CDI University

You da' man

That's really fun, as long as neither party has a big ego investment in the existing schematic. With touchier folks it's usually better to do it at the white board at an earlier stage of the design.

I'm all for it, as long as your computerized version has as much work put into degrading gracefully as the original!

Cheers

Phil Hobbs

(Who finally found a microprocessor-controlled toaster that he likes.)

I tend toward simple-minded...

Built around 1970.

Tricky part was sensing collector current... HV NPN Beta ~3

I can saw Plexiglas with it ;-) ...Jim Thompson

He asked me for help to do just what I'm doing. His theory on building reliable circuits is that the only component with an infinite MTBF is the one that you don't build into your circuit. So any time I said "we can take out this chunk" he'd just grin and go on to the next part of the circuit.

And yes, this is going to put me on my mettle to make sure that the micro and its software don't become a rich source of failures.

CDI ignition in cars pumps the cap up to several hundred volts, then dumps it all into the coil primary using an SCR.

Part of the reason for the micro is for various bells and whistles that increase the spark energy when the vehicle really needs it, but doesn't leave it at high power all the time, to prevent erosion of the plugs, and excessive heating of the circuit.

"CDI ignition in cars"? The posted example, run on Ford experimental cars from 1970-1971, was deemed too expensive at the time ;-)

Study this sequence:

Q1 turns on, current builds up in L1

When current reaches 5A, Q1 turns off, dumping L1 into C1, to 350V (single stroke :-)

Next time Q1 turns on C1 is dumped into ignition coil and L1 "charges" again.

In front of this was a simple bipolar chip that measured the currents, and accepted pick-up timing from a "star-wheel" in the distributor.

Featured multiple firing at low RPM's... considered desirable for smog reduction back in those days... 40 years ago :-)

"Heating"? Doesn't happen significantly in properly designed circuits. Use a PowerMOS in place of Q1 makes the current sense trivial. ...Jim Thompson

He isn't by any chance related to Earl William Muntz? :-)

I used to have a Mark Ten CDI system on my old Triumph coupe. That really helped reduce the number of unplanned stops. Good medicine.

Cheers

Phil Hobbs

t,

ty

The transistors that could survive in that environment were too expensive - at least back then. The SCR-switched capacitor discharge ignitions that were commercially successful were a lot cheaper, essentially because the SCR switch was a lot cheaper - more than enough to cover the extra cost of a Royer inverter to boost 12V up to somewhere between 250V and 400V.

We built a bunch of them in 1970 at my first job in Australia - they all worked fine, but the one's that dumped too much energy into the spark-plug chewed up the spark-plug electrodes embarrassingly fast.

At low revs, the switching spikes on the inverter had time to charge the capacitor to about twice the nominal voltage, which was handy when the battery was being pulled down by having to drive the starter motor.

We just used the standard distributor, and put 100mA through the points to keep them clean. They didn't wear when used to switch 12V into the SCR gate and ignition timing would stay good for years - without the ignition coil to sustain an arc across the points they didn't erode at all.

Sounds like a bit too much for good spark-plug life.

A suitable crummy Royer invertor will do that without any help from a micro.

Obviously Jim managed to miss the SCR-switched scheme, which lasted long enough to show up here in 2000.

The thread was "Capacitor value in CDI" and it ran from September 25 to September 27 2000.

-- Bill Sloman, Nijmegen