CDI for Microsquirt?

[nemiro]

Hello All-
I have a uS on my 1997 Seadoo GSX 787 PWC. I have been running it in a fuel only mode for almost a year with great success. I decided I wanted to make the leap over to full ignition control, and hit a snag. The existing coil is rather small, and is currently fired by the OE integrated CDI (I cannot separate it from the rest of the electronics). Since the goal is to remove the original computer from the boat completely, I need to come up with a standalone CDI that can be triggered by logic level, +12v, or a ground signal, so that it will be compatible with the uS. It also needs to not have a built in advance curve or rev limit, as I want to control all of that from the uS. Oh, and it has to be water tight.

I tried to trigger the existing coil straight off the uS with less than perfect results, to say the least. At 7500rpm, I have a total of 4ms cycle time (8ms per revolution, two firings per revolution, 2 stroke). This does not give enough time for the typical charge time and recovery before the next charge of an 'average' coil. Any suggestions anyone has would be most welcome.

[Matt Cramer]

MSD 6A?

[nemiro]

I have thought of that, but I need three things out of a CDI. First, it has to be water tight. Second, it needs to hit the coil fast and hard - I do not have enough time between cycles to give it the full 3.1ms+ of time it would take to charge, and then proper discharge time. A stock style CDI used in many motorcycles would work, if it did not require a VR sensor or magneto to trigger it.

[24c]

Any suggestions anyone has would be most welcome.

How about motorcycle stick coils, coil on plug types, they have a fast charge time. They are usually "waterproof" because they seal into the head.

...otherwise a DIY CDI module, schematics are everywhere on the web.

[dontz125]

Most bike CDIs are AC-powered, getting 200vac from the alternator. Not particularly usable on a vehicle with a 60vac alternator...

As 24c said, you might be able to make a bike TCI coil work; my FZR250 runs wasted spark at 18500rpm, so each coil (driving 2 cyls each) is firing every 3.2msec. The current Yamaha R6 (NOT R6S!) uses individually-controlled TCI COP at 16500rpm, or 8250 pulses per minute. That should be just right for your PWC.

[Matt Cramer]

I have thought of that, but I need three things out of a CDI. First, it has to be water tight. Second, it needs to hit the coil fast and hard - I do not have enough time between cycles to give it the full 3.1ms+ of time it would take to charge, and then proper discharge time. A stock style CDI used in many motorcycles would work, if it did not require a VR sensor or magneto to trigger it.

The basic 6A isn't watertight, but they make marine and off road versions that are. They definitely don't have much charge time - they're used on V8s with the same sort of redline your engine has. And they can be triggered off the Microsquirt ignition output.

On the inductive side, you can get a pretty hot charge on our IGN-1A coils in 2.0 msec...

[nemiro]

Thanks for the replies! I don't really want to build a CDI from scratch (although I could), but a COP setup may be the way to go. Matt, can you tell me more about your coils?

[Matt Cramer]

Sure, they can be triggered by jumpering the Microsquirt for logic level output. Were there any particular specs you needed for the coils?

[nemiro]

Are they a COP coil, or are they a coil pack that I can use to fire two cylinders at once in a wastespark configuration? On this engine, all the Microsquirt knows is that one cylinder is at TDC, it does not know which one, so I will always have to be in a wastespark arrangement, as a result.

[Matt Cramer]

IGN-1As have a single output, but can be fired in a wasted spark configuration. The Huber Mustang uses them this way, and he's running about 40 psi of boost.