rusefi.com

Can Frankenstein do an in-line 2 cyl 2 stroke with 2 spark plugs per chamber, where the 2 plug fires about 10º after the first plug.
Each of the 4 plugs have to have separate channels, even though you could get away with waste on the primary plug, but definitely not on the second firing plug in each chamber.
Both plugs can have the same map, just the second has to fire after the first.

That would really be software since thats what orchestraits the signals?

Not possible right now but sounds like an easy change if thats something useful. What is it?

Mazda Rotary.

I had to disguise it because people sometimes get scared if they know before hand.

There is one ignition per rev per chamber on the 2 rotor engines, 13B, but they are twin plug.
You can waste the leading plug because it is always in the bottom half of the chamber, but the trailing will be in the next newly filling chamber when the crank turns 180º.

That's what I thought I've seen rotary represented as a virtual 4 cylinder. I can do that one you have some hardware wired with shaft signal on it.

Just curious at the moment.

Has rotary engine logic been implemented yet? I don't have an engine to test on but I think I can contribute to this somehow.

Joeldc510 wrote: ↑

Wed Apr 25, 2018 10:54 pm

Has rotary engine logic been implemented yet? I don't have an engine to test on but I think I can contribute to this somehow.

No one has looked into rotary since without an engine to test what's the point really?

On the other hand is not it that some 2 "cylinder" rotaries could be simulated as a four cylinder not rotary?

There's already support for per-cylinder ignition timing offsets, right?

Might this combo work?
- Set to 4 cylinder, 2 stroke (90 degrees between cylinders)
- Ignition outputs 1/2 go to the 1st rotor, and 3/4 go to second
- Set extra ignition advance to 0/80/0/80 for channel 1/2/3/4 respectively.

I think this will make outputs 1/3 fire 180 degrees apart, and 2/4 fire 10 degrees after their respective first channels.

EDIT:

mck1117 wrote: ↑

Tue May 08, 2018 5:54 am

There's already support for per-cylinder ignition timing offsets, right?

Might this combo work?
- Set to 4 cylinder, 2 stroke (90 degrees between cylinders)
- Ignition outputs 1/2 go to the 1st rotor, and 3/4 go to second
- Set extra ignition advance to 0/80/0/80 for channel 1/2/3/4 respectively.

I think this will make outputs 1/3 fire 180 degrees apart, and 2/4 fire 10 degrees after their respective first channels.

I already wrote this reply before I realized what you were doing here. I think it could work like that, but when we have direct access to the code and hardware I don't see much of a reason to not simply implement the feature properly rather than trying to work around it.




http://www.megamanual.com/v22manual/rotary.htm

According to MegaSquirt, rotary motors should be configured as a 4 cylinder 4 stroke (with 2 coils per "cylinder").

I think the trick with understanding these motors is to ignore the rotor entirely and only look at the eccentric shaft (b) and spark events, which gives 2 sparks for 2 rotations, or 2 four stroke "cylinders" per rotor. (Could this be simplified to a 2-stroke 2 cylinder?)

The only real trick would be matching coil configuration to the cylinders, but that should be a simple matter to do. Essentially you would need to have each "virtual cylinder" have two coils with a configurable timing offset (either via a fixed offset, 2 spark tables, or an additional offset table)

mck1117 wrote: ↑

Tue May 08, 2018 5:54 am

There's already support for per-cylinder ignition timing offsets, right?

yep https://github.com/rusefi/rusefi/commit/d2c0373d7588a3791f16f125ad637d35ae5d4bad