Thorspark Electronic Ignition

[groily]

I've seen the screw mod done to give 5+ degrees of movement (10+ on the crank) for this very reason/circumstance. Requires drilling and tapping out the little hole, using a larger screw and machining the offset onto the end bit to get a goodly arc of movement. And probably making up a locknut which will fit snugly without looking like a carbuncle! The one I know of used a 5/16 setscrew, so quite chunky. A good mod, and a lot easier than resetting the whole kit and caboodle for the sake of a couple of crank degrees.
But actually, thinking of fine tuning on the strobe or by ear for that matter, wouldn't lots of adjustment be available on a machine with an ATD if a manual end housing were used, the appropriate Thorspark stator fitted, and the cable attachment modified to operate as a 'vernier' type adjuster a bit like the things on old Lucas car distributors? With an extra locking grub screw somewhere to eliminate the risk of a bit of chatter? Just thinking . . . dangerous pursuit, must desist.

 

[Alan @Ncl]

I think yours are great ideas to explore Groily.  

I just hate messing around with the conventional 'timing on the taper' business, primarily because before I got a new autoadvance (many years back) it use to slip all over the place.  Now I loctite it out of paranoia  but then always worry the damned thing might not come off when I want it to. I ended up making slotted bolt holes in the flange of the mag too, which is another approach I know others amongst you have also adopted. Unfortunately the replacement mag I have on now has not been thus treated as I don't have access to machining facilities anymore and found the filing option tedious and messy (at least, the way I did it was).

One of the things I liked about Thorspark is the potential ease of timing adjustment relative to the traditional option. Coupling this with the Groily vernier  suggested approach looks good to me.  I suppose all this depends quite a bit on what sort of fit the Thorspark drum makes in the end housing though as it would have to be able to rotate slightly but not be slack enough to wobble about?

Alan

[groily]

The slotted flanges are better Alan if the mag guts are being retained as, er, a magneto, as it's silly to interfere too much with the internal timing of the thing and lose spark intensity.  The mods I have seen and touched were on a V twin, that old old problem, where with a KVF the second cylinder is expected to thrive on a lousy 25 degree late spark and sometimes resents its 2nd class status. The problem is that if you try to get one cylinder to spark ahead of the 'right' moment, it won't as there's nothing there. You can't 'divide and rule' so the trick is to get the first one working strongly as 'soon' as possible, and then get the second one to fire as little late as possible in terms of the armature:magnets, before the spark has dwindled to the level of an old man's passion. Damn fool idea to use a mag designed for a parallel twin in a V, explains twin mags and expensive-to-make bevel gears on some engines, and also explains the early adoption of coil ignition despite inadequate Miller electrics by that much-vaunted purveyor of V twins from Stevenage!
But with an electronic set up, free thinking is the order of the day. And it's fun! (No idea how tight a fit the Thorspark bits are of course, but presumably they won't rattle around in a half-decent housing . . .)

[kiwipom]

hi guys, this subject has turned into an interesting exersise pitty more members have not joined in. Alan you will see from previous posts that i have only tested this Thorspark system on the bench to see that it did actually work ok so fine tuning on the bike has not happened as the bike is not finished. that extra bolt is where i have replaced the screw offset adjuster with a locater one, it could have been more discreat but i could always cut the head off, the other chrome domed nut on the opposite end is just filling the hole where the carbon brush was. Alan the idea of setting the timing/test light is good but can`t claim knowledge here or wil be able to try it yet. As a matter of interest with your electrical expertise would it be possible to run a Tachometer from the electronic ignition? Groily your surgestions of an ajusting mechanism would be a good improvement perhaps we should send our thoughts to Thorspark, they may have already thought of it,cheers
Bob     

[bsa-bill]

this subject has turned into an interesting exercise pity more members have not joined in.

Would like to join in Bob, but think I'll learn more reading what you guys are saying, as previously stated I'm interested in the Thorspark ignition (and others that are out there) and if genuine parts keep going for silly prices it could shortly possible to sell a good maggie for more than an electronic kit would cost and make a profit (ha - it'll never happen I know).
Would not consider this option on my Flash but as the RGF project is non stock in so many ways it is worth thinking about should the maggie give trouble , it shouldn't as it has been rebuilt,
And now you mention a Tachometer sensing from the plugs - these I have looked at but they do come at a price, I need to wait until the Chinese are making them.

