Thorspark Electronic Ignition

[kiwipom]

Hi guys, I also have converted to ?Thorspark ignition? also to 12v dynamo. I haven?t finished the restoration yet so obviously not sure how it will work on the bike but have had it set up on the bench and it sparked well. I couldn?t see the point of having the old armature whizzing around doing nothing so I took some measurements, drew a picture and had a shaft made to take it`s place also did away with those expensive mag bearings and replaced with sealed ones which were much cheaper, still had to use insulators with bearings because of sizes. Have a look at the pics and make comments if you wish, the wiring diagram pic is the one that came with the kit, could have been computer drawn in my view   

[Alan @Ncl]

I really liked that Kiwipom.  Very professional and extremely useful to those of us who may be contemplating this route.  Loved Thorspark's technical manual too; clearly they are well into computer aided design.

Looking at this it looks as though there is no intermediate control box between the coil, and the sender unit in the CB housing, on the end of the mag. This would suggest that all the electronics is in there with the pickup. Is that really the case?  I imagined it was only the trigger that was in there (which I was guessing would be a rotating magnet and a static Hall effect transistor pickup). It looks much simpler than I would have expected. If that is the way it is done and if it does what they say, then more credit to them for design simplicity.

Out of interest, are you able to measure the coil resistances of the the primary (red and black) and the secondary (which I am guessing is from one HT lead to the other, if its a dual output coil)?  I could possibly do some sums on this to try and understand how they get such low power consumption. 

Any more close up pic of the magnet, the pickup and the coil could be handy too if not too much trouble. Thanks again for the info.

[kiwipom]

hi guys/Alan, Alan you obviously know more about electronics than i and what you say is probably right, everything is in the little red case, and yes the aliminium bit is as you say a rotating head with two magnets, i am incuding some more pics but in two posts as numbers dictate, cheers
Bob

[kiwipom]

hi guys/Alan, here is the rest of the instructions,Alan i am not electronics savvy so hope the pics will answer your questions, if anyone wants AutoCad 8 file of shaft i can e-mail it then you can play/adjust at will,the 2D dwg is 239kb, the 3d dwg is 519kb,cheers
Bob

[Alan @Ncl]

Thanks Bob
That is great.  Exactly what I wanted and very good of you to do all this.

I am not an electronics engineer but I can now do a bit of crude analysis based on this data and at least come up with some reasonable questions (possibly).  Meanwhile, others on the Forum might feel like doing the same and no doubt some may be better qualified than me to do this.  

To recap, what I think this tells us is that the coil has a 5.5 ohm resistance so will pull about 2 amps when turned on (12 volt system). The HT side is about 17.6 kohms (which roughly suggests about 36,000 volts, I would guess).  

The red box probably contains a Hall effect transistor which is basically a switch which turns on as the magnet approaches, thereby starting the coil charging, and turns off again as it passes.  This in turn triggers the HT spark on both cylinders together, on each engine revolution, using the wasted spark principle. The timing of the spark occurs as the trailing edge of the magnet passes (the equivalent of the contact breaker opening) so it may be possible to set the timing with a 12 volt lamp, just as we can with a mag (although obviously a strobe is probably better).

I will report back if, and as soon as, I have any more but it will be interesting to see if the guys think this makes any sense.

Alan

[kiwipom]

hi Alan, as a matter of interest do you Geordy`s not sleep? it must be halfway throuhg the night, it`s 2.45 inthe afternoon here,cheers
Bob

[bsa-bill]

Useful post Bob, I think I'll print it off for future reference as I halve shelved the Thorspark idea for the present as both maggies seem to be performing.

btw it's Geordies and it's difficult for us to sleep at night when Sunderland are four places above us in the league

[Alan @Ncl]

You'r quite right Bob.  The Geordie nation never rests.  Thing is, time you get in from the pub and wait for the wetsumping to run its course, its getting on towards one o' clock. And I have a confession for Bill.  Despite living in the great city of Newcastle since 1978, I was born and raised in Sunderland!  How is that for truly cosmopolitan?

Now back to Thorspark.  I did a little check on the data (and can produce the details if anyone is interested) but here is the upshot. Looking at Bob's data I am guessing the diameter of the Thorspark Ally rotor is about 12 mm and the magnet about 3mm across.  Say the engine is running at full chat, 6000 rpm.  On this basis I reckon the coil only gets about 1.6 msec to charge.  This is a very short time as the stuff I have seen on the web seems to suggest that ideally you need about twice this long (the time constant of an R/L circuit is resistance times inductance but we can leave that for now). At 1200 rpm, it gets 8msec which is ample of course so this seems to suggest good fat starting sparks but possibly tailing off a bit at speed.  Remember my initial curiosity was fired by that add for a second hand unit being dumped on ebay because the owner blamed it for mid range misfiring (rightly or wrongly).

However, if my estimates above are correct, this does indeed support Thorspark's claim that the unit will only consume an average of 0.25 amps (my crude estimate is about 0.35 amps but I have made a few speculative assumptions in here). In fact, all my statements should be treated as speculative as anyone who can find good words to say about both Sunderland and Newcastle is obviously a two faced devious scoundrel.

