My big Magneto repair project

[a10gf]

So, finally got the magneto beast dissected (once more), after postponing it way too long, A10 stuck in the garage. This time needs a slipring, capacitor and shimming (and very unsecure about the state of everything else...). Any good tips welcome.

Will tell how it went (either success or to have to send it away).

Before I proceed, some questions (have searched the net a lot, but not 100% sure of the values found):
What's the best value and type for a capacitor.

Primary winding gives approx 0,5 ohm, secondary approx 5,1 kohm, are these readings ok ?
e

Edit: I really like the old lucas equipment (when it works, which it does when maintained). Have evaluated, (and of course see advantages in) going 12v + electronic ignition and everything else available, but always revert to keeping the 6v, lucas regulator, dyn & mag, feeling it's part of some 'soul' this bike maybe has   and getting things to work is part of my soul

[Triton Thrasher]

Check the state of the bearing tracks.

Radial play between the cam ring and its housing and wear on the points pivot do strange things to high speed timing.

[beezalex]

What's the best value and type for a capacitor.

Not sure what the value is, the ones I've gotten aren't labeled, but they have to be high-voltage and heat-resisant, so generic caps won't do.  Bob Kizer of podtronics fame sells them here in the US at a reasonable price.

http://www.podtronics.net/magnetos.htm

[a10gf]

Excellent link, beezalex, lots of info! There I read 'A resistance of around 5K ohms indicates a good winding.' + important capacitor info.

Triton wrote:

Radial play between the cam ring and its housing

Yes, another possible source of timing errors. ok here, just moves freely.

And remembered I've got Bacon's 'Restoring Motorcycle Electrics' with it's very good k2f general restoration procedure with pictures,  incl info about capacitor\condensator value, 0,2 uF 400v.

[a10gf]

Continuing the investigations and reading Bacon's book mentioned above, I came across this, which explained a (for me) longstanding big mystery: why are magneto bearings isolated...

 "to insulate the bearings and stop current passing trough them, as this would damage the balls and tracks" 

[beezalex]

Yup, particularly with grease, it would weld little carbon pits in there.  BTW, I used mylar shim stock which comes in many thicknesses to make shims that would fit perfectly.

[Triton Thrasher]

I used paper cups, then dripped Loctite into them.

[a10gf]

Nest step, getting a spark from the armature while out of the mag body. Connecting a sparkplug between armature body and slipring output of the secondary winding, then shortly 'flashing' the primary winding with the + of a 6v battery. And yes, got sparks !

Question, has anybody done this test, and how strong a spark is to be expected?

Then in an attempt to spot any heat related failure, the armature was slowly heated to approx 75 deg.c. then redid tests when hot, cooling down and cold, still sparking.

[fido]

I used a metalised polyester capacitor and potted it into place using epoxy resin.

[a10gf]

Ok... next ! All this may be common knowledge, but new to me, and others may find it useful:

Difference in points opening gap size between top and bottom cam lobe. First eliminating armature endplay, then could it be:

- eccentric points plate
- bent armature shaft
- nonalignement of cam housing at mag body side
- eccentric cam housing at cam side
- bearing cup eccentricity
- camring outer clearing to camring housing > movement
- difference in cam lobe thickness

...and maybe even more possibilities.

The problem has irritated me for a long time, and wondered what the heck could be done to get the points gap decently equal, without destroying otherwise good parts in the process. Took the time these days to take a good look, and some good measurements, eliminating as much as possible.

The fix was easy, slight difference in insulation paper thickness top\bottom of bearing cup, in relation to the points gap difference between top\bottom cam lobe, took a few attempts with different paper thickness to get the right bearing offset, and eureka !

Browsing The Haynes manual, about the gap it says "if there is a variation the cam ring has worn unevenly and has to be replaced", no other possible causes mentioned. Well, I'm glad I didn't buy the cam first and realizing afterwards it was not at fault at all.

Important detail for manual timing control magnetos: The timing control plunger needs a strong spring, or else the cam could move when the points hits the lobes, thus pushing timing way off randomly.

Edit:
Contacts, plate and cam:

The original contacts plate, with nos lucas points from ebay, pricey but very good, points both lining up quite perfectly (pattern points I've tried were a catastrophy). Also tried a new pattern cam, but the old one was actually fine and went back into action.

Measured cam gap to housing, around .002, could be better, but much timing testing showed it had little or no impact in practice., as the cam is firmly locked downwards by a strong retard\advance spring. Points spring is well clear of camring.

Upon reassembly, got the oiling wick at the bottom of the camring inserted, a 2 parts wick, one filling the cavity in the housing, and one small length inserted afterwards trough the cam hole. Apart from some normal signs of use and maintenance after like 50 years, the plate and cam is still in very good condition.

-------------------

Edit: added text to arrows on the housing, the 'thick' resulting in moving the points plate slightly down relative to the upper camring lobe, equalizing the points gap difference on this particular mag.

[manosound]

Erling,

Does this mean that, from your list, "bearing cup eccentricity" is the chosen diagnosis, with the eccentricity caused by the paper insulator or cured by modifying the paper insulator?  

