Odd cut out

[Ramblin MAn]

I have an m42 from a 91 318is that has been transplanted into 1982 e21 320i that has been used as a track/street car for a couple years.

The previous owner installed a custom intake manifold on the car right before I bought it and didn't have time to sort things out.

The engine has a wierd stall and recover problem that I thought was  the air flow meter. I used the bently manual to check the resistance across pins 1 and 2 and as others have stated, resistance was stuck at 11k ohms throughout the range of travel. I took a chance on an ebay air flow meter and it reads 500 ohms and steadily rises til about halfway through its travel and then starts to fall again to nothing.

Interestingly, I disassembled the one that came with the car and it behaves exactly the same when I test without putting the little printed circuit board that is contained inside in the circuit.

The ebay AFM is an earlier unit than the one that came on the car.

Anyway, back to the problem. Car starts and idles nicely, however if I slowly open the throttle, when I get to about 2800 rpm the engine stalls and then at about 900 rpm it recovers and revs back up and then at 2800 it stalls again. It keeps repeating this process as long as I hold the throttle steady.

I can rev past it. When I watched the original air flow meter in operation the stalling would occur right where the resistance started to fall off.

I checked the TPS and it tested fine, no jumping or flat spots. Both air flow meters run exactly the same.


Trying to find out if anyone has heard of anything like this before. Also does anyone know what the resistance values on the air flow meter should be? Should the values rise contantly to the very end of vane travel?

If anyone is interested, here is a couple pictures of the intake.

http://pg.photos.yahoo.com/ph/ramblin_man_us/detail?.dir=5374&.dnm=5029re2.jpg

http://pg.photos.yahoo.com/ph/ramblin_man_us/detail?.dir=5374&.dnm=9e8cre2.jpg

I know this isn't an e30, but the m42 is from an e30 so Thats why I didn't post in the non e30 section.

[bmwman91]

Interesting setup on that engine.

Couple things...has it ever run right?  When you go & do something like that you MUST get custom software to account for the changes in induction dynamics.  The ignition timing maps will be way off for this motor, for one.

That AFM LOOKS like an M42 unit, maybe from a later model?  If it is not from an M42, it will not run right.  They all put out similarly shaped curves, but were scaled differently.  You'll likely run lean if it was a unit from a 6-cylinder car.

I'd almost guarantee that the problems are from running un-modified fuel/ignition maps in the ECU and from using the wrong AFM.  Now maybe the thing is custom-managed and needs some re-tuning, who knows.  Do you get a check engine light when running?

Interesting rig anyway though, get more pics!  I can host them for you if you want.

[Ramblin MAn]

I don't know if it ever ran right with this intake. I bought the car sight unseen from an advertiesement on the bmw320i.com forum and it had the stock intake on it in the pictures. The collection of parts was worth way more than what I paid for the whole car. The car was in Denver and I have a brother there. He went and drove it and said it was fantastic. However, I've found a list of issues. I haven't even driven it yet further than driveway to trailer to driveway while I slowly try to sort things out.

I was thinking about going to Megasquirt for fuel managemet. I'd like to get rid of the AFM all together and be able to tune the fuel curve.

I've also thought about simply going back to a stock intake manifold and throttle body. I'm not sure what might not be working with out the two stage throttle in the stock set up.

I don't have a check engine light to come on. That is one thing I'm trying to figure out how to add. It would be nice to get codes, if any from the ecu.

[bmwman91]

Well, get yourself a Chilton's book.  It has all the wiring diagrams and pinouts from the Motronic.  You can find yourself the output for the CEL and hook a bulb up if you just want to check.  There was a built-in diagnostic system where you could get the ECU to dump out the faults stored in it, but that requires a working throttle potentiometer...which brings me to my next bit.

Granted the pics are small, but I did not see anything resembling the original TPS.  The M42's ECU (Motronic M175) supposedly is self-calibrating for the TPS to account for variances that might be experienced if it is replaced.  Unfortunately, any old pot won't work as there is still an expected range of resistance expected, and most others are pretty different.

My recommendation...get yourself a stock manifold (looks like you'd only need the upper unit...appears that the lower one is on there still), TB, vacuum lines, a couple little coolant lines for the de-icer on the TB (although they are probably plugged already since the TB is not there) and intake boot.  An OE air box is a good idea, too.  Cone filters on these motors do little to nothing (esp w/o a heat shield).

Once you get it 'restored' and running flawlesly, then tinker with other options.  I did a MegaSquirt on my first car.  It was a real PITA, but so would any stand-alone system.  Anyway, keep us posted.  Get more pics man!  I'd like to see that intake setup a little more closely!

[Ramblin MAn]

Here's what I just measured.

18 volts going to hot side of the injectors.

6.75 volts measured at the 5 volt supply at AFM and TPS.

TPS is 4k when closed and 1k when open. Is that not backwards? Remember this is on the custom intake and the PO may have put the TPS on the wrong side of the throttle.

When I disconnect the TPS the engine does not stumble and revs free and easy, at least sitting in the driveway.

Anyone?

[Ramblin MAn]

forget the voltages above. My meter's battery battery was going dead. I got my good meter out and compared readings and my cheapy meter reads 12 volts as 18 volts while my good meter reads correctly. I've been busy relining my box gutters (100 year old house) so fun projects have had to wait while I'm 35 feet up in the air. I'll remeasure everything when I get a chance.

