1 VR with 2 flying magnets 180 apart too complex? Really?

[racingmini_mtl]

Untimed injection is that: untimed. So you can't rely on injection happening when you want it. If you want a specific timing location, you need a setup that allows you either semi-sequential or fully sequential.

If you use a setup with simultaneous injection and the same number of squirts as the number of ignition events (4 squirts/sim in the 4 cylinder/4stroke setup) then you will have a repeatable and consistent injection location every time you start but you still have no control over the location. Moreover, each injector will be fired twice as often as they would in an alternating setup with a resulting smaller pulse width. If all the injector characteristics are set perfectly, you should still be able to get a correct fueling but with large injectors and resulting short pulse widths it becomes increasingly difficult to characterize the injector correctly as you get into a non-linear region. And this cannot be corrected for on the MS2/Microsquirt.

So while a 2 flying magnet setup is the minimum setup you can use on this engine, it does come with some compromises that can end up being a pain in the ass and not due to a limitation in the Megasquirt hardware or firmware. First, you have the lowest possible resolution for ignition timing since the engine position and RPM is only updated once per ignition event. Second, you either have a random location of each injection pulse each time you start the engine due to not knowing where cylinder one is located or you double the number of squirts and you have a repeatable injection location which is still not controlled (and possibly far from optimal) and can be difficult to tune correctly at low load/RPM due to the small pulse widths.

So you either live with the compromises or you change your 2 flying magnet setup to something better. At the moment there is an issue with the code (with the workaround I mentioned) but the real limitation is not coming from it.

Jean

[yellow-fever]

Awesome! Everything you mentioned made total sense. I realize this is a compromise with the low resolution and no fuel timing control with only 2 flying magnets, but I'm going to give it a try. If I run into any fueling issues due to injector size/pulse width simultaneous, I will try to modify a flywheel some how. Simultaneous with 2 injection events per 360 I think will work ok with the engine I have. It has a rotary valve style intake and when the piston is on the intake/compression stroke, it will have a better chance of getting at least 1 good atomized shot into the base just before the rotary valve closes the intake off, or at least when it's open. The other shot of fuel may just hang out behind the valve until the next squirt where it will be joined by the next good atomized squirt when the rotary valve is open again. I know it would be better with a fuel shot just at the right time with sequential timed injection, but with 2 squirts per 360, at least 1 good atomized shot will hit the base per 360. Any idea if the injectors fire with any relation to ignition timing? Does it follow ignition advance in any way, or is it fixed at a certain time after each trigger event? I also got to thinking that I have a high quality 2 channel oscilloscope at work that I can wire the stock CDI coil trigger signal and the megasquirt coil trigger signals to. Then I could overlay the 2 and adjust the megasquirt spark tables at any rpm range I want so that the 2 signals line up perfect on the scope. I think you could get the trigger angle and stock factory CDI spark map exactly where it needs to be. Does that sound like it would work to create a base spark map with? Wish I was this eager to learn back in school! LOL!
Thanks for all the replies everyone!

-Dan