Kill switch and Flywheel fix

The handbook for this bike has  nearly a full page on starting techniques... but absolutely nothing on how to stop the motor when you arrive at your destination! There is no ignition key, no decompression valve and although some bikes (My NSU Quick 50 for instance) go to  a fuel-starve position when the throttle is wound back and pushed past idle that too seems to be missing. If that system had ever applied to the Plover, then its not in the handbook and I can find no trace of it in the carb or twistgrip. When I asked the FB forum, the response was that these bikes are stopped simply by stalling them; brake on, in gear, idle speed and clutch out!... How primitive is that! I wanted to fit a kill switch and the simplest way is to wire the moving point directly to earth via a kill switch. This should probably be an actual on-off switch for use in case of genuine emergency, but hey, I've got a button. A button kills the ign only as long as its pressed. Consequently, the motor could restart if its released  before the motor has stopped turning so I will need to remember this! I could always stall it out if it becomes a problem.....

First step is to remove the magneto cover as already described, and then remove the flywheel using an extractor... but here comes the BIG problem. The flywheel is held on by a single nut torqued heavily to the crankshaft to hold the rotor onto its taper. The Woodruff key ensures timing not attachment.

Magneto cover removed, flywheel nut exposed.

This nut from the factory is a left hand thread. I loosened it using an impact wrench... but it wouldn't undo! Instead after an initial tension it span freely... and then the light dawned. At some point in the past someone had stripped the original nut. They had then filed the threaded section of the crankshaft down to a smaller diameter and cut a new thread on this. Sadly the thread they cut was conventional RH and they did not think to mark the  rotor with this fact! Consequently I have now stripped the repair thread on an already reduced crankshaft! This is potentially a disaster.  I did attempt to repair the thread using a Whitworth die of the same size (5/16) with apparent success although I had my doubts that it would hold. 

In the meantime I pressed on with fitting the kill switch. I connected a red wire to a spade connector and then connected this to the moving point beside the capacitor.

Kill switch wired to fixed point (red wire) beside capacitor lea (orange).

I could then lead the wire out of the magneto housing via a grommet installed behind the capacitor. This ensures that no new wiring will contact the flywheel

Red wire exits magneto housing through a new grommet lined hole.

The red wire could then be lead up into the battery compartment. 

Kill button fitted to handlebars

Meanwhile the kill button was mounted to the handlebars and both leads from the button were fed along the bars and below the tank, into the battery compartment. Here I could connect both red wires together, and connect the black button wire to earth.  

At this stage I was ready to try a road test on the bike and I reassembled everything but I was not able to get much torque on the flywheel nut. I refilled the bike with oil (gearbox and primary chain-case) and attempted to start. It refused!! The bike fired at least once on each kick but refused to run on and emitted a puff of white smoke. This was a real blow as starting and running was one thingthat the bike had done readily even before I started work. The usual checks revealed fuel in the carb and sparking at the plug. The spark was stopped by the kill switch so at least that was working, but despite all my attempts, the motor soon would not even fire and sparking had actually stopped. This left me no choice but to strip the magneto again, although I already suspected  what the problem would turn out to be... and yes it was! Removing the flywheel once more showed that the nut's torque had been insufficient, and the heavy flywheel had actually rotated on the crankshaft, shearing the Woodruff key in the process. This had naturally damaged both mating tapers but these are probably still serviceable even though I might need to grind them both together. However the damage to the crank threads looks almost terminal and clearly this is going to be a problem. there are several potential solutions but oonlyno 1 is guaranteed.

1. Fit new crankshaft, bearings and seals.

2. Drill and tap crankshaft and fit an internal screw to clamp the rotor via an adaptor. This is an approach used on mopeds like the NSU Quickly.

3. Build  the threaded section back up with weld, restore as closely as possible to circularity (I cannot machine the entire crank so this would have to be done by hand filing) and then cut a new thread in the added metal.

Number "1" remains my backstop but new cranks for these motors are no longer available and I will have to wait for a serviceable used example to come up for sale, renovate it as required and fit. This would of course also necessitate a total motor strip changing bushes and bearings as needed. Lack of a crank means I cant take this route until one is available so I left this solution for the time being.

Solution "3" is also problematic as welding on the crank risks destroying the crank magneto-side gas seal, which would again necessitate a complete motor strip and rebuild to replace it. It may be possible to weld without destroying the seal IF the crank is wrapped in ice cold water-soaked  rags, but its a risk I'd rather not take unless I have to. Accordingly I'm left with solution 2 as my first choice, however before I tackle this it was necessary to check that this magneto seal is actually intact at this stage. When the flywheel slips the timing becomes random and it can ignite the mixture through the ports into the transfer chamber. An explosion here could blow out the seals on both sides of the crankshaft, and I did experience a substantial backfire in my attempts to start. I checked the mag side seal by dangling a flag of tissue beside the  crank as the motor is kicked over. A leaky seal would cause the flag to blow away from the motor cases... luckily no leakage was observed this side at least. I'll check the clutch side at a later stage.

