The magneto

The magneto ignition system has always mystified me. So I was quite concerned to discover that I had no spark. Eventually I summoned up the courage to start to examine the problem. I removed the flywheel using some flywheel clamps to prevent rotation whilst I unscrewed the central nut.

 I could then remove the flywheel using a puller.

The key in the taper fit to the crankshaft fell out when the flywheel was removed and has clearly been damaged and will need to be renewed. Its a 3mm thick key and they are readily available. I took several views of the back plate and it's components before removing it from the crankcase.

Contact points

Bosch coil (black covered) is soldered to the press fit capacitor inserted in the base plate.

I could then remove the base plate of the Magneto by loosening the two screws (one top and one bottom) to remove it from the crank case itself.

Once I'd removed it I was able to check for spark using a pencil battery (see on) but this failed to produce any spark whatsoever. Using a meter I was able to measure the resistance between the HT lead (without the cap) and earth (base plate) and also the lighting output lead and earth; both of these should pass through the coils and so give me a measure of coil integrity. The lighting coil had a resistance of 1.1 Ohm and I can confirm that when spinning the Magneto by hand I could detect and AC voltage output on this wire. The HT coil gave a resistance of 6.7 kilo ohm which I believe is about right.

However, it was difficult to test the LT coils whilst they were still connected  so I unsoldered the ignition coil LT wire from the condenser  and unscrewed the two screws passing through the terminal horseshoe plates in order to remove the coil. Once it was removed I was able to test resistance etc more easily. First of all the to end horseshoe plates were continuous electrically with each other giving a value of 0 ohms resistance. Surprisingly, testing the low tension lead; the brown lead now unsoldered from the capacitor, I found no resistance between this lead and the end plates whatsoever. This is clearly not right since this wire  lead joins  to the coil and  so should have no direct connection to earth but only via the primary coil windings.  Hopefully this is my only problem. I also tested for continuity between the low tension coil and the high tension coils inside this unit. These should not actually be connected, but since both are  connected to the earth horseshoe plates they can be linked via these and so once again I obtained s value of 6.7 kilo  ohm. This is different from say a traditional battery powered points ignition system as in a car where the two coils can be tested separately. It  is obvious I was going to need to change this coil but it wasn't clear whether I should simply obtain new coil or send the whole Magneto away for reconditioning. The magnetism on the flywheel did appear better than I had suspected once the flywheel was removed from the Magneto core. However I placed it back on the core for storage.

Disconnecting the baseplate

Wear marks on crank taper. These were probably caused by failure in the Woodruff key which was clearly worn. Hopefully this will not be critical once the key is renewed.

Removing the backplate. 

Capacitor. This is a press fit and can be drifted out for replacement if required but capacitors can be fragile and I think this procedure would be risky. Luckily these rarely fail.

Removing the ignition coil- held on by only 2 screws

Pattern ignition coils of the Bosch format are around £15 from Mopedland. Its a direct fit replacement.

I was able to verify function using the spark test. Reconnect the cap and fit a spark plug. Rest the plug so its in contact with the base plate or something that's earthed to the base plate. Then flash a 1.5V pen-cell from the base plate and capacitor terminal. This should produce a weakish spark at the plug. My old coil didn't but I'm pleased to say that the new one does!

I held a pen cell in contact with the base plate, route the HT lead with plug in the cap such that the plug earths to the backplate. Then run a wire from the other terminal of the pen cell and flash that onto the capacitor connection to the ign coil. You should see a spark at the plug, although this is driven by only 1.5v so it will be weak.

For what its worth I present the following values from coils that worked.

Lighting coil: test between yellow  output lead and base plate earth  1-1.2 ohms

HT coil: measure  after removing the spark cap between end of HT lead and base plate earth  6.6 k-ohm. (My old coil which didn't work was 4.6 k-ohm).

LT coil: This can't be measured unless disconnected from the capacitor then test between the freed end lead and base plate earth, previous coil; 1.8 ohm. The new coil was lower at 1.2 ohm.

Capacitor 330 nfarad. 

Now refitted, had to adjust the position of the new coil. Adjustment is limited but as originally installed it fouled the magnets. I set it with a 7 thou gap top and bottom which equalised the space available. I also had to set the points gap to 12 thou at TDC, and set the timing. This last is very awkward because you can't access the rear plate easily to adjust it once the flywheel is on. However using a dial gauge through the plug hole and a Magneto boys magneto box I was able to set it to 2.3mm btdc. Although this was a full movement of the back plate. Once done I trimmed the HT lead both ends and attached it to coil and cap before testing for spark.... Yippee I got one!!! I then used some old 2 stroke mix and a temporary fuel tank to attempt to start... And yes I got firing!!! I can't run properly as my throttle cable is incorrect but the motor did at least run. So milestone achieved!!! I will replace the points plug HT lead and cap though as these are all very old. 

Since I fixed this magneto, I restored another using the old style, tape-wrapped coils. Unlike the bonded plastic ones above, this coil requires the HT lead be soldered into an eye mounting. I dismantled the old coil to see how this had been done. The inner wire from the HT lead was itself soldered into a narrow metal rube. This tube was then inserted and soldered into the eye. I'm guessing that this arrangement gives greater mechanical strength to the junction and also reduces the amount if solder (and therefore) heat needed to make the join. I was unsuccessful in finding a source of new metal sleeves for the inner wire, but I found a small roll pin could be opened out to allow the wire to enter. 

However, these pins are coated so I had to rub them down to bare metal using rolled up Emery paper.

Widening the slot allowed the wire from the HT core to slip in smoothly and ialso let the solder flow in well to hold it.

 

This then fitted nicely into the eye mounting on the coil. I did try to solder but the insulation resin seems to melt very easily so I crimped the eye onto the soldered roll pin and applied the minimum of solder just to the open end of the mounting eye.