If I can prevail upon Bob's good will with a few more requests:
1  Could you confirm/correct my assumptions about the rotor geometry?
2  I see the instructions lightly refer to the removal of the slipring (camring).  How easy is this in practice? I had a little poke about and it did not seem very anxious to be disturbed.
3  I see the instructions also require the body of the coil to be earthed. I assume this is for the LT circuit as with a twin output coil, I think the HT is a closed loop through both plugs and the cylinder head.  Could you just stick your tester on between each of the four terminals and the coil body to see if any of them are showing any resistance readings?
4  I also notice that to adjust the timing, you have to loosen the centre bolt and move the rotor.  This implies it does not have the little key as in the CB unit and sounds fiddly.  Are these assumptions correct?  Conversely the pic shows a couple of crosshead screws holding the sender unit and I imagined they would be perhaps slotted to permit some adjustment here instead (sound potentially easier).

Thanks again and best wishes to all.

Alan

[bsa-bill]

In fact, all my statements should be treated as speculative as anyone who can find good words to say about both Sunderland and Newcastle is obviously a two faced devious scoundrel.

The fact you own an A10 allows forgiveness

To those of you wondering  Newcastle ( Geordies) and Sunderland ( Makems) share a sporting relationship akin to Australia and England Cricketers, usually friendly banter, match day and Newcastle Brown fetches out the worst in a few hotheads unfortunatley, none the less it good to see both teams doing well so far

[kiwipom]

hi guys/Alan Bill, thanks for your humerous note, anyone who comes from the same place as Bobby Charlton can`t be bad, i digress. (1)The rotor is 12.7mm with as you correctly say 3mm magnets, (2)as for the slip ring from memory it was no problem with the end piece removed, a little heat and away it should come, if all else fails brute force and ignorance works every time,lol.(3)there was no resistance between earth and any of the other four terminals,(4)you are correct that there is no key to where the rotor sits on the shaft/no adjustment on the sender holding screws, and yes it would be a good idea to have them slotted for adjustment however moveing the rotor is probably easier. I am thinking of sending these comments to the makers of this unit for their advice,thanks guys,
Bob

[Alan @Ncl]

You have delivered the goods yet again.  Thanks Bob (thanks also to Bill for his words of reconciliation).

Just about to turn in so here is tonight's shot.
1  With the exact measurement I now reckon the charging time is about 1.5 msec at 6,000 engine rpm (minor change).
2  In this setup, I don't really see why the coil has to be earthed to work (if it is not completing any circuits which is what I expected) but it can't do any harm and seems intuitively sensible.  Forgot to ask what amperage fuse they give you?  Also how much clearance is there between magnet and trigger unit?
3  Tried warming the camring on my old mag but its very tight. As for the brute force and ignorance, I have an abundant supply of the latter but cant see how to apply the former as I cant get behind it to pull it or whack it?
4  I am guessing you might be able to check the timing as shown in my diagram below but may be wrong as Thorspark do not mension this method. I would expect it to turn on as the magnet approaches then turn off again, very shortly afterwards, as it passes, triggering the spark as it goes.
5  I was also thinking of talking to Thorspark as soon as I feel far enough up the learning curve.  Perhaps we can perform a 'pincer movement'.  So far, I am quite impressed although I still have a couple of niggling doubts.

Over and out

Alan

[kiwipom]

hi guys/Alan, very impressive dwg Alan, Thorspark could use it.  Removeing the camring from a spare mag was easyer than i remember with the housing removed of course(pics included)no heat required just a few gentle taps with sharp tool in the two small holes that are in the ring itself. A gap of 15thou between trigger and rotor. Fuse is 15amp, i think that has covered it,cheers,
Bob

[groily]

Underneath the little removable bobble on the underside of the cam ring housing (see lower pic), I think you'll find a lock-nutted small screw with an eccentric tip engaging in the cam ring Bob- which is there to give very fine adjustment of the ring-to-mag-body on initial assembly to ensure that points-opening and point of max spark power coincide. Not saying you have to, but I've tended to remove that before pulling the ring!

[kiwipom]

hi groily, yes you are right but this one is munted, i was just demonstrating to Alan that it does come out easy and yes Alan will have to note that,cheers
Bob

[Alan @Ncl]

Thanks Bob and Groilly.  The Kiwi method worked a treat and it is out.  Noted that screw but thought it was locating rather than holding (I've learned a new word here too, 'munted').  I definitely think the Thorspark instructions could be a bit more explicit about this as there must be others out there like me, with limited imagination or initiative, who would benefit from the advice you lads have given.  In due course, perhaps we may offer them the benefit of our 'collective wisdom' for inclusion in their documentation (I expect they will love that).  I am going to update the diagram by the way so can you check to see I have it right please?

I was thinking about this screw business (noted Bob's lovely gleaming mag seems to have an extra bolt in there too). It seems to me that if you could move the stator a little bit within the casing, possibly even with the engine running under strobe, it would be a good way to fine set the timing then pinch down the screw when finalised. I would have thought it might be quite tricky to get it just right by moving the rotor on its taper and keeping it there while you tighten the centre screw.  This said, it does not seem you (Bob) had any problems with this.

Also wondered what you thought about my suggestion to use a timing light in the switching wire, as in the diagram, to set the timing without a strobe? Perhaps Bob might like to try it and comment if it works?