Luckily there is a centre drilling in the end of the crank which acts to locate a drill bit for the drilling... but the metal is case-hardened and many drills wouldn't penetrate even when sharpened. I found a cobalt drill could pierce the hardening, and once through the initial coating it penetrated relatively easily into the crankshaft itself. (Note to self... is the crank hollow?) I drilled a 3mm bore appx 3cms deep and subsequently widened this to 4.5mm. I could then tap to M5 using a hand tap and bottoming out with a second tap. M5 is smaller than I would prefer. The original nut (bsw 7/16") was closer to M8 and would have readily accommodated an M6 machine screw. However, this part has been reduced in thickness by earlier machining- it might just accept an M6 (and this remains a possibility if the M5 doesn't work) but the end of the shaft is potentially weakened and its safer to try a deep M5 first. I drilled a deeper hole than I'm really comfortable with because the longer the threaded section, the more torque the screw will be able to hold. For this, the hole needs to penetrate into the thicker sections of the crank beyond the taper where the thread will be more solid. 

Tapping the 4.5 hole in the crankshaft

I also needed to design and produce an adaptor that would allow the screw to clamp the flywheel. I settled on making a steel plug bored through to 5mm and counterbored at 8.5 mm deep enough to clear the threaded spigot of the crank. This also showed a small opening between the crank and the flywheel when it was pushed back, so I machined a raised section in the base to interlock with this gap and hopefully keep the whole plug centralised.

Adaptor plug- 8.5mm counterbore with 5mm through bore at base. Note ridge machined to engage with gap between flywheel and crank

Adaptor plug seen from above- 5mm through bore.

8

in position

Bolted on to retain flywheel

I cant find a torque setting for the original flywheel nut, the manual simply describes it as "very tight" but I am now using an M5 bolt and this will not take the same torque. Manuals suggest that the max torque for a high tensile M5 bolt is 10Nm... I didn't want to go so high as I really don't want to strip this new system, so torqued it to 7Nm. I will test and increase if necessary. I may still have the option to increase the thread size to M6 if I have to but this does risk weakening the crank threaded section..

However... and despite these efforts I could still obtain no spark. Having removed the flywheel I jammed the points open with an isulation strip and detached the primary coil connection from the points. I could then measure resistance to earth across the primary which came out at 0.8-1.2 ohms. I detached the spark plug cap and measured resistance to earth through the secondary coil and this was 4.75 kilo ohms. I also measured any continuity between the primary and secondary coils by measuring from the end of the ht lead to the primary coil points connection. Clearly there should be no connection between the coil windings, but I found this to be 4.75 kohm. This concerned me until I realised that both ends of the secondary coil are earthed; one via the ht lead to sparkplug and the other is twisted around the earth lead from the primary coil thus providing the continuity and resistance value I had determined. I dont know what the values of these coils should be, but these dont sound a million miles from correct. However I carried out a spark test. To do this its necessary to connect the earth side of the primary coil to a metal structure or the baseplate and bring the sparking plug (or naked end of the HT lead) into contact with the same conductor. You can then flash the other teminal of the battery against the other terminal of the primary coil. This should induce an ht voltage in the secondary coil and thus a spark at the HT terminal or plug. Sadly, altjough I did observe a self-induction spark at the primary coil, no spark occurred at the HT lead. Consequently, It seems likely that the coil is faulty, but I have no explanation of how it has occurred. I assume that it must have occurred as the rotor slipped on the crank taper in my attempted startup. 

I ordered a new coil (Wipac type S0155) and Ill check the resistance measurements on the new one when it comes. In the meantime I removed the existing coil. 

The stator is held to the cranksae and generator shroud by 2 screws into column nuts. These permit limited rotational movement which presumably gives some slight adjustment of timing. I noted the position of the mounting screws to try and get this at least partly correct when I rebuild.

Once these screws are removed the stator lifts off the spigot behind.

I loosened the plastic ht cable clamp which sadly broke in the process. I also detached the common earth connection of both coils shown below. This also serves as the attachment for the lubrication pad bracket. 

Once the wiring has been loosened i could examine the back of the stator where there was an obvious gap between the armature and the coil.

This could be levered carefully using a screwdriver.

Unfortunately the coil to contact breaker wire (which was already fragile) broke off. This weakness could well be something to do with the lack of spark, but the broken contact and detached wire arecshown below.

The ign coil is a tight press fit on the stator projections and could be levered off with a screwdriver. This did crack the insulation but since Im not going to reuse this coil it didnt matter. If you do wish to reuse then some protection would be needed. Coil coming off below...

... and removed.

I did solder a replacement wire to the stump of the broken contact to repeat the spark test. Again no spark was observed and again I concluded that this coil is broken.

I ordered a nos wipac coil and fitted this to the stator. However checking the resistance values of the coils on this gave values very similar to those of my old coil; 0.8-1.2 ohms primary and 4.75 kOhm in the secondary. I cant explain the lack of spark from my coil but I haven't been able to check insulation between the two coils because their earth wires are connected. Perhaps this has broken down but towards the end of the secondary coil? Alternatively perhaps the coil fly leads are compromised? In any event I replaced the coil and the HT lead.

It was possible that the capacitor, coil and kill switch leads were shorting to earth via the contact breaker post. The isulation here did seem to have thinned so I added a fibre washer to distance these connections from the post.

I fitted a new HT lead and covered this in new heat insulating sleeving as the old had frayed.

Finally I reassembled the cases and tried to start the motor. It fired up first time! It's possible the timing may need some minor tweaks as (very) limited adjustment is possible by rotating the stator. However I had photographed the position of the stator relative to the mounting screws and reproduced that on reassembly. I doubt it can be very far out.

Stator mounting screw positions are shown below.