seperate little sparks be - plink, plink, plink, in ze dustbin mitt dem?

So is the answer to use an AC (unrectified) flyback transformer, or to opt for a push-pull AC transformer?

I assume it will be easier to reach ~50kV using flyback topology. Do the integral diodes reduce self capacitance in the flyback?

Proud Mary

Mon May 20 2013, 05:22PM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

Ash Small wrote ...

My point, not perhaps well made, is that as the LOPT frequency increases, the amount of filtration required to turn the output into 'pure' DC decreases, and this is one of the reasons that makes using these much higher frequencies attractive to circuit designers. At 150 kHz, I would expect the amount of filtration provided by stray L and C would tend to produce a 'raw DC' waveform - no clearly separate current pulses as you are hoping for.

I'm no expert on flyback technology, but there are many here who don't feel happy if they haven't fried an LOPT before breakfast, and picked the bacon from their teeth with the charred shards of an overcooked SMPS board. Ask them, Mr Ash! There's lot of 'em in here!

Ash Small

Mon May 20 2013, 06:06PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

Proud Mary wrote ...

My point, not perhaps well made, is that as the LOPT frequency increases, the amount of filtration required to turn the output into 'pure' DC decreases, and this is one of the reasons that makes using these much higher frequencies attractive to circuit designers. At 150 kHz, I would expect the amount of filtration provided by stray L and C would tend to produce a 'raw DC' waveform - no clearly separate current pulses as you are hoping for.

Well, that might not be a problem. If we put a capacitor accross the spark plug, it will charge up until breakdown is achieved. Current will continue to flow at a reduced voltage after breakdown. This is pretty much the ideal scenario. Thanks PM.

first previous 2 3 4 5 next last

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