Fattening-up an HT spark

It's Paul Burridge ;-)

...Jim Thompson

Cripes! 1.8 Joules, as opposed to typical 65mJ... but a 65mJ CD ignition WILL make a juicy spark... just not a plasma (which is why your "gap wear is significant" ;-)

...Jim Thompson

Back around 1970, a bunch of us made our own CDI systems, all slightly different, and it turned out that 250V and IuF (31.2.5mJ) or 300V and

Jim is talking nonsense about the spark/plasma distinction. There is a real difference between the spark - the initial breakdown of the gap - and the subsequent discharge, but essentially all the energy that is deposited in the gap is deposited during the glow discharge period, which goes on for a great deal longer.

Almost all the energy is - in any event - dissipated at the electrode surface, by the positive ion bombardment that generates the secondary electrons that cary the current through the gas. More current does deposit more enegy in the gap, and it helps if you let the induction coil ring down so that most of the energy stored in the coil ends up at the spark gap, rather than being recycled into the capacitor for the next spark.

---------- Bill Sloman, Nijmegen

I ran about 80 mJ on my motorcycle CD's, but the bikes got wrecked or stolen often enough that plug wear wasn't a big issue.

John

Like this?

The plug fires via a conventional ignition coil. Connected to it is a diode, inductor and a 10uF capacitor charged to about 600V The diode must hold off the HV from the ignition coil and carry the current from the capacitor. The inductor limits the current. The spark plug must not have a resistor inside. The picture is a photograph of a single shot. The gap wear is significant. I have not tried fancy electrode materials.

Regards,

Boris Mohar

Got Knock? - see: Viatrack Printed Circuit Designs (among other things)

void _-void-_ in the obvious place

The electronic ignition system I've been developing (just for a bit of excitement) is firing up in a timely fashion; the only downer is the spark's a bit on the weak side. Is there any way of fattening-up a spark from an HT coil driven by a switching IGFET so it's nice and chubby and blue rather than skinny and sort of yellowish? I've got some half-remembered concept about using a capacitor in series with the coil that resonates with it in a Teslaesque fashion, but am stumbling about in the dark here, quite frankly. Any suggestions, chaps? p.

^^^^ Correction! I meant IGBT, in fact.

Increase the peak current in your coil (leave your IGBT on longer or use a higher voltage).

Cheers, John

Does it fire the plug at running pressure? If so, there's enough voltage.

I assume this is not a CDI design but stores the energy in the inductance.

Does your IGBT have a high reverse blocking voltage? It should be able to block a few hundred volts in each direction when off.

Just about any spark is as good as any other spark for engine ignition. If you want the spark to be more impressive, keeping it going a bit longer can help.

Instead of just pulsing the IGBT off, cycle it several times at just a little under the resonant frequency.

I could but I won't. This is not out of malice but kindness. Since you have embarked on electronics as a carrier, you need to learn how to learn. I have given you enough clues there for you to figure it out. When you do, post the circuit on ABSE ng.

No can do, John. Some bastard in China might copy it and mass produce it. Next thing you know, it'll wind up in the next car you buy. And you really don't want to be driving around in a car with an ignition system designed by me, do you? I'm sure ol' Jim wouldn't! ;-)

Um, yeah.

Thanks, Boris. Sounds just the job. But I'm having difficulty with the above description. Can you post the relevant circuit fragment?

Working in a lab that uses a lot of strobes, I needed a cheap spark on demand circuit. I dump a AC line powered voltage doubler into a ignition coil, so about 380V at ~ .5 uF actual capacitance. This makes a nice fat "hot" spark with a short gap. Two 1K 1 watt resistors limit current into the doubler and a SCR tripped by a optocoupler does the firing, off a 5V signal from a frame grabber board. The opto is powered from a small amount of charge stored on another capacitor and clamped by a 5V zener. This has proved so useful around the lab that we have done a stock PCB for it. I dont have a web site, but I could send it to Jim Thompson if he wishes to host the circuit and layout, its in eagle PCB format.

Steve Roberts

Send me a PDF (or some other common graphics format) and I'll take a look. (I don't have Eagle.)

...Jim Thompson

Hell no! I was trolling. I've been retired for 15 years (and electronics is just a hobby for me which I'm not particularly good at anyway but just enjoy a challenge).

Paul Burridge wrote:

Get an ignition coil driver IGBT. Made for the purpose. It will have a high enough standoff voltage, which could be your problem or part of it. I used HGTP14N40F3VL in some ignitions I made, it's a good igbt but I don't think it's any longer available from the major suppliers. I have some MGP15N40CL that I ordered because I couldn't get the Fairchild ones, but I haven't used them yet. If you want to make sure your ignition lasts, use the right parts. These coil driver IGBT's have clamping. They will work in this rather nasty application and they should last (car manufacturers use them). I wouldn't use vanilla IGBT's because you would have to add the extra circuit elements for protection. You're not paying significantly more for an ignition coil driver anyway. I'm an amateur and I built and sold a few points-triggered electronic ignition modules to my fellow antique motorcycle enthusiasts. The spark isn't necessarily that much better than you get from points, if the points are properly tuned... but you can run more current with the igbt than you can with points, and that will get you a bigger spark. Plus you eliminate the hassle of points (I hate points). How do you turn your IGBT on and off? I just used a pull-up resistor on the gate and a small npn transistor controlled by the points that pulls the gate to ground when the points open. I had a pet project on my own bike. I took the points out and installed magnets, hall sensors and a shutter disc. It ran quite well but the engine heat caused the hall sensors to fail after about a year. Here's a link to a photo

Interesting. I'll have to look up 50 kV diodes, see how much they cost. With the right inductor, could you eliminate the diode? I mean use your inductor to block the hv shot from getting into the storage cap; kill two birds with one stone.

(snip)

It isn't safe to speak ill of the God of Darkness.

Hadn't occurred to me, to be honest. I've been driving the IGBT with a 2N7000 D-mos device.

Okay, thanks. Can't really see the gubbins in that shot as your stator plate (I assume it is) is fouling the view. I once fitted a Boyer Bransden electronic ignition to a Bonneville, which proved to be a very stressful experience. Mounting the stator plate (which had magnetic pins for the hall-effect sensors) always seemed to result in actuating the sensor, resulting in a huge backfire from one or other of the cylinders. After repeated attempts at static timing and several changes of underpants, my nerves were quite shattered. Still, anything's got to be better than Lucas parts, I guess.