Thorspark Electronic Ignition

[Alan @Ncl]

I was wondering about a variation on Jim's version as shown in my diagram  below (Battery&DualCoil).  I have an old mag with a knackered slip ring and thought about feeding the CB through the kill circuit as shown.  Could either hide the capacitor in the rotor or mount externally. Note this version does not use the HT pickups at all.  

Another possible further development of this is to add an extra stage using a £10 Maplin electronic conversion so the CB carries only a small current that drives a transistor which then drives the coil. In fact, its almost a Thorspark!  Any thoughts on these ideas?

Alan

[trevinoz]

I don't know if any of you have seen a magneto set up on a dedicated tester but it is very interesting to watch.
The spark can be made to jump a set gap or can be set so that the angle of degree of firing can be compared between cylinders.
The primary voltage on our K2Fs is in the order of 200 volts so may be a little high to apply to normal coils but may be OK for AC ignition coils.
  Trev.

[Alan @Ncl]

That's interesting Trev.  I was wondering what sort of voltage a mag primary could manage.  It must de able to push a hell of an instantaneous current if only its own internal resistance stands in the way.  I expect the inductance holds it down a bit but wonder if you have any more data around this area.  Presumably the points are having to carry all this current too.  Any idea what it typically manages at kickover?

Alan

[chaterlea25]

Hi All,
The method I used is using separate coils for each cylinder, current is generated by the rotating mag windings
this flows in parallel into the remote coil while the slipring is in contact with the relavent brush
when the points open the collapsing field in the mag  primary causes a pulse into the remote coil giving a spark
The circuit is the same as most flywheel magnetos
The important thing is that the coils Must be from something which had this type of ignition in the first place
6v or 12v coils will not give satisfactory results

The method of using the points only is back to plain battery and coil ign, capacitor can be in the rotor or externally

Regards
John O R

[lawnmowerman]

Hi Alan

Your version 2 diagram summarises what I was trying to suggest in my earlier post (but much more clearly - a picture is worth a thousand words) and I would think is the better of the two versions as it does away with the mag completely and just uses it to generate an interrupted earth to the coils using the existing HT leads and pickups so keeping the original look. Perhaps the coils could be tucked under the tank or seat front to hide them away.
We are almost reinventing the wheel here as we are getting close to the modern electronic versions which are currently available (and probably more reliable) and use contacless switching thus doing away with points adjustment and the timing inaccuracies which result if not set exactly right or become evident with wear of the contacts, pivots and heel over time.
I am trying to keep my SR as original as possible so I am sticking with the standard mag solution. I do not use the bike on a daily basis so points wear is not a great concern and once a mag is rebuilt properly and set up right, it should give years of good service.
When I dropped the mag off for rebuild my mate had a quick look and said that the HT slip ring was pretty badly grooved around the circumference with the grooves filled with carbon which is almost impossible to get out without either replacing or re-skimming. He also said that in most cases, if the old type capacitor was still being used it would be breaking down and causing intermittent faults.

Jim

[Triton Thrasher]

I am trying to keep my SR as original as possible so I am sticking with the standard mag solution.

Jim

   Even if, or perhaps especially if you want to use the bike on a daily basis, nowadays the route to reliability and riding happiness is a a standard magneto refurbished by Tony Cooper.  I too have jury-rigged a failed mag to work with coil and battery in the distant past and it did work, but with a good magneto, you have less worries and fewer roadside problems and not much maintenance either.                                                                                                                                                                                                                                        I'm delighted to see entusiasts use their ingenuity to fit super ignitions to their bikes, but those who just want to ride an old BSA twin will be better served by the K2F.

[lawnmowerman]

Totally agree TT - do it once, do it right and forget about it. When you consider how much we have probably already spent on restoring our bikes then it is false economy trying to save money by not properly refurbing the mags. They are the heart of the bike and can cause so many misleading symptons when they are not working 100%.

Jim

[Alan @Ncl]

I have just updated the diagrams in my previous post as having read John's comments, I realised there were omissions in one of them as I had forgotten the HT (now LT) slipring is not continuous but will actually only deliver a pulse over the period that the brass contact lines up with a brush.  I am also including in this post yet another option which I think is the one John is suggesting (and he has actually built one and made it work).  John, is this the way you did yours?  Also what value of capacitors do you recommend and can you fit them near the coils, out of sight, as in the diagram?  Also, do you know what resistance the coils are (or even what inductance if possible?).

Jim, I agree that the battery version diagram in the original post, it is getting close to a conventional electronic ignition but, like a Thorspark, looks original externally.  I reckon this would only consume about 1.1 amp (say 14 watts) on average if used with a Thorspark type coil.  Note it is a wasted spark type as I have drawn it. However, if wired through the HT pickups and made more like John's version, which I think was your original intention, it need not be a wasted spark.

