RPM unstable but no sync loss

[Kyle]

I have been working through so sync loss issues, which I finally seemed to have figured out, but I have an oscillation in my RPM signal that I don't really know what to do with. RPM indication fluctuates by 100 to 200 rpm. My log file is to big to attach and I can't figure out how to clip it but here are a couple pictures. I will make a shorter log tomorrow evening.

Has anybody seen something similar? I haven't seen a sync loss. Should I be looking for electrical gremlins, checking grounds and disconnecting potential feedback sources? Should I be adjusting the VR circuit pots to try and clear this up? Currently they are set with R52 all the way out and R56 all the way out then two turns in (per instruction http://msextra.com/doc/ms3/ignition.html#vrv357pull). I'm using a 36-1 toothed wheel with a hall sensor. Trying to get this signal solid so I can rip the carb off an install the throttle body injection. Right now the megasquirt is just data logging.

I had a similar problem before, when I was trying to get coil negative triggering to work, but I eventually gave up on that.

Thanks for any suggestions

LogView.png

LogView2.png

[Matt Cramer]

If you could post a tooth log, that would show how stable the RPM actually is.

[Kyle]

Are you suggesting that the engine is actually running that rough? It is an odd fire V6 and really needs some carb work. (I'm determined that the next time I take the carb off it is never going back on) That being said, it sounds smooth when held at a constant throttle position. I was expecting that you would see the engine rpm change in the map reading if it was fluctuating that badly. But I really have no idea what to expect.

I took a tooth log with the idle screw turned out to about 1800 rpm. I also took two logs, one running and one cranking (distributor disconnected at the coil). Don't know if the cranking log means anything, but it looks like I would expect it to. The tooth log does show the RPM changing. I just can't bring myself to believe the engine is actually running like that. I guess its almost 50 years old. If its running like that is it going to be a problem for the injection?

I've been considering trying to power/ground the MS from a separate battery to see if I can remove some electrical noise from the running engine. Not even sure that makes sense, just swinging blindly at this point.

Or maybe I should just rip off the carb and see if it runs with the TBI.

[Matt Cramer]

IIRC, there have been a couple changes to the oddfire wheel decoding since MS3 1.0 firmware - have you tried the latest code? The RPM itself looks pretty stable.

[Kyle]

Updated to the latest firmware. I didn't realize there were so many updates. I'm a little sad now that the project car has been a project so long.

Anyway, I took a couple logs but didn't notice a significant difference between the new files and the old. Still no sync loss though. I guess that's a bonus.

Matt, I just bought a fuel pump from you. When it shows up I'm going for broke. Hopefully Its starts with the TBI when I'm done. Thanks for your input.

...If it doesn't, there's always that 3800 Buick engine I've been fantasizing about. I hear they run pretty well on an ECU.

But until then, any thoughts or input would still be appreciated.

[Kyle]

Just posting an update in case any one comes across this in the future.


The engine runs fine on throttle body injection with the seeming unstable rpm, at least as good as it did with the carb and I still have a lot of work to get to a decent tune. I'm reasonably sure that it is just an effect from having an old oddfire engine.

Also, previously I posted that the MAP reading did not seem unstable consistent with the RPM. Turns out I had it hooked up to an incorrect port. With it hooked up correctly, you can see the rpm variations in the mainifold pressure. I guess the only importance in this is that it makes me more confident it is not a problem with my wiring or sensors.

[Kyle]

As you can see in the original post, when I started on this project I was concerned about the way RPM was oscillating while the jeep was running. It was never a problem and I have run fuel only on TBI for years successfully with the original distributor. Logs show that RPM was varying by as much as 80 rpm over tenths of a second. Whatever, its an old odd fire v6 engine that runs rough.

In the last couple weeks I successfully converted to LS truck coils and it's now running coil near plug sequential ignition. I'm loving it. But take a look at the logs, I attached a before and after. The new logs have no oscillation.

Its pretty cool, but my mind is blown. Is this just a software thing? But I didn't update the firmware. I have trouble thinking that the engine is actually running that much smoother, maybe it is. Why would it though? Distributor lash or something?(is that even a thing?)

Not looking for anything really, just though some of you might be as entertained by this as I am.

[prof315]

it's running smoother because you have way more spark and are getting a better burn. Spark timing will also be far more accurate which helps.