2004 Sherco has lost spark

[GeoffreyLebowski]

I'm hoping someone can offer some help on my Sherco 290 which has no spark. Last time out it was troublesome to start and kicked back an ran backwards a number of times which I haven't had before. I then got it running forwards, but when I stopped it to refuel I couldn't get it to restart. Checking the spark plug has shown there is no spark when I kick it.

I Have checked over all the wiring and cleaned the contacts and earth points. I disconnected the engine kill switch to confirm it wasn't causing the issue. I have done some basic checks on the stator resistances and voltages and these seem to be within a spec I found on the splat shop website and an earlier topic. My ignition system is the Leonelli one, both the stator and the cdi unit. It is the one with the two 3 pin plugs and the seperate yellow ac wire.

Unfortunately I don't know anyone with a similar system which I can interchange parts with. Is there any way of bench testing the leonelli cdi unit. I don't want to spend hundreds of pounds buying a new unit, or stator, only to find the problem is elsewhere.

[lemur]

"have done some basic checks on the stator resistances and voltages and these seem to be within a spec"

More on that; what was the coil resistance readings, were they all the same and how are you measuring the AC voltage coming off the alternator?  

... It's not suppose to be all dirty and rusty like in there 😐 and one thing you never want to see near the magnets and coils is metal shavings that can short things out.

The ignition pickup coil produces a brief electric charge that is voltage amplified by the CDI and the ignition coil to generate the high voltage spark at the plug.  The pickup coil has a resistance check but the CDI boxes on any bike is near impossible to meter or scope test, you need a known good unit to really be sure, that's why I like to buy trials bike in pairs.  Without a replacement CDI unit you can only process of elimination troubleshoot whereby if everything else electric tests good then it must be the black box.  

Engine running backwards would have me checking to see if the ignition timing is way off or the engine is stuffed full of carbon.

[konrad]

You have my sympathies! That has got to rank as one of the worst ignition systems ever designed!

The "pickup" is actually a pair of Hall-effect sensors.

You really need specialize knowledge and equipment to diagnose these ignitions, otherwise you just have to throw parts at it.  I made some breakout cables to probe the signals and spun the system with my vertical mill.

[GeoffreyLebowski]

Konrad, Thanks for that info. I now know what is inside the pick up for the black, brown and blue wires. Did you use an oscilloscope to check for the early and late signal pulse traces?

I checked my stator resistances and diode voltage checks with my stator in situ.

I got the following results which I compared to the data on the on the splat shop site as follows.

3 way R W Gn connector: All well within limits

R to W        78.5 ohms             Spec. 50 - 100 ohms

W to Gn     586 ohms              Spec. 540 - 640 ohms

Gn to R      509 ohms              Spec. 450 - 550 ohms

3 way Bk Bn Bl connector: Readings slighly lower than spec.

Diode check. Red lead to connector pin, Black lead to earth.                     Black lead to connector pin, Red lead to earth. 

Bn                 919 mV                Spec; 1000 - 1600 mV?                              650 mV             Spec; no data

Bl                 OL                        Spec; no reading                                         648 mV             Spec; no data

Bk                 911 mV                Spec; 1050 - 1650 mV?                              650 mV             Spec; no data

I also checked for ac voltages when kicking the bike over. Difficult to get a steady reading but both W to Gn, Gn to R and the single yellow wire were all indicating instantaneously between 3 and 20 V so presume they were suppling unregulated ac.

[konrad]

Great you were able to get some test values from Chris, he's the best resource in the business!

Regarding the early and late timings, that's my nomenclature -- not official Leonelli documentation.  I was trying to figure out how the system worked.  Pretty sure I used a degree wheel, and supplied the signals marked Vdd via a 5 volt bench power supply and turned the flywheel by hand.  Hall-effect sensors can work at zero rpm.  The position (timing) of the stator plate will affect the exact high/low transitions.

The 3 identical stator coils provide power for the fan and are not related to the ignition.

