Under what conditions will the DME shut off the injectors?

[Ramblin MAn]

I'm trying to sort out an idle problem that I'm documenting in another tread, but one of the things I am seeing is once the car warms, the injectors will be shut off for about three seconds which drops the idle then they are turned back on, the idle raises then the cycle repeats.

I originally had tiltled this thread the DME and how it works, but I didn't want anyone thinking I knew.

Certainly one of you has been able to find out some info about this kind of thing.

It seems obvious that the DME is going into an "alarm" contition and turning off the injectors. If I just knew what could cause the alarm then I would know where to concentrate first.
SInce it doesn't happen until the car has hit the same spot on the temp guage, i'm thinking it's related, but that may just where it switches over to the input that is causing the problem.

Does anyone know the program the DME runs?

[bearsbmw]

this is your ICV sticking and most likely has nothing to do with your injectors

[Ramblin MAn]

I didn't say it was a problem with the injectors, I said I can hear the injectors being shut off when the idle drops. Cleaning the ICV/IAC is on the agenda today along with several other items. I'm covering that in a seperate thread, but if it is the IAC, disconnecting the hose from the MAF boot and and plugging them should quickly show if thats the culprit. I thought I did that already, but I will try it again. I know I unplugged the electrical connection.


But that's not the topic of this post. What I want to know is the program for the DME.(Ha, all I want to know) The flow chart if you will. Under what conditions does it shut off the injectors.

If the only way it can tell a lean condition exists is from the O2 sensor then unplugging it should keep it from making that "choice" and not turn off the injectors. But that doesn't work.

It shouldn't be that difficult. There are only 7 "sensors" : crank p, cam p, temp, rpm, AFM, TPS and O2. I have to take cam p and crank p out of this equasion although crank P is, I assume, providing  RPM. Most out of range problems would/should cause the computer to go into limp home mode, I would think. Only a few conditions would cause the injectors to cut out.

IF AFM + TPS x temp / RPM <>X  where X is some range THEN turn off injectors. Something like that.

The computer has a map that says how much fuel to use ay any given RPM. It will adjust that based on the O2 feedback. It would also take the change in readings into account so it knew if you were mashing on the gas, to dump more fuel like anaccelerator pump, or shut off the injectors while you were coasting.

Is there a provision for vehicle speed? I didn't think about that. Maybe I need to drive it. I remember now my 91 mustang had to be driven to at least 35 miles an hour before the computer would reset after a disconnect.

[bearsbmw]

I'm highly familure with the maping as I tune the m42 all the time...you must have one hell of a set of ears to be able to hear the injectors turning off ...There is nothing in the programing that will shut off injectors but there is thing's that will make the car not start at all witch are some you mentioned above... I would look at your ICV, Failing injector ? ,fuel pump relay? and fuel pump

[Ramblin MAn]

Actually I do have some pretty good hearing, or loud injectors or both. I have never had any problem heaing the injectors clicking on any of my euro cars. On my SAAB its the loudest noise there is.

The thing is, had I not had this exact same problem on my other m42 powered car when it had the custom intake and 66mm throttle body I would say I have a fault. But the other car did the same thing only at 2800 rpm. The big single plate was letting too much air in as apposed to what the TPS reading mapped out to and the injectors were being shut off. At 2800 rpm I couldn't hear that, I had to hook up a dwell meter to see that happening. The only fix I had available to me was putting the stock intake and TB back on which cleared it right up. I could have gone with megasquirt but I didn't have the cash or time.

But, that aside, I would think that you would agree that the maps are just tables that the program running on the computer uses to set injector duration and spark advance, and, there are a whole slew of things happening behind the scenes that if we knew the "program" would help us pinpoint perfromance problems.

I have the PDF for bosch Motronic 1.5 and it gives a general explaination of what it does, but I can't find one for these.

Ya I can and will check and swap every component until the problem goes away, but I would rather know that when "this" is happening, the comuputer is seeing "that" so here is where you should look first.

[Ramblin MAn]

Bearsbmw,  you will be happy to know that you were dead on and it was the ICV. 2 years of sitting had created so much oxidation that the vane was frozen. A couple shots of Freeze Off and working it back and forth a little has it working fine and the car idles rock solid at 850.

