No spark

All Engine, Clutch, Chains, and Sprockets Stuff Here.
Post Reply
499_cc
Posts: 4
Joined: March 24th, 2016, 12:31 am

No spark

Post by 499_cc »

This bike of mine (CR500 '87) has always been a pig to start, which prevented me to enjoinng it to the full.
All the checks have been performed as per manual, and all the items are within specs.
I did the checks with a peak voltage adaptor, so the readings should be reliable.
I got brand new stator and coil pick-up.

With the old stator, the bike started maybe after forty kicks, with the new one I was especting a two or three kicks start, but after the tenth I gave up.

The only reading I am suspicious of, is the one from the stator itself, before the signal arrives to the pickup or the ignitor; there are a few volts only, but got no reference on how much the transistors in the ignitor are going to amplify that signal, so, AFAIC it could be ok.

The checks have been performed with two available ignitors, of whom, at least one, I know for sure is working.

HT wire and cap are also brand new as is the PWK 40,5 bought as part of a strategy to eliminate possible causes of malfunction (old PJ was bored out...).

I can only think, (if I have to be positive at all costs) the flywheel might need some revolutions to remagnetize fullly?

Kicking the bike further is not an option (too much morale sapping) :( unless I know I'm headed to a solution of the problem.

Thanks
Last edited by 499_cc on May 30th, 2016, 4:39 pm, edited 1 time in total.
499_cc
Posts: 4
Joined: March 24th, 2016, 12:31 am

The ICM rant

Post by 499_cc »

Well, as it seems this request of help is doomed into oblivion, what I''ll try and do is, understand the functioning of the ICM (which fits the culprit role reasonably well) reverse engineering it, in an attemp to reproduce it with discret components.

I'm what you could call a noob as far as electrons are concerned, but I wont let this deter me (or, so I''ll try).

CDI of the era, (I will assume it's a CDI and not a TCI, but to be shure I'll have to carve into the original box itself) seem to employ an SCR to switch high currents and tensions, from the lower ones provided by the flywheel, to the HT coil, where the much sought after spark is provided.

Here the schematics of an SCR employed to switch on and off, like normal points would, and, to amplify, the current.

Image

The current flow, from the anode A to the cathode K is controlled by the gate G which acts as a switch;

So, when we apply a tension to the gate, the junction between the latter and the cathode, becomes conductive and, current flows from A to K.

Once this happens, the SCR goes into a latching mode, meaning that it keeps conducting current until power is switched off.

Albeit this paper does not want to be academic for the shortcomings (of mine) pointed out, earlier, it will strive to be clear and concise and, above all adhere to the stark CR reality.

Enter two components of the CR combo : the PICKUP COIL and the KILL SWITCH.

We (didn't we?) said, the gate needs a tension of sort, in order to give the consent to the flow of current, from A to K and from K to the HT coil; what better then, that the signal from the pickup coil (which is in tune, with the position of the piston, relative to TDC -Top Dead Center,- in actual fact starting to signal a few degrees before it -in our case 22* BTDC at 4000rpm-, so as to provide some opportune measure of advance)?

So, the signal from the pick-up coil, very small in voltage (less than 10V) should be channelled and filtered for noise removal by respectively: a diode, a resistance and a condenser (capacitor), to the gate of the SCR.

Once becoming conductive in the gate-cathode junction, the flow-path of the current, will be from the main condenser to ground, via the SCR anode and cathode and, from ground to the coil primary windings.
prolungare
In fact, the main capacitor, will have in the meanwile been charged by the action of the kickstarter through the flyweel-exciter coil combination, which produces a tension in thsincronizzazionepe region of 100VAC converted to DC by a diode before entering the capacitor.

So, we introduced the pickup-coil, now, a brief explanation of the functions of the kill-switch.

Please remember that I'm drawing from similar experiences, gathered from the net and from reverse (engineering) thinking.

The SCR once triggered from the gate voltage, goes into a latching mode, I.E. we could see current rushing from the anode to the cathode in a self sustained way, even after tension to the gate has been removed.

So, how do we stop the CR short of dropping fifth gear with no gas, brakes pulled?
I suppose we could use the kill-switch, wired in a way, inside the igniter, so as to short the A. more or less in this fashion:

Image

But, prior to that, is more important to concentrate on how to synchronize the SCR. Currentrly. I am investigating the posasiibility of the ignition, supplying an alternating discharge tension, instead of the negative only one, supplied by the SCR.
Ideally, one ought to have a "twin spark" system, comprised of a point ignition, for the fatter spark that's able to supply on starting-up a cold engine, with the electronic one, kicking in, soon after the first spark, to prolong the latter, and have superior performance and better fuel efficiency at higher revs.
...

A degree of leniency is required of the reader, because of the steep angle, the learning curve is taking, at this stage.
Thanks

ADDENDUM (last in first out)
inserted: (I will assume it's a CDI and not a TCI, but to be shure I'll have to carve into the original box itself)
Post Reply