Finally, I agree with all that a properly refurbished authentic mag is probably best especially for some of the original beauties you lads have.  But these options may be useful for skinflints like me with rather un-original looking beasts (though I would still like it to at least look like a mag).

[chaterlea25]

Hi Alan,
Your nearly there 
The wire from the primary to the remote coil in your drawing is broken by the points
This is incorrect

The wire coming from the primary to the points should be  also connected to the coil via the slipring
As the current flows in the primary when the points are closed it also flows in parallel throuhg the primary of the remote coil (via the slipring)
the current flowing in the magneto rotor basically magnetises the rotor spool (as happens in a normal mag)
when the points open the collapsing magnetic field sends a voltage spike into the remote coils primary
this spike is transformed by the remote coils secondary into the high voltage spark

I dont have an AC coil to measure the resistance here to measure, next time I visit my old homeplace I will have a look for one
The capacitor should be 0.22 micro farad, As used in all old battery and coil or flywheel mag setups
I wonder if a CDI coil would work? maybe not as the charging coils on some CDI units generate around 100volts???


In last months Classic Motorcycle magazine there was an article on energy transfer ignition, the circuit is basically the same, today  I read a letter in reply to the article in whick the writer  rekons the phase waves in the ET system
are too short because of having six magnets on the rotor, reasoning that the waves on a magneto are much longer
as there are only a north and south magnets within the casing
It makes sense in laymans language

HTH
John O R

[Alan @Ncl]

Still not sure I have got this John but have changed my original post and added a new diagram here plus a simplified version to define the basics.  Not sure if I have the condenser right either. Do'nt want to overstretch your patience but if its still not right, and if I had your email, I could send over the powerpoint file.  

Was also wondering if this does not behave more like DC rather than AC in practice as the brass slipring is a bit like one big dynamo commutator segment so the current might always flow into each external primary in one direction only?

Alan

[Stu55Flash]

Thats the cct I had in mind. The collapsing field when the points open fires a current through the remote coil.

Stu

[chaterlea25]

Hi Alan,
Yes the circuit is now correct
On a twin mag as wew are on about k2f's the current will be going in opposite direction every half revolution
the rotor poles pass the N and S magnets alternatively  ???? making AC

on a single mag one wave will be interupted by the points the next will just flow through the circuit in the opposite direction, so the  spark pulses on a single mag will always be the same polarity

HTH
John O R

[Alan @Ncl]

Thanks John and Stu.  Seems I have finally managed to accurately portray what you are saying in my diagrammatic form.  The only other thing about this is the capacitor location.  If we don't want it inside the mag, I think we could get the same effect my mounting it in the kill wire (out of sight) and earthing it (not shown in my diagram yet).  What do you think?

I am also returning to the original Thorspark issue here (have not heard from KiwiPom Bob for a while and wonder how he is getting on with his?).  I attach my speculations about how they actually get such low current but also detailing my provisional concerns about rise time and stability of timing with this device.  I should emphasise this attachment is for critiquing and discussion, not a definitive explanation.  I may put these points to Thorspark in due course, once I have established that I am not talking  crap.

Alan  

[lawnmowerman]

Hi Alan

If you put your observations to Thorspark they may well offer you a job as a consultant 

Jim

[Stu55Flash]

Alan

My reading of the Thorspark ignition is that it uses the battery to power the generation of a spark through the HT coil. So all the transistor cct does in the magneto is switch a short burst of current through the HT coil when it senses the magnet or slug on the rotor arm. Hence this current is independent of the magneto. All the magneto is there for is to mount the rotor arm to and time the spark. The magnetic cct in the magneto is not used. I don't see how the rise time of the rotor arm affect current in the cct. The transistor would be calibrated at a trigger level not necessarily at the peak of the rotor signal.

So therefore the current flow is what Thorspark have designed through their device - the little black box that is mounted inside the magneto. I suspect this would on a very short time interval but just enough ampage to provide a good spark from the HT coil. Too much and the life of the coil will be limited and/or battery current wasted. The HT coils are from CDI type ignitions and have very short rise curves [many vK/us] low reluctance coils- not like conventional DC coils or LC circuits.

Sorry but I don't think the calculations are relevant. Anyway for applications such as these the manufacturer would quote RMS values for this kind of application which would be much less (0.707*peak).

Also the capacitor is there to reduce arcing at the points in this respect it provides a short cct to high frequency components of the generated voltage and hence extends the life of the points. The discharge push is a secondary affect. 

All that said I stand to be corrected - Stu