MicroSquirt.com read only forum (Jan.15,2018)

Hi -

I'm working on a 1200cc v-twin engine. I actually have it running, but I am completely confused on the trigger wheel setup. I am wondering if someone could suggest correct ignition settings for this application?

The engine is a 90-degree oddfire v-twin (common crank pin, 90 deg. angle between cylinders)
Trigger wheel is a 36-1 using hall sensor and hall offsetter. My incoming signal scopes out clean and MS reads it great (more later on that)
When the missing tooth is centered under the hall sensor, the crank is located 135 degrees BTDC #1.
The coils are dynaspark 3-ohm units being driven by the vb921 onboard the uS.

I come from the MS1 world, so I am having a hard time grasping the ignition algorithm and config on MS2/uS. Any help would be appreciated.


The funny thing is, I have it running now, but there are 2x as many spark pulses as I should be seeing, Megatune is showing double the actual engine RPM (or some multiple), it's strange but true! There are spark events timed about 10 deg BTDC for each cylinder that allow it to start, idle and rev just fine. Fuel tuning by ear was easy (engine is on a test stand).

I am specifically looking for Ignition Options and Advanced Ignition parameter help!

Code version is 2.886. I am confident the inputs, grounds, and wiring are sound (nothing weird going on so far).

I am welcome to all suggestions and I have attached the "running" config to this post.

thanks,
-scott

Scott,

I don't understand the double rpm. I just ran your msq exactly as is with a 36-1, 3200 rpm signal and it works fine. Each output fires once every 360 deg, and so the total no. of firings of both outputs is 4 per 2 rev engine cycle. MT reads 3200 rpm and the timing of the outputs on the scope is correct for the rpm. Are you sure you are running this the same as the msq you provided - 18 skip teeth, odd angle = 90, 2 cyl, 4 stroke ? Those are the variables that determine rpm.

You need 2 coils because you have odd-fire, but you don't have cam synch. So you have to have 2 separate coils each fired in wasted spark mode. If it were not odd fire, you would just need a single wasted spark coil.

The only problem I see is that you must have a negative trigger offset for the dual spark mode, and you have +30. That will likely cause you problems at certain rpms / outputs by not providing enough time to complete the required dwell. The skip teeth, in conjunction with the delay teeth, define the reference tach pulses. The trigger offset then locates TDC relative to the tach pulse. The code operates on a two cycle ahead basis. You want to take advantage of this to get the fullest possible dwell time, because you may need it at high rpm with wasted spark. So when the tach pulse for cyl 1 arrives, you want the next tdc for this cylinder to be about 300 deg away. Then you have 300 deg to set up, dwell and spark with plenty of advance. It's way more complicated than I would like, but it is very flexible and there are timing diagrams to illustrate this at http://www.microsquirt.info. There are also many threads in this forum on V-twin setups like this with final msqs that worked.

Thanks Al.

Correct me if I am wrong, but you're suggesting I adjust the Delay teeth, such that I end up with a -300 degree trigger offset? I did this some last night, and it actually seemed to lower the threshold on the "phantom rev limiter" behavior. I'm not losing tach signal, with my first MSQ I was able to rev to about 6500rpm, but I went as far as a -200 degree trigger offset, which was a delay of 30 teeth, and my rpm limit went lower, this was maintaining perfectly accurate ignition timing though.


The "double rpm" thing was an error on my end! misconfigured optical sensor for an external tach.

If you can confirm the approach I am taking, or correct me, I appreciate it!! engine runs perfectly otherwise, no hiccups etc.

thanks,
-scott

Scott,

If you are getting to 6000+ rpm with perfect timing throughout, and then you are running into an rpm limit, then there are several possibilities. If you datalog this, does the dwell appear to be going down as you get near 6000 rpm as compared to say 4000 rpm. If there is a dropoff, then that would indicate we need to tweak the config a bit (delay teeth and trigger offset). The other possibility is that there is too much dwell and the coils are on all the time - which will quickly burn out the VB921s. See if they are getting hot at high rpm - they will naturally get warmer, but if they are too hot to hold then that is a problem.

That makes sense! through all this I have neglected to datalog but tonight I am working on it once more and will watch the dwell, seeing how changes affect the limit. I am also running this without an alternator, so there can obviously be many things influencing my minor issues.

Tonight will be the first night I have the 'scope on it, and do some datalogging.

I'll keep this thread updated!

thanks,
-scott

Edit - we have a runner!!

I never did have a problem with tach input or dwell. Stupidly, I had a FUEL TUNING problem.

So, we have this 90-degree v-twin running with a 36-1 wheel, hall sensor, and the hall offsetter circuit - and it runs beautifully!

I did however finally come to understand the config of the wheel decoder, and must say it works just fine for us!

thanks again for your help!
-scott