'82 450 Igniter - Repair, Replace, or Make?

[pete]

Well, skipped watching the MotoGP again tonight (still don't know the results) and gave it another burl.

No spark with W disconnected or connected to 12v, only when it's connected to ground.

However, I did notice tonight that I was getting some ghost sparks.

When cranking around past the right pickup coil, the right plug would spark, but then it would also sometimes spark going past the left coil.

I know I read something on Lou's site about ghost sparks, so I will have to revisit and see where that was and how he fixed it, but I think it was the Honda CB circuit which means I'm going to struggle without rewiring the pickup coils because the CB has four separate wires, whereas mine only has the three with the black/white in common to both.

[pete]

This is the first way I saw the HEI module simply connected with W and G directly to pickup coils, and this is for a CX650

http://webpages.charter.net/n8nxf/HEI-CX650C.pdf

He notes at the bottom he had issues coming from high RPM's down and ended up biasing what seems to be the negative common point of the pickup coils, so I may need to look at this next. Note that the end result is a bias of 1.4v to the common side of the pickup coils and W is disconnected.

I'll feed it to LTSpice and see what it looks like as I may need to drop the existing diode in this configuration.

[Matchless]

Pete,
If you compare the GS1000 HEI method you will see a note in red. I also experienced "ghost" sparks and Lou advised me to add the R2 resistor to change the bias on his method to better match the HEI on and off threshold.

On your method you may just need some fine tuning of the trigger voltage from the transistor output to the HEI.

[pete]

Pete,
If you compare the GS1000 HEI method you will see a note in red. I also experienced "ghost" sparks and Lou advised me to add the R2 resistor to change the bias on his method to better match the HEI on and off threshold.

On your method you may just need some fine tuning of the trigger voltage from the transistor output to the HEI.

Cheers Matchless.

I need to look at it again for sure, but I'm actually thinking in hindsight that I may have bumped something on the breadboard.

What I'm actually thinking at the moment is to basically make version 1 on the vero board I got the other day and get some soldered connections going to rule out any dodgy connections.

[bakalorz]

John, I tried your updated 741 circuit to no avail, and in case my signal readings were wrong in that I do get negative as well as positive signals, I tried both variations.

Then, I got so frustrated I just went bugger it and back to the beginning...

I know that when I put a AA battery's negative to W and positive to G, when I let it go it gives me a nice fat juicy spark.

I also know that the stock ignitor sparks with a positive going signal.

I also want dwell to be controlled by the pickup coils, rather than having to stuff around with biasing the HEI module if I can avoid it (I want simple stuff).

So, working with 0.08v as my input, I decided the most simplest and straight forward way is to go back to the original suggestion of a simple transistor circuit to both amplify and invert the signal.

I played around with LTSpice (thanks again for the heads up on that one John) and used the diode coupling from the German fellow's ignitor circuit to couple the pickup coil to the BC547.

I played around with random resistor values (I won't pretend to know what I'm doing here) and ended up with the attached circuit.

LTSpice showed me an inverting signal dropping from 12v to ground which is definitely right for firing the HEI module and avoiding any ambiguity.

I hooked it up on the breadboard for both pickups and coils, and just cranking by hand quickly enough fires a spark every single time as it should.

So, my question is... even though it appears to work correctly in terms of sparking every revolution, what are the drawbacks of this? Any clues?

If the amplitude of the signal increases with RPM's (which I expect it will), will that have any detrimental effects? I increased it to 1v in LTSpice and it looked ok, but of course it's hard to tell.

I will say this much though... geez it feels good to give it a crank just by hand and see those nice big blue sparks

It all depends on how the pickup is supposed to fire the coils.

It sounds like you now have it firing after the pickup voltage goes below zero, as it comes up again. if that is the correct way, you should be pretty much good as is. (if that gives a spark at the right time, but the dwell is off, you can adjust it by replacing the diode with a resistor, by picking the right value you can adjust the dwell)

If the pickup is supposed to work the other way around, the way you have it now is firing at the wrong time, and the timing will be way off. In that case you will have to get the two transistor circuit going.

I believe the reason it didn't work before is because I forgot that the pickup coil has a good bit of resistance, which I didn't think of when I suggested trying the german guys biasing. that means that the voltage drop through the pickup plus the diode voltage, never let it let the first transistor turn off. If that's true, you should be able to get the two transistor circuit to work by replacing the diode with a suitable resistor, probably 300 ohms or less, possibly 0 ohms (i.e. just short out the diode)
If even 0 ohms is not low enough, you could raise R1 in your drawing.

But one way or another, by adjusting either the resistor replacing the diode or R1, you should be able to get the two transistor amp working too.
(and I still suspect that that's the one that's probably right ...)

A far as figuring out which way is right, the easiest way would be to use a timing light and see which one gives the right firing time (although if you can adjust the timing far enough, you might be able to make the theoretically wrong way work too).

[pete]

It all depends on how the pickup is supposed to fire the coils.

It sounds like you now have it firing after the pickup voltage goes below zero, as it comes up again. if that is the correct way, you should be pretty much good as is. (if that gives a spark at the right time, but the dwell is off, you can adjust it by replacing the diode with a resistor, by picking the right value you can adjust the dwell)

If the pickup is supposed to work the other way around, the way you have it now is firing at the wrong time, and the timing will be way off. In that case you will have to get the two transistor circuit going.

I believe the reason it didn't work before is because I forgot that the pickup coil has a good bit of resistance, which I didn't think of when I suggested trying the german guys biasing. that means that the voltage drop through the pickup plus the diode voltage, never let it let the first transistor turn off. If that's true, you should be able to get the two transistor circuit to work by replacing the diode with a suitable resistor, probably 300 ohms or less, possibly 0 ohms (i.e. just short out the diode)
If even 0 ohms is not low enough, you could raise R1 in your drawing.

