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

Try the start settings i said.

You could up the fuel pressure but since you are only using low rpm, it doesn't matter. They are just open longer all the time.

You won't use anymore fuel because you tune the VE table to what fuel you need. Firing the injectors every rev will fix your firing vs stroke problem. It is a common intake so what ever valve is open will pull fuel from the whole intake (both injectors). You just want both valves closed during injection or one just opening.

You just need to time the injection event right, and inject on the exhaust strokes (BTDC intake). Try between 50-75%. And pick your smoothest idle.

Injection occurs at tach. Yours is set at 90* (9 teeth) after missing teeth which is TDC. You should delay to get back BTDC and not start injection on valve overlap. Since you have a low rpm engine, your cams should be timed near TDC. (you want to inject before the intake valve opens) A delay of 100% or 360* from your tach will put injection back at TDC on the opposite stroke.

But if you have an odd angle engine???? #2 TDC will occur before this point. (2.9 code always fires #2 injector 360* after #1). So take your cylinder offset and minus from #1 delay. This will fire injectors at TDC #2 and your fire #1 x* before #1TDC. (x = cyl offset). Injection delay is a percentage of 360*. So if you have a 60* vtwin. If you delay injection 300*, #1 will fire 60* BTDC and #2 will fire at TDC.

I don't mean to confuse you. If you just switch to simul injection and back off your req fuel until you get the best idling and rev, then play with injection timing to get the best idling and rev- you are good to go. And switch the settings for the start up. Then tune in your VE table.

I believe this is a 60 degree vtwin motor but I will double check to make sure. I am doing a test tomorrow, so I will get to try to make these changes.

Now since I have my trigger wheel setup so that TDC is 9 teeth after the gap and the delay teeth is 9 does that mean that it injects fuel at the gap or the 9th tooth? Do I need to change the delay teeth or is that where the injection delay comes into play?

Tach is 1st tooth after missing teeth. Delay teeth electronically moves this. Since you set it at 9 teeth. Tach is now at the 9th tooth. Injection happens on tach.

I fixed previous post so you can reread it. I think i was looking at a different msq before and thinking your tdc was elsewhere. - sorry i said injection happens 90* after tdc. I meant after missing teeth. It happens at tach tooth so, Injection is happening right at tdc of number one. I apologize for the confusion.

Since you are using fuel only - you can set injection with delay teeth OR injection delay. eg:If you leave injection delay at 0. And you move delay teeth to 6. You will inject 30* BTDC #1 & 30* ATDC #2 on a 60* vtwin.

After some thought - using simul inj firing is alot like using a carb. And having an odd fire engine is compromised too because injection will alway be off for the number 2. Fuel is kinda always going to be in the intake and you can't pinpoint it to get the fuel savings you want. But you can just use the amount of fuel you need. But you may always have excess fuel during overlap. And the reason for this project was fuel savings. Simul will fix your stroke / inj event / changing tune problem but may or may not be better on fuel. Expecially since this is a low rpm engine and you have low fuel pressure and very long injection times.

If its not better on fuel, You would be better to use the cam sensor as a reference point and then you could pin point the fuel better. And raise pressure to shorten your injection times.

Well I got to test the machine today and do some adjustments that you suggested. I was able to test the machine against another machine that is identical except it is still running the carb. After I adjusted the VE table and going to simul injection the machine performed really well. It definitely outperformed the carbed machine but I still don't feel like it is running correctly. Of course, I forgot to data log the test. We were going up and down a pretty steep hill so we were able to put a heavy load on the machine. I was able to get the machine to go up the hill full force and only loose about 500 rpms where the carbed machine was loosing almost 1000. That being said, a few hours after the test I got a phone call saying that the machine was not running very well. Of course I wasn't there nor was I able to get to it. I was told that it seemed like it was running rich, it wouldn't rev up all the way. I think I still need to get the injection timing corrected. I am suppose to be able to run the machine tomorrow so hopefully I'll remember to datalog it.

This is a long-shot, but if your processor is getting resets while the engine is running or starting the MAP sensor may be giving the processor a "new" initial baro reading that isn't correct because by then the intake is already under vacuum.

My motorcycle project has suffered from this problem. I've been unable to completely cure my random processor reset problem, and I noticed that when the bike was running from initial startup it had good AFRs in the datalog, but then it will get a momentary reset (marked by a red line in the datalog timeline and the Tachcount resets to zero) the AFRs immediately go rich and the bike runs like crap. So my assumption is that when the processor resets it takes a new "startup" baro reading which is totally wrong because the engine is producing vacuum anywhere between the 90s to the 60s kpa.

I'm still working to get a reset-free situation, but I'm also thinking of installing a second MAP for full-time atmosphere sensing.

EWflyer wrote:This is a long-shot, but if your processor is getting resets while the engine is running or starting the MAP sensor may be giving the processor a "new" initial baro reading that isn't correct because by then the intake is already under vacuum.

My motorcycle project has suffered from this problem. I've been unable to completely cure my random processor reset problem, and I noticed that when the bike was running from initial startup it had good AFRs in the datalog, but then it will get a momentary reset (marked by a red line in the datalog timeline and the Tachcount resets to zero) the AFRs immediately go rich and the bike runs like crap. So my assumption is that when the processor resets it takes a new "startup" baro reading which is totally wrong because the engine is producing vacuum anywhere between the 90s to the 60s kpa.

I'm still working to get a reset-free situation, but I'm also thinking of installing a second MAP for full-time atmosphere sensing.

As long as it is running the machine runs great.

800vtwin:

I was able to get two data logs. The first one is long. It is from cranking, warm up and running. I also made a change to the ve table during this data log. The second one was after I let it sit for a short period of time to see how it started back up and revved up. However, the first data log is too large for me to upload so I uploaded the second one. Is there a way for me to downsize the first datalog? It has the most info in it because it is has from a cold start around 40* to running it under a pretty decent load.