Newbie: ignition only setup on a small 4-stroke single

[saNgai]

hye.. newbie also here, can u post the full picture how the wiring lookalike
and then test result, really appreciate it thanks.
actually i really not understand the teeth part..

[tech3sandcars]

Please reveiew the other 2 threads going on, " 2009 harley davidson" and the " need help with Vtwin" lots of good information on those.

[grippo]

Also read http://www.microsquirt.info/ - the dual spark link on the left side scroll bar and
http://www.megamanual.com/seq/multichannel.htm

[mfro]

Hi everybody!

It's been a while since I posted here. I just wanted to give some feedback about the progress I've made.

DSC_2809.jpg (21.6 KiB) Viewed 1690 times

That's the finished generator & ignition setup for my single cylinder 153cc engine. It showed its first signs of life just yesterday . I'm currently running an ignition-olny setup to get a decent starting point before trying to setup EFI.

It's still not perfect (I have stable idle and can easily rev to about 6000 rpm where the engine starts to suffer from misfires as if timing retards) and I guess I'll need to now dig into VR sensor distance adjustment and look after whatever timing circuit optimization tricks described in different threads.

Just wanted to say thanks to this forum - without the hints available here I wouldn't have been able to come this far. Thanks!

Regards,
Markus

[myk777]

Have you looked into using a gear tooth hall effect sensor instead of a VR sensor, I can't for the life of me figure out why so many people continue to struggle along with VR sensors when a hall is so simple and effective.

Mike

[mfro]

Have you looked into using a gear tooth hall effect sensor instead of a VR sensor, I can't for the life of me figure out why so many people continue to struggle along with VR sensors when a hall is so simple and effective.

Mike

I really wouldn't have known what to answer to this question two hours before when I was close to frustration with misfires and false triggers.

Then the parcel from the electronics shop arrived with my 10kOhm resistor.

No joke: five minutes later the engine changed completely and now it behaves as it should. Revs easily into 5 digit rpms and no misfires anymore .

[mfro]

I do have a question that bothered me the last few days.

With a wasted spark setup, what boundary is a "trigger - dwell - spark"-cyle not allowed to cross? What numbers add up to an ignition cycle's time?

I did observe rev limiting behaviour (like retarded ignition) above about 10000 RPM now that the new enginie is broken in. Tried to move the tooth gap around by adjusting the trigger wheel which helped a bit but not much.

What helped, eventually, was to reduce dwell. Dwell for my coil is originally specified to 2.5ms, by reducing it to 1.7ms the engine still runs good but now easily revs 12500 RPM and more (didn't dare to try where the limit moved to now with an enginge still new).

I have a wasted spark setup with one single coil alternatively charged by IGN1 and IGN2. The tooth gap/trigger point (18-1 wheel) is located roughly at -180°BTDC/+180°ATDC. Max advance is about 30°.

With the observed behaviour, I'd assume that IGN1 and IGN2 cycles overlap above a certain RPM leading to interleaving charging/discharging cycles for my coil.

I did some rough calculations adding up times for advance, spark and dwell, but didn't really find an explanation.

Bildschirmfoto.png (22.56 KiB) Viewed 1655 times

Although 2.5ms dwell equals to 180°at 12000RPM, there should still be plenty of time without disturbing the other IGN output.

What am I missing?

[grippo]

In dual spark mode, Microsquirt calculates each ignition output timing beginning at the start of a tach pulse (-trigger offset ATDC) and the dwell and spark, including a min spark duration, must be completed within 2 tach cycles. The factors which affect this are:
- trigger offset
- user specified max dwell, max(really min) spark duration, and ignition advance
- wasted vs sequential mode
- whether the ign outputs are separate or tied together (only possible with 3.x code)

If you can conveniently set trigger offset to a small negative number, say -5 deg, that will give more time because the dwell can't start until the tach pulse arrives. With min trigger offset, outputs tied together, and sequential mode you can spark every 720 deg and get a window of 2x 720 = 1440 deg in which to dwell and spark. If you use wasted you get a window of 720 deg, and if you have separate ign outputs the window becomes 360 deg with a spark every 180 deg, every other one being unused since ign2 is unconnected. Timing diagrams for your configuration (config 4b) can be found at http://www.microsquirt.info/dualspark.htm

If you are interested in tying the 2 outputs together, it would be best to use the latest 3.7x code which we are going to release as soon as the production run for the new microsquirt is complete, but this code will work on existing microsquirts. If you want to start sooner I can send you the code download and documentation for it.

[mfro]

Thanks Grippo!

From the linked page, I actually do have configuration 5a currently (Skip teeth=M, two outputs tied together, wasted), but I understand I might as well use config 5b with my setup, avoiding any possible interleaving problems of cycles. Will try at the weekend.

Thanks again!

[grippo]

5b should make things worse because the tach cycle (Skip_Teeth) is only 180 deg (M/2). Since you have already tied the outputs together, only going sequential can help, if you are really at the limit. If you post your msq I can look at the ign outputs on a scope and tell better what is going on.