Also, I can't figure out what the arrows in the photo are pointing at, perhaps because I don't have a manual advance magneto.

Richard L.

[beezalex]

Good stuff.  I've often wondered about doing this, but it just seemed way too fiddly.  Glad someone decided to nut up and try it.

[Triton Thrasher]

Difference in points opening gap size between top and bottom cam lobe. First eliminating armature endplay, then could it be:

- eccentric points plate
- bent armature shaft
- nonalignement of cam housing at mag body side
- eccentric cam housing at cam side
- bearing cup eccentricity
- camring outer clearing to camring housing > movement
- difference in cam lobe thickness

Not the first two. They can't cause timing or gap discrepancies.

[a10gf]

Triton, yes it was an elimination job, looking trough every part of the magneto that could have anything to do with it.

Richard, nothing to do with manual or not, picture just giving some rough idea of

slight different insulation paper thickness top\bottom of bearing cup

Fixed (on this particular combination of parts making up this mag) by inserting a 'shim' by the bearing cup top (resulting in moving the top lobe of the cam downwards vs the armature\points plate).

Beezalex, glad to be of help.

--------------

I guess the whole 180 deg point gap issue comes from many magnetos currently in use being assembled from parts from here and there or are worn. I see now how even any tiny tolerance deviation somewhere, in some part of any mag, will affect the points behavior at the 2 cam lobes, f.ex., with cam just 0.01'' off-center relative to points, setting the points for correct 0.012'' gap on one side could then result in 0.022'' at the other side.

Continuing the matter, I see one can really finetune the 180 deg. points gap this way (alternative would, I think, be grinding down one side of the camring ). Am not yet down to 0.00 diff. (guess it's impossible!), but will try to get close, still have a little to go on on one side of the bearing cup before there is no place for some suitable thin isolation (cigarette paper won't do....) and of course, there is another limit to moving the bearing, the armature may touch the magnet.

Anyway, armature shimming sorted, points gap close to equal, capacitor and slipring is in the post, soon the moment of truth:

1. Jez, no spark at all
2. if sparking, will it be consistent at high rpm + heat
3. will it have acceptable L\R timing

[Triton Thrasher]

3. will it have acceptable L\R timing

You can check that now with a degree disc.

[a10gf]

You can check that now with a degree disc.

Got my special setup, see timing, running it off a drill, checking at all speeds.

btw, nice signature animation!

[a10gf]

Am thinking about how much strength may be left in the mag body magnets, and possible need for remagnetization.

How much 'resistance' should I expect to feel when turning the armature past the point of max flux, turning the armature shaft with two fingers pinching the threads at drive side ?

Yes, maybe difficult to evaluate this with words (ideally I should have a known good mag for comparison, but those mags I know about that are in seemingly good working condition are on some friend's bikes, and can't expect anybody to remove one so I can play with it!)

Thanks
e

[a10gf]

Parts are in place, slipring & brushes, shimming, 0.22uf 600v cap supposedly of the 'good' type, initial test gives strong sparks turning by hand and very good at various speeds, l\r looks close to 100%

Regarding magnetism, I suppose if one gets a good spark at slow rotation by hand, it must be ok.

Playing with the advance\retard with mag running, get approx. -16 degr. difference between full retard and full advance. What was quite interesting is it was impossible to detect any visible difference of the spark strength at the plugs, just as strong in full retard, rotating by hand or with the drill at any speed. The saying goes that the spark should be weaker at retard due less magnetic flux at that point, could really not spot any difference at all. Could maybe differ under compression though.

As an anecdote, if it is the original label, type nr shows this mag (or at least the body) started it's working life on an Ariel 500, with autoadvance\fixed camring. If anybody wants to check their mag, Lucas lists 42248E for auto and 42263B for manual (1951 parts catalog).

Pictures of the not-so-scientifical but very useful test setup, l\r timing check with strobe + the 180 deg marks on the drive pinion. Using plugs, caps and cables that will go on the bike later on so they get tested as well.

Will post some pictures and details of the inner work + tweaking later.

[a10gf]

And just for fun, was it possible to get a photo of one of today's sparks...

Some really closeup sparks : topic=1412

[a10gf]

...update & results:

-New capacitor stopped the heavy sparking at the points.

-Offsetting slipring-side bearing cup + finetuning the armature shimming equalized and stabilized l\r points opening gap, and got l\r timing spot on.

-Occasional misfiring fixed, main probable remedy the new slipring and brushes, in combination with the above fixes.

-Fixed tendency of timing to jump totally off now and then (up to 15 deg.) This was (I think) due to the points opening with the very uneven l\r gap, hitting the lobes of the cam very unevenly (one very hard, one very soft), and influencing how the points spring would control the points.

Parts presumed in advance to be defective or worn (armature, points+plate, weak magnets, worn or asymetrcal camring) were actually fine.

Test run with the drill setup for some 15 hours in total, looks stable as hell at any speed. Next week I guess I'll have time for real life test, getting it back on the engine, with good hope the sparks will be as strong under compression.