In the mean time, I have a question about the factory throttle body. There is what appears to be a small primary and larger secondary throttle plates. How does the secondary throttle plate actuate? Is it mechanical and connected to the primary through some linkage or is it opened via manifold vacuum or some type of servo controlled by the ECU?

I'm trying to track down why the the weird cut out goes away when I disconnect the TPS. I'm assuming the ECU is going to a limp home map that is richer then the one it runs on when the TPS is connected but I'm making wild guesses.

So nobody has a flow chart or anything that shows what the ecu is doing with the sensor data? We all know that the computer has fuel maps and ignition maps and uses the sensor data to adjust the maps. Doesn't anyone have the speciffics? My Yamaha FJR service manual has an in depth section on the fuel injection and how all of the sensors affect the fuel map. Someone has to have this information. After all it was programmed into every DME 1.7 ECU.

I've thought about just getting a Techlusion fuel nanny to adjust the fuel, but I think there is more going on than just needing a re-map.

I have an osilliscope on the way, and I guess I'll have to invest in a wideband O2 sensor or at least one of the cheaper EGA's to figure this out.

Those of you that have used Mega Squirt, what functions did you let MS handle and what did the stock ECU keep? I don't see where MS will handle multiple coils yet. I don't want to go to MS until it will do it all.

[romkasponka]

TPS on pin 2&3 should be 800-1200Ohm closed and 3200-4800 Ohm fully open.

[Ramblin MAn]

It appears that when the stalling occurs, the fuel injectors are not opening. According to my fuel pressure gague, the pressure goes up when the stall occurs. Spark continues so it is only the injectors that stop firing.

Again, if I unplug the TPS this does not occur. The engine will rev just fine.

Anyone care to speculate what could be causing this?

One more thing, if I unplug the O2 sensor the car dies after about 5 seconds.

[Ramblin MAn]

I replaced the O2 sensor with a new one just to be safe. I'm using the one for a Mustang and pulled the pigtail off of one at the junk yard so I can just plug and play.

It had no effect. The reason I got a new sensor is that I was only measuring 0.16 volts max at the O2 sensor. I even dumped raw fuel into the intake through a vacuum port to the point it wanted to die and still didn't get a reasonable reading.

Of course I do everything half ass at first so I didn't check to see if I was getting voltage to the O2 heater. Since the new O2 sensor didn't fix it I have to go back and do it right this time.

I also put in a different ecu, it ran the same so I installed the chip that came with it which makes it run really rich/bad and it still cuts out.

(remember this is a transplant into an e21 and I don't know what might be wrong)

In the mean time, I just bought a used complete intake with everything still attached and the crank and cam sensors thrown in plus a wiring harness with relays included.

I will track this down even if I have to put a stock intake back on.

I'm gravitating towards the crank sensor (guessing) but it would be nice if I knew how the different signals interacted with each other. If I unplug the TPS, no stalling. I hear it stumble and recover as it compensates for no TPS input and goes into what I assume is limp home mode and it seems to run great.

With the tps plugged in and measuring the O2 sensor output (pre replacement) when the engine would stumble, the O2 sensor would go to a steady .1 volt. In the rpm range below the stumble voltage would fluctuate and above the rpm range it would fluctuate. Nowhere near what it should, but it was obvious that the ECU or some component was doing this "on purpose" meaning it was looking at the various signals, seeing some problem and shutting off the injectors and setting the O2 sensor value, if thats even possible.

This is at what appears to be the magic 2800 rpm that gets mentioned so frequently.

I'll figure it out eventually. I was supposed to be getting an o-scope but that may never materialize. So it's replace one thing at a time and see.

[Ramblin MAn]

if anybody cares.....

I think I've been missing the boat here. I was just looking at the difference between the stock throttle body (now that I have one) and the aftermarket one on the custom manifold.

My crude measurments showed that the primary and secondary throttle plates on the stock throttle body measured 32mm and 54mm respectively while the single throttle plate on the aftermarket throttle body measures 60mm.

At the point where the stock primary and secodary start to open together but the secondary is still closed, I "calculated" that at the same point in travel, the afterarket throttle was flowing three times as much air, (well, three times as much open area) but the combined area of the two butterfiles is bigger than the single larger butterfly.

SOOoooooO, I'm taking the stock throttle and the custom intake over to my brother in law, the machinest, and seeing if he can whip me up a an adapter plate. The only other thing I can think of would to be a progressive TPS switch that changes quicker sooner. I think it will be quicker to just machine something.

If that does not work I'll bolt up the stock intake.

[Ramblin MAn]

So I think this has been my problem all along. I'll know on Thursday.


http://new.photos.yahoo.com/ramblin_man_us/photo/294928803805066294/2



The new to me intake all painted up and ready to assemble. That new paint is sure to add at lease 5 HP.

http://new.photos.yahoo.com/ramblin_man_us/photo/294928803805102812/3

[Ramblin MAn]

Well that was it. Car fires right up and revs freely without issues.

http://new.photos.yahoo.com/ramblin_man_us/photo/294928803849920417/2

http://new.photos.yahoo.com/ramblin_man_us/photo/294928803849920432/1

http://new.photos.yahoo.com/ramblin_man_us/photo/294928803849865676/0