Now back to the topic.

So the idle is high, the computer has told the ICV to close but the rpms have not changed. The computer sees that the throttle is closed and that the AFM is flowing too much air because the flap is open too far and that the rpm that it should be seeing at that TPS value is too high. So it shuts off the injectors for a few revolutions allowing the RPM and AFM readings to drop and match up with the TPS value in some three dimentional table. Once it sees everything back in range it then turns the injectors back on.

With the sticking ICV the idle climbs back up, the fault condition re-occurs and the computer shuts off the injectors again. Rinse and repeat.

I guess if no-one is interested in this I'll shut up about it. It's just I have been bitten by the same bug twice and I would like to know more about it than this time I just cleaned the ICV and now it's fixed.

[bearsbmw]

Glad you figured it out...I still don't know of any table that shuts off injectors under any conditions... Must be a map I have never seen in the hex ... good luck!!

[roundel318]

I am reading, so is your TPS "high" ?

[Ramblin MAn]

Not maps. Maps are just tables that the program the cumputer runs uses to make it's calculations. As I understand it, the computer runs on a cycle, a "read" and a "write" cycle. On the read cycle the computer looks at the rpm, AFM, TPS, coolant temp and O2 sensor values for those equipped and then based on the change (delta) in those values from the last read and the values in the maps, "writes" the injector pulse and spark advance and then repeats the cycle.

Thats a simplistic view of it. You could call it "If, then", whatever. But the computer does a lot of other things like shut off the ac compressor at WOT etc...



And no my TPS was not high, what I was saying is that the computer looks at the values it measures and compares those values to the ones it has stored in the maps then it "does" something. In my case it saw the throttle was closed via the TPS reading, but the values it was seeing for AFM and RPM were too high when compared to the mapped values. Since the throttle was closed all it could do was shut off the injectors until the AFM and RPM values fell back into acceptible range and then turned them back on. and then repeated the cycle.

Thats just my guess based upon observation.

[quinn11m20]

I have read this thread and I am impressed with your knowledge of the DME system. I am not an electical engineer but a serious computer enthusiast. That being said, I follow the "in conjunction with" topic about the DME, AFM, TPS. They all talk to each other and the DME sends out the info accordingly.
But on the "shutting down" issue, my opinion is that they are not really shut down and or off per-say. I think they go into a standby mode waiting for the "pulse info" from the DME. To my knowledge the have 2 leads: 1 is a ground and 2 is a positive.
They pulse to say at the least, so......thats how i kinda think of it.

As the shutting off theory. I would say not until the ignition was turned off. Hopefully I helped with my thoughts. Tony. OUT!

[quinn11m20]

Awesome post as well.

[fiftytakedowns]

I think this whole topic may be relevant to what is going on in my troubleshooting thread here:

http://www.m42club.com/forums/showthread.php?p=101624#post101624

[MLM]

I think I have experienced the same thing.

I installed ITB's and had the same effect ititially until i turned the idle position of the throttle plates until they were all but closed.

A larger throttle plate would have the same effect. Bassically the ecu sees the throttle is closed but the rpm is to high. At this pint it cuts the fuel.

This is because of the overrun fuel cut mode where the ecu tries to save fuel when coasting up to lights or down hill etc.

Your trying to opperate the engine in a zone where it thinks your cruising (high rpm) without load (closed throttle) and cuts fuel until its in a region where it needs to add fuel again to keep the engine running (below a limit rpm). At which point because your bypassing to much air it accelerates again.


If you can adjust the throttle plate so it is closed at idle (or very close to it) allowing the icv to do its job you should get arround the issue.

[fiftytakedowns]

My TPS was showing 1.8 through 2.5 ohms at the connectors and. 1.8 ohms through 4.8 ohms at the sensor itself . (bentley instructed to check with the sensor harness connector off)
My Spare TPS showd 1.6-4.6 ohms . I tried adjusting the smaller throttle plate as closed as I could and It still looks like It is out of hte parameter

[fiftytakedowns]

Did you figure out what would cause the injectors to shut off (in my case during cold starts?)