But one way or another, by adjusting either the resistor replacing the diode or R1, you should be able to get the two transistor amp working too.
(and I still suspect that that's the one that's probably right ...)

A far as figuring out which way is right, the easiest way would be to use a timing light and see which one gives the right firing time (although if you can adjust the timing far enough, you might be able to make the theoretically wrong way work too).

Cheers Martin, that makes sense. It's only 65 or 67 ohms across the coils, but that will definitely make a difference!

As to the firing direction (negative vs. positive), with the stock ignitor the test is to put a 1.5v source across the ignitor to fire the spark, so the positive signal should fire it.

With the HEI module, removing the 1.5v battery is what fires the spark, so that's why I'm confident I need the inverted and amplified signal.

I'll endeavour to start soldering version 1 up tonight after catching up with the MotoGP, and then I will definitely be using a timing light while cranking before actually attempting to start it to verify everything.

[pete]

Ok, watched the MotoGP finally and while this isn't thread for it, let me just say next round should be fun...

Then I got a start on version 1:





Ugly, but should be ok, and i really need to refresh my soldering skills.

I've marked out the vero board so I can cut it with the cutting wheel on the dremel at some point and I've hooked up all the wires except those that need to have spade connectors to go on the HEI modules.

All I need to do now is those two for the HEI modules and another two that need to go from the transistor collectors to G on the appropriate module.

I got some hookup wire from Jaycar along with the vero board, but somehow managed to end up with 0.2mm wire when I thought it was 1.2mm wire, so not sure if that'll be any good or not.

I'm sure I'll have some other wire around somewhere.

[pete]

And John, I get your comment now, and no, my ambition didn't outweigh my talent... at least not yet...

[john82q]

pete,

.2 will do for a test, kind of poor mans fuse, just keep it short, you will need thicker for real though.

Liked how casey handled it, all smiles and handshakes for the camera, but have a dig.

john

[pete]

pete,

.2 will do for a test, kind of poor mans fuse, just keep it short, you will need thicker for real though.

Liked how casey handled it, all smiles and handshakes for the camera, but have a dig.

john

Cool, if I can't find anything more suitable that'll have to do.

And yep, he certainly served it up well... reminded me a bit of Rossi vs Gibernau except more polite...

Also, need to get your scope back to you soon too. If I can get a chance to fire it up on Saturday I'll try to get a better look at the signal too as well as getting the timing light on it.

[pete]

Good news! First basic testing of version 1 was successful. Very rudimentary testing with hand cranking again, but no ghost spark so hopefully that really was just the dodgy bread board connections.



I obviously didn't put a lot of thought into wire lengths and was just using whatever was handy, so this will only be good for initial testing as I won't be able to mount it to the electrics tray like this:



But I was able to connect it relatively smoothly to the bike, I just had to use an aligator clip to earth it:



I'll probably need to look at the coil relay mod at some point as there's about a 0.4v drop when turning the ignition on with the circuit connected at the moment, but the spark is still big and blue.

At this stage, I'm on track to try the first start on Saturday I think and the way the modules and circuit are should be fine for a 15 minute run to give the engine its first heat cycle and check the timing as well.

[Matchless]

Hi Pete,
Excellent job! You must have that after action satisfaction feeling now!

I cannot recall them getting really hot when in use, but a simple bit of aluminum sheet should be a good enough heatsink for your testing. As long as the metal part is bolted to it firmly and some heatsink compound is used.

I used two lengths of U shaped aluminum with a HEI module bolted inside each and then sort of boxed them together with the parts potted in silicon.

[pete]

Hi Pete,
Excellent job! You must have that after action satisfaction feeling now!

I cannot recall them getting really hot when in use, but a simple bit of aluminum sheet should be a good enough heatsink for your testing. As long as the metal part is bolted to it firmly and some heatsink compound is used.

I used two lengths of U shaped aluminum with a HEI module bolted inside each and then sort of boxed them together with the parts potted in silicon.

Cheers Andre, and yep, I do believe I have that after action satisfaction feeling Something akin to after the risotto I believe, but different...

I have some compound ready, and have a proper heatsink ready also:



I need to do some work to the electrics tray, but I need to know what I'm mounting first so I can get bolt holes sorted and work out how to attach the modules to the heatsink etc.

Once I confirm timing etc. and know that the circuit is good across the rev range, then I can finalise the layout.

I'd also like to try using a more generic plug and socket for the connection to the harness so I can build two and carry one as a spare.

[pete]

Well... my first start happened today and wow, what a disappointment that was!

I don't believe it's an ignition issue, but I won't swear to it.

It won't run long enough to really check the timing, and the load of the timing light makes it die as well. I also still couldn't get a good signal reading on John's scope from the pickup coils.

The little bit of time that the timing light worked seemed to indicate I wasn't getting ghost sparks, but I can't rev or anything to see what happens either.

This is of course all info for a different thread, but I just wanted to update you all on what happened today.

On the upside, one thing I'm not lacking in is big fat blue sparks!

[Matchless]

It won't run long enough to really check the timing, and the load of the timing light makes it die as well. I also still couldn't get a good signal reading on John's scope from the pickup coils.

.......
On the upside, one thing I'm not lacking in is big fat blue sparks!

Pete,
Are you using an induction timing light? If so the only extra load will be on the 12V side and that should be negligible. Do you have good battery voltage on the coils?

You also need a good ground on the HEI modules as you are pulling about 4-6 Amps every time the coils are powered. The rest of the ignitor circuit draws very low current and can use thin wires.

If you are getting spark when turning the motor, you should be able to fire a timing light and should be able to check the timing without it running... I think!

Then maybe its not electrical?
Good luck!