gemellocattivo.com

I've been away on a family camping adventure.....I really don't enjoy camping but the kids had fun so...

Anyway, that means no car time recently. I did get to play with the collector NBO2 data collection a bit more and have decided its pretty unlikely I'll ever manage to wring cylinder data out of it so I'll stop that nonsense.

The ecu does has a feature that will report out the time between teeth at specific times relative to spark events....I may play with this a bit to see what I can do with it. I also need ot get the CAN stuff working. And compression and wiring tests I think.

As in cylinder compression and if so why given that all is new?

Exactly because ALL is new and untested.

There was a lot of variation on the first compression test, which is expected with brand new unseated parts. Now that its run a bit I'd expect things to be getting more consitand and that is almost what I found.

9 cylinders were in the 150-160 range. Cylinder 6 where I see oil was 175, which means the oil is helping it seal and therefore most likely coming in through the intake for some reason.

Cylinder...4(? without my notes in front of me) was 140, not awful but something to keep an eye on.

Then there is cylinder 8 at 110. Last test it was 125 so its gone backwards and I'm certain I hear a leak during the test. I need to go back through my notes, I think this one was prone to a valve leak but a gentle tap on the bucket to pop it open and let it slam fixed it so I thoght it was fine.....that may have been a bad call on my part. For the moment I'll watch it, but if it persists this and the oil in 6 mean the engine needs to come apart, the only question is now of winter.

After all that I put clean plugs in, loaded the lastest version of the model into the ECU and....couldn't get it started but did manage some really loud back fires. WTF is it now?

I need to go through the log data. I had cranked it a lot before trying to start it so the battery was low and at the very end when I was done and picking up I saw that the wire I had clipped the timing light to had logged itself in the throttle linkage preventing at least cylinder 7 throttle from closing so I don't know what that means and will need to take a closer look tonight.

The model I loaded last night remover the cylinder O2 stuff that the sensor response time was just not fast engough to support and added new code to do misfire detection using crank speed variations. The new misfire code seemed to work....there was a problem with a math function that worked on the simulator not being accepted as valid by the ecu, but a quit edit and it was solved and pretty obvious that a couple cylinders were firing and the rest weren't but until its running I can't be certain the ones I think are firing are the ones that actually are. I found some data that says velocity peaks around 150ATDC so that is about where I'm looking....but with 12 cylinder one hits the top of compression which should cause a slowing every 60 degrees and at 150ATDC, 3 or maybe 4 cylinders have fired depending on timing advance.....but it seems to be working during cranking at least with the crank moving about 30% faster if the cylinder fired (though I'm not currently positive I'm calling it the correct cylinder) . I need to see it in action on a running enine though.

I also added a misfire count.....but thinking I'm not sure how useful it is since the function its not synced to engine cycles its just running in a timed loop so during crank and low rpm the same misfire coud be counted multiple times and at high rpm it might not be counted at all. I need to rethink this part but its not important at the monent, I just look at a graph of the log file of misfire vs crank position and its obvisous whats going on, but long term I'd like to light a malfuction light on the dash so this will need to be sorted.

Tonights focus will sorting out or trying to sort out why it didn't start last night.

It fired right up tonight, so it was the battery I guess. I didn't have a chance to really dig through the logs so I still need to do that.

Tonight I shut down cylinder 1 on purpose to check the misfire detection and #1 generated a misfire indication....but it looks like #7did to.....i need to look at that.

I'm also confused....the front pipes are hot, I climbed under the car to check. The misfire stuff says they run.....

Keep up the good fight, my man. Finally caught up on all of this again. Congrats for bringing her to life!

I knew there was a good reason I wasn't really working on it....now that you're caugth up I can get back to work!

I looked through the daya files.....I need more sensors!

The misfire data is interesting. It looks like the controlled misfire shows on 4 cylinders...1 before and 2 after. That makes sense I guess, there is power for 180 degrees and it takes time to get the flywheel back up to speed. I had it set to look for 1% drop, it should probably be 1.5% or 2%, at least near 2000rpm and no/light load. That change should clear the false reading I got yesterday.

This whole thing is a bit of a catch 22 game....everything is new so every problem could be almost anything. The rear bank pipes are clearly getting hotter than the front bank, but cyl 12 almost always has the shortest time (so its firing the hardest).

I just ordered a differential pressure gage that should be here saturday to sync the TBs....start with the basics. the actuator is attached to #12 so my plan is match them all to #12...should be straight forward. This meter is much more sensitive than the MAP sensors, long term I'd like to use the MAP sensors but right now I want to be sure.

If that doesn't make things more clear I'll install O2 sensors on each cylinder. I probably won't use the NB sensors becaseu the ECU doesn't read then well.....there is a 5V pull up and high reading can mean rich or the sensor is cold becasue cylinders are dead. The front bank sensor take much longer to get to a reading I believe, so more evidence of a large temp diff front to back. I think my CAN expander can read them...but its not online yet. I'll have to figureout what I want to do here....but first sycn the TBs

I guess you don't have 12 EGT probes.

Going to hook up plug gap impedance sensing knock/misfire detection too?

And have you checked that all your injectors are pulsing?

I do not. The cylinder O2 bungs I have are in a good place for EGT, so that is an option but I'd probably do WBO2 first.


I've not seen a good add-on system for that?


Yes. I did that way back. I wrote a little program that runs the fuel pump and lets you pusle the injector whatever time in msec you like. Everything looked right. Something could have failed since buy it was all good once upon a time.

The crank speed data seems to say everythingis running at times but cylinders are coming/going, particularly on the front bank. The WBO2 sensor say the front is stupid lean, which I thought has to do with cylinders not running but the crank speed data says no, its just stupid lean. So I THINK the TB sync will make a big difference and is key to the problems I'm having.....it can't possibly be very good at the moment. I odered a digital differential pressure meter and I already have a pressure hose to each cylinder so that is set 1.

Once I know the TBs are matched I'll see where I am and decide the next step if things haven't dramatically improved....but next will likely be 12 WBO2 sensors if it remain unhappy.

This morning as I walked still 1/2 a sleep from bed to the shower I stopped as I realized a differential pressure sensor maybe wasn't my best idea becasue each cylinder pulls at a different time...DOH!!!!

I do have some little check valves I could use to get the peak from each cylinder but I'd need to unpull and connect for each reading. It might be better to just put a scope on the MAP sensors....or maybe easier yet is just buy a pair of flow meters to set on the stacks which was my first thought but there was only 1 in stock for saturday delivery at amazon when I thought of it plus long term I really want a solution that will let me monitor sync and hopefully correct out of sync in the ECU.....so either the MAP sensors of O2 sensors need to get connected to the ECU.

I'll try a scope on the MAP sensors and see what I have signal wise I guess and hopefully this is a major source of my problems.

Your MAP gizmo isn't "cylinder aware" I guess.
Can your ECU figure out which cylinder is pulling and log the MAP reading at that time?