Well that's my tuppence worth, now back to lurk and learn

[kiwipom]

hi guys, good one Bill, yes this is an information highway and like you i have gained a lot from looking and learning, Alan has also joined the club with learning that new word "Munted". It`s a sad day when you don`t learn anything, I await Alan`s opinion on the Tachometer he seams to be the electrical guru, cheers
Bob

[Alan @Ncl]

Hello again Bill Bob, Groily and others.  A few points have been raised so here is my latest 'tuppence worth', (to quote Bill).

First, although flattered to be accorded the title of 'electrical guru', I am afraid this is an overstatement of great magnitude. I was a Marine Engineer by background (so am really more comfortable with big diesels and steam plant than electronics).  The little bit I have picked up comes purely from association with others I have worked with (and its over 20 years since I did any real engineering anyway).  

So in my role as student rather than teacher, I like everyone else am learning all the time (and I have learned a lot since coming onto this forum, thanks to all of you out there).

Second point is Bob's Tacho idea.  Bearing in mind reservations about my credentials, as stated above, I think it should be possible to do this quite easily and cheaply working with the low voltage side of the Thorspark.  If it is producing a short 12 volt pulse, once every engine rev, it should be possible to pick this up and compare it against an electronic time base.  Dividing one by the other and outputting as a digital (or maybe even an analogue) signal, suitably scaled, would give speed.  I think there will be chips available at low cost to do this and we have a 12 volt supply readily at hand to drive the thing (power consumption would be negligible).  You could read off with your voltage tester or rig up a permanent display.  Sounds really promising Bob!  There might even be something on the market already made up to do most of this for us.

Related to the above, I had noted from Bob's earlier posts that his bike is still work in progress but the test I was suggesting can be performed on the bench anyway as all you need to do is connect a test lamp (or use the volt tester) and see if it does indeed turn on then off as you slowly turn the magnet past the spot on the red pick up unit.  Its the same idea as testing for a contact breaker opening except there is a much shorter dwell period.

Now to Bill's point.  As you say Bill, you can get devices that pick up speed using an inductive pickup off the HT side but as you also say, these are expensive so if we could do it with what is already available, courtesy of Thorspark, on the LT side, that may be simpler and cheaper.  Come to think of it, its another potential market opportunity for Thorspark to exploit as a 'bolt-on'?

Finally to Groily.  Yes, I miss my slotted mag already as, if I can ever get the bloody thing started (I am going back to my other thread about this soon) I think I may want to retard a bit from the current 3/8.  And it used to be oh so easy before when I had the slotted flange.  

I had not really thought much about the issue of where the optimum magnetic flux comes on a mag and the potential problems with V twins, until you alerted me to this and its a good point that you make. I have very limited practical experience with mags but from a purely theoretical perspective, I would expect the flux to build up following a nominally sinusoidal profile so ideally we try to catch it (i.e. open the points) at exactly 90 degrees and 180 degrees.  However, if we look at a Sine wave, it only drops by about +/-2%  either side of 90 degrees (i.e. Sine 80 and Sine 100 = 0.985).  So it might be more forgiving than may at first be apparent.  Even if out by 25 degrees, Sine 65 = 0.9 so its only lost 10%.  I wonder how these speculations match with reality?

Thanks again to you all

Alan

[MG]

Alan,
I talked to an old Bosch service man a few years ago. He said they used to set the mag internal timing 5 degrees late to achieve the best spark. This is to compensate the lag in phase of the current compared to tension caused by the coil inductance. You would want to open the points at the moment of maximum current passing through the primary coil.

You have to keep in mind that the ideal timing is strongly depending on the mag rpm, but 5 deg proved to be a good setting to give a strong spark at low rpm, while the higher voltage at higher rpm will compensate much of the non-ideal timing anyway.