The coil wrapped in white glass-reinforced tape provides the high voltage for the CDI.   There is one winding for high-rpm running and another for low-rpm running.  One end of each winding is connected together and I think also connects to the chassis.  So you should see a very low resistance from one of the charge coil wires (red or green or white) to the aluminum stator plate.

[GeoffreyLebowski]

Thanks Konrad. Some more useful information on understanding how the system operates.

You may well be right about the wiring for the red green white plug, except that I think the ground is at one end of the two coils. If under a steady state resistance check, the coils approximate to resistors my deduction would be that the system is wired as my drawn diagram. Within experimental error, the resitances measured between the pins and the measurements from the pins to ground tally.

Gn to ground     589 ohms

R to ground         79 ohms

W to ground          0.2 ohms

 

[konrad]

That wiring scheme is completely reasonable.  Ground in the center was an assumption on my part.

I'll also take this opportunity to mention that you'll need a pullup resistor (say 10k ohms) for the blue wire if you test the Hall sensors as I mentioned previously.  Search "Honeywell SS543" if you need a datasheet for the Hall sensor. 

[GeoffreyLebowski]

Thanks Konrad, It has been 40 years since I did electronics at college or played around with circuits and I've forgotten most of it.. Could you explain how I can test whether the hall sensors are working correctly. I have looked at the data sheet for them, but am not really much wiser as to what I am looking for.

I am presuming I could test on the bike and put a timing disc on the flywheel to monitor when the signal swithches.

For the "early" trace; Do I put 5v on the black wire and measure the output from the blue wire via a 10k ohm "pull up" resistor? Turning the flywheel to monitor at what degree of rotation the signal switches.

Presumably the applied voltage would be ok at about 7v ( a 6v dc battery charger)?

What voltage would I expect to see on the output? the datasheet mentions 0.4 Vdc max. would it switch from 0.4 v to 0 v

[konrad]

You can use a degree wheel if you want, but I don't think the "when" is much of an issue.  What you need to test is that the Hall switches actuality do switch.  Test each Hall switch independently by only powering them one at a time.

The top of the pull-up resistor also gets connected to the positive side of your power source.   Monitor the voltage between ground and the output of each SS543AT.

You want to see the output swing from roughly your power supply voltage to roughly 0.5 volts (or less) as you rotate the flywheel.

I would suggest using a battery rather than a battery charger as a power source.  Any battery between 5 and 20 volts would be fine.

[stevec]

I haven't heard of this being an issue with Shercos but have you checked for a sheered woodruff key?

[GeoffreyLebowski]

18 hours ago, stevec said:

I haven't heard of this being an issue with Shercos but have you checked for a sheered woodruff key?

The woodruff key is still intact and undamaged.

[GeoffreyLebowski]

I took the decision to bite the bullet, and have sent the entire ignition system to Steve at Bradford windings for assessment and repair. When I spoke and mentioned the running backwards issue, he suggested that pointed towards the timing sensors.

It was going to involve a 20 mile trip to find a suitable pull up resistor, and even if I had tested it, I then have the issue of where to go next regarding fixing it. If it was the hall effect sensors, sourcing them isn't easy and de-potting and replacing with my equipment looks tricky. If they are working fine then I have the issue of a new or refurbished cdi unit. Hopefully I'll have it back and running within a fortnight.

[RastoM]

Hi Geoffrey, do you know what was the issue? And if so how did you resolve it. 

I am having a similar issue but it did run perfectly, switched it off, removed the tank, put it back together and it has no spark.

Thanks for replying in advance 

[GeoffreyLebowski]

RastoM, I did resolve it by having the pick up timing sensors replaced on the stator. It now runs fine with no issues at all. Also changed the spark plug cap, so I'm hoping it wasn't just that. It may be worth looking at that first, or that you haven't dislodged some wiring when you removed and replaced the tank. It is pretty congested for wiring under the tank, specially if you have the full lighting loom still installed.

Edited April 29 by GeoffreyLebowski
typo