Villiers Mk10 Flywheel Magneto Spark Autopsy

1 post / 0 new
Last seen: 1 week 22 hours ago
Joined: 04/10/2010 - 13:15
Villiers Mk10 Flywheel Magneto Spark Autopsy


My old Colwood Motor Hoe Single Wheeled Garden Tractor, powered by a Villiers Mk.10 had fallen back into regular use after restoration, working hard pushing it's two wheeled trolley loaded with 60 shovels full of road plaining material to maintain our allotment track.

On a recent session, after stopping the engine to reload it decided not to play any more.

After exhausting my extensive stock of expletives on it, it was time to get serious. I tried the obvious quick fixes of plug and points cleaning, fuel down plug hole, but still no spark.

I measured the coil and also checked the points gap, difficult at near ground level in the mud and not helped by the flywheel spokes obstructing the view, without finding the culprit.

With the complete electrics assembly removed to make access easier, setting the points was then not possible as the removed flywheel incorporates the cam profiled boss and can not be fitted to the crankshaft which is of course still on the engine.  Either the complete engine has to be removed to the bench unless you happen to have a spare engine handy. Maybe there was a special tool available to provide an easy method to set the points during manufacture.

I timed out to consult Mr Google, where lots of information was found, but scattered over numerous sites and mostly covered specifics rather than comprehensively. I resorted to my usual method, writing down my thoughts to help clarify things. These may be of use to others so are copied below. Please feel free to put me right as I am no expert.


The condenser measured OK on a Cap meter so things needed eliminating independently.

The primary measured about 3Ω, or a dead short with the points closed. The secondary about 15kΩ, measured at the coil stud rather than the HT lead to avoid any continuity fault.

It is possible to perform a crude static function test to prove if the coil is serviceable, bypassing other possible causes of no spark.

Use a 12 volt battery from a car/motorcycle/ lawnmower/buggy or whatever.

(The big square Ever Ready lantern/roadworks battery may also do, but not tried this).

Firstly ensure that the points are isolated with a piece of card put in between the contacts.

Lay the plug body contacting ground and the HT lead connected but not contacting ground.

Wire the battery negative to ground and connect the battery positive terminal to a flying lead. Watch the plug for a spark as you momentarily touch the flying lead to either the coil LT contact or alternatively the insulated tit on the cut-out contact on the back plate if fitted, taking care not to let any stray whisker find earth on the cut-out arm.

This proves the coil and plug OK. Beware that the flying lead will also arc as it makes contact, demonstrating how a capacitor would otherwise prevent points from arcing.

So, what was left​? Clearly the primary coil was not getting what it needed; either a strong enough pulse or at the correct instant - time to get technical!


The flywheel incorporates a cam lobe on the centre boss to actuate the points.

The cam measured 1.250” diameter across where there is no lobe and 1.265” where there is. This difference of 0.015” is therefore the movement transmitted to the cam follower end of the points arm, which pivots centrally so is the same amount that the points move.

It follows that if the points gap was to be set in excess of this distance, the points would never fully close when off cam. With any smaller setting, as the follower rides up the cam the points will open; the smaller the gap, the sooner and for a longer duration, consequently influencing the timing. Assuming that a spark is produced, this would then occur earlier than the ideal point before TDC, making for difficult starting, under powered, over heating, even kick back when operating the starting device or may run for just a few revs.

The gap should always be slightly less than the 0.015” movement, so set to about 0.012”.

Timing not only relates to the spark moment coinciding with the correct piston position, it can also influence the spark quality. The rotating flywheel magnets need to be in the position where they will produce the strongest flux, coinciding with the points opening and this voltage being fed to the primary coil winding. It is this change in voltage which is then induced into the secondary high voltage coil needed to fire the plug.

Incorrect adjustment could result in the flywheel magnets not being in the optimum position to generate adequate flux to induce a good voltage in the coil as the points open and the field collapses, resulting in a weak or non existent spark.


This is often given in degrees° of flywheel rotation, not so convenient if you want this as a distance on the flywheel rim and not already marked. To calculate this I measured the flywheel diameter as 6½" (or 650mm). Multiplied this by 3.14, (Remember Pi =3.1414r at School), which gave 20.41" circumference and noted it. (Optionally wrap a strip of masking tape around the flywheel, mark one rev and measure it to get the same figure).

To convert the 5° advance given in the handbook into distance on this size of flywheel. Divide 360° by this 5° (which gives 72), then divide the 20.41 circumference noted earlier by 72 to give the equivalent inch distance on the flywheel 0.283" to be exact. (A tad over ¼”).


For part of each rotation the points are closed, providing an earth path for this capacitor, but as the points open the capacitor prevents them from arcing by absorbing the current, a bit like a shock absorber. These can degrade with age and also go either short or open circuit. If you suspect an o/c capacitor it is possible to simply mount another from your motorists shop on the outside of the magneto plate connecting the lead to the convenient cut-out contact where it feeds through the plate as this goes to the same connection inside the points housing as the original. Arcing points are unlikely to kill the spark, but a bad capacitor can.

So what determines the desired capacitance value? Capacitors are measured in Micro-Farads (1F = 1,000,000µF) and usually of a value about 0.7 to 1.0µF. It forms part of a tuned circuit in conjunction with the coil winding inductance so as to momentarily resonate, subjecting the coil to what is akin to an AC, necessary to act as a transformer to produce the high tension secondary voltage for the plug.

The capacitor discharge via the primary coil gives a Back Voltage, causing a rapid collapse of it's magnetic field, inducing the momentary HT voltage in the secondary winding for the spark.

Compare the decaying ring given off if flicking a wine glass as opposed to a bucket or a thimble. Visualize a capacitor as if it were a water tank or cistern that emptied as a toilet was flushed, so enhancing the flow more than if just from the supply pipe. It then needs to fill again in time for the next flush, so if not big enough it would not provide enough to flush fully, or if too big would not have time to fill to capacity. To understand the resonance visualise pouring water into a bucket at high velocity so that it sloshed back and forth – resonating! (Remember 1/2√LC)

In our application the capacitor needs to fill quickly, otherwise the current flow would be prolonged and smooth, so degrading the fast current collapse as the points open, required to make it resonate. If it was of lower capacity it would fill before absorbing enough to give a powerful resonance and also the points could still arc to an extent.


It is difficult to see when the piston in set on TDC as the plug hole is offset. Poping a screwdriver in to detect this is impossible, but to avoid the need to take the cylinder head off a way around this is to poke a cable tie in the plug hole. Not only will this flex and even retain a kink put in near the end to help guide it into the bore, small ratchet groves found all the way along it's stem also help show movement. Gently turn the flywheel to bring the piston up and the cable tie will also rise up to indicate TDC.


Having put it all back together it fired up on first pull, so the actual culprit remains unknown, but I learnt a lot from the experience. The points gap maybe just got the wrong side of 15 thou as I filed them, so exceeding the 15 thou movement input from the cam.