[Alan @Ncl]

Ah Yes MG.  As usual, there is obviously more to this than greets the eye, and I know from previous posts that you are a good authority on these things.  I see now that you want to break the LT circuit at the point where current flow (i.e. magnetic flux) is highest, and that this will not coincide uniquely with the point when the winding passes the magnet, taking inductance into consideration.  Presumably as speed increases, the inductive component increases also and therefore pushes the offset even further around (i.e. further away from point of closest geometric proximity)?  Can I ask, do you know which way it goes?  I.e. if the ideal position is 5 degrees after the point of closest geometric proximity (is there a name for this point?) at say 750 rpm, what would it be at 6,000 rpm?

Alan

[MG]

sorry Alan, can't help there.
Building a level shifter to measure the spark intensity directly with the oscilloscope is on the to-do list for a loooooong time now.
The plan is to have the mag on the test stand, and check spark intensity at different rpm, to find a good compromise for the internal timing between slow and fast running.

I'm a mechanical engineer myself with in fact some, but still very limited electrical knowledge, so empirical testing it will be (if I ever get 'round to it).
Only being anal there anyway, in fact a few degrees off aren't really making much difference, like you have stated before. But interesting stuff nevertheless.

Cheers, Markus

[lawnmowerman]

I am following this one with interest - not that I am converting to electronic ignition as I am currently getting the mag rebuilt (good enough for aircraft engines then good enough for me) but there are some good learning points regarding mags.
The mag internal flux timing may explain why mine and others' A10s run on one when the manual retard is operated - especially if the mag is on the way out - as the spark will be weaker and due to manufacturing tolerances may only affect one side before ultimately giving up the ghost completely and affecting both sides.

Jim

[kiwipom]

hi jim, i get your point but just to add a bit of humour, You don`t have to kick start an aeroplane engine,cheers
Bob

[MG]

You don`t have to kick start an aeroplane engine

http://www.youtube.com/watch?v=hnd05v361qc

jump to 1:50

 

[Alan @Ncl]

Greetings again chaps
Loved the video clip and was very relieved to see him get going eventually (better luck than me in this respect).

Noted Jim is in rebuild mode and await with interest the results.  Did his mag tester mate identify any specific problem upon examination?

Also will be interesting to see any results Marcus might come up with in due course with his oscilloscope.  Practice (i.e. empirical testing) is always what matters most in the end but meanwhile, while we await the real world results, here is bit more theoretical musing for our 'delight and delectation'.

If we start with the assumption that the lag is 5 degrees at around starting speed, and then speculate an increase by a factor of about 8 to get from starting to full speed, then the inductive component of the armature impedance (which is the cosine of the lag angle, if I remember correctly) should also rise by about 8 times.  The resistive component stays the same.  On this basis, I reckon the lag will swing to about 35 degrees (tangent comes in here).  So if it had been set up optimally at 5 degrees geometric lag, then at speed it would now be an electrical 30 degrees away from optimal. Since Sin 60 = .87 this would mean you are then potentially down by about 13% from optimal, by this effect alone.  However, as Marcus says, voltage will be much higher anyway at the higher speed so you will have compensated for this many times over.  It would be interesting to see whether any of this speculation bears out in practice?

Meanwhile, it occurs to me that  there is possibly another very practical implication arising from our debate which may appeal to a broader audience.  I read on the forum somewhere that a 1 thou change in contact breaker setting equated to 2 degrees of timing.  If this is correct, we could get quite significant timing adjustment just by fiddling the points gap.  Now I think the argument against this was always that we might spoil the magnetic flux unless we stayed at 12 thou but I think that my point about the relative flatness of the sine function, around the 90 degree position, means that this effect is negligible (I think Marcus agreed this point in his last contribution). In this case we could set up for say 12 thou and 5/16 BTDC (32 degrees) then get all the extra advance we may want up to 3/8 (35 degrees), just by cracking the points open a couple of thou.

Does any of this make sense?

Alan

[MG]

Yes, absolutely. I do the final timing by strobe light, and if it is only slightly out, say about 1, maybe 1.5 degrees, I use the points gap tweak to get it just right.

Cheers, Markus