M50 sensor problems. need some wizardly help.

[Mordan]

EDIT: FINAL FIX is Crankshaft Sensor. New Camshaft sensor didn't do much. Then 2 weeks later, I had an idle issue (yoyo + stalling). I fixed that with O2 sensor and TPS.
So I can say now that the car is fixed.

Hello,

For those who are quickly bored reading here is the youtube video I made about my symptoms.

[ame]http://www.youtube.com/watch?v=Ie63HlN210s[/ame]

I'm here because the E34 guys are knowledgeable and mature. Besides I own an E34 and it is my favorite.
But this issue in this thread is about a M50 Vanos in a E36 with Siemens MS40.1 engine management. A little unknown prick that fitted only OBD I E36 320i and E34 520i everywhere in the world but the USA. Yet the basics of the M50 are still there. Since many of E34s have M50 engine, I took the liberty to post in this sub forum for the reason outlined above.

Here is what happened in chronological order.

Bought the car 2 years ago at 40k miles. Cheap price and low mileage are the only reasons why I bought that car.

1 year later and 10k miles later during winter, when overtaking, the engine would stumble/hesitation at 3000-3500. Similar behavior to a rev limiter. The engine would sort of cut off. I couldn't get past 3500 anymore.

Back then I changed the plugs,fuel filter,air filter and cleaned the MAF.The stumble disappeared. However the car would still take 3-5 seconds to start when engine was hot.
That issue disappeared on its own after 10-15 trips and weather became better. Never knew why it was hard to start hot.

A mechanics told me the difficult start issue is probably caused by a temperature sensor (intake or coolant).

It enriches too much the mixture. At the time I forgot about it since the symptoms disappeared.

Fast forward 8 month.

November 2010
Noticeable increase in fuel consumption. 10% over what is usual. I write down statistics in a booklet.

5 December 2010
When temperature got very cold, same issue came back while I paid a guy to install LPG gaseous injection. Bah. He said it happened just once when he first started car in the parking lot before the LPG installation.

13 December 2010

Now I also have the following

Car is now difficult to start cold and hot. Takes 3-4 seconds.
The car would sometimes goes into limp mode when starting it. You could feel each time, the engine management would hesitate between limp mode or normal mode.

In the beginning, it wouldn't go into limp mode when cold. But eventually it did.
Around 3000 rpm, the rev needle is not so smooth anymore, like it hesitates.
The car often idles roughly when hot. Then the car shakes.
The stumble/rev limiter sometimes comes in around 3500 rpm.

It is clearly not a clear cut fault.


17 December:
Put in a new air filter. Engine run better for awhile before issues came back.

24 December:
I went to the dealer and they didn't see any code. OBD I sucks if your car is not made for the USA.

5 January 2010
Replaced Valve Cover Gasket (VCG) was changed because a little oil was leaking in the 2nd cylinder plug well.
Replace 6 new coils,boots and NGK spark plugs because it seemed some cylinders were not firing when hot engine is idling roughly.
I previously did the test by removing coil connectors while engine is idling roughly shaking the car. Removing on coil #2 and #5 didn't change the idle.

Morevoer the 2nd cylinder coil had this bright yellow oil around where the boot and coil mate. That yellow color is too bright to be engine oil. But I'm not 100% positive. Changed the coil for the peace of mind since when they really go bad I hear they can wreak havoc in your ECU.


7 January 2010
The weather was freezing but no moisture at all, the car started cold right away. Smooth idle.
I thought wow! It didn't do that for some time! Let's take a test trip. Everything went fine at first.
When the engine got hot, I got some serious stumble/hesitation around 2500 rpm. After that, the engine got into limp mode while running in a few seconds. I lost all power.

The engine even died by itself. I restarted and limped back home.

After doing the coils, I did not reset the ECU. The ECU learning must have been bad.

Therefore I unplugged the battery and took the opportuinity to remove and check the DME.
It is very clean. No moisture symptoms. I remove metal plates using a Tx 10 socket but could not see the coil drivers. Tried to remove the top PCB but afraid to force it. The Siemens DMEs have different clips than the Bosch ones.

I wished I had a breaker box! That would be the best tool to debug the electronics in that ODBI car.

19 January:
So now after a reset, new coils, new boots, new plugs, and a new cheap MAF, I still have a difficult cold/hot start. Needs 4-5 cranks before engine starts like its going to choke and then stabilize. I still have the hesitations.

Here are all observation I made today

-if I remove the break booster hose, engine dies right away.

-when I remove the oil cap, it has no effect whatsoever. No air moving in or out when I put my hand over the ole. Is it normal for a M50? On my M20/M40, engine hates it when i remove the cap.

-when i plug my Mytivac vacuum pump, i see a healthy 20 psi

-when i accelerate in idle, vacuum needle first drop (normal)

-During deceleration, vacuum needle is dropping slowly from 30 psi. So the exhaust is not blocked.

-from the DME connector, the following sensors ohmed nearly the same as NEW OEM dealer sensors I bought to be used as yardsticks. Bentley does not cover values for Siemens MS40.1
-- IAT - 9k at 7 degrees Celsius. Same as new dealer sensor
-- CTS - 4.5k at 7 degrees Celsius. Easy to reach, it ohmed 200-300 ohm when hot so this one is good at all time! Same as new dealer sensor.
-- hall effect CPS and Cam sensor at 15 ohms primary 2 ohm secondary
But is the ohm test reliable for hall effect sensors? Those are the readings I get, even from the new OEM sensor.


-when hot and i accelerate in idle, engine hesitates or wants to die (video). It does not do that when cold. That's definitively not good.
-still have engine hesitations starting at 3500 RPM
-I was running a tad rich. see the pictures in my post about my plugs. http://forums.bimmerforums.com/forum/showthread.php?t=1589106
Don't know yet if the new coil improved that.

From my research it could be

O2 sensor : running rich but not extreme. Exhaust tip is not sooty. The o2 does not impact the starting process.
RPM sensor : I read that RPM needle symptoms are always because of a bad CTS,Wiring or DME
Knock sensors : retarding ignition. They say they rarely fail. Ugh!

Could it be a relay issue?

For the needle hesitation, it should not be related to the fuel system because it does the same when car is running on LPG. However the car is always starting on gasoline. Still the LPG thing is a nasty variable that adds incertitude to the diagnostic.

The ECU slowly learns a bad mapping from all its sensors. Because my car runs better after a reset. Limp mode appearance disappears for awhile.

For a 55k miles car, it breaks the myth that old low mileage cars are a safe bet.

After reading MBman post, I bought a second hand DME that should arrive soon. At worst I will be able to make a breaker box out of it. Could the coil drivers fail only when engine is hot? Or is it On/Off?
You can never be sure about electronic gremlins.

I know I will find the goblin but I'm writing this post for the record and maybe someone will be able to pin point the problem right away from the symptoms described.

Here again is the youtube vid with all the symptoms together.

[ame]http://www.youtube.com/watch?v=Ie63HlN210s[/ame]

Thx for your time and reading.

[Jay547]

Can be Cam. Position Sensor or relay problem, PM me if you need any assistance with parts.

[Mordan]

Can be Cam. Position Sensor or relay problem, PM me if you need any assistance with parts.

Which relay would it be?

How would you describe the RPM needle behavior between 3k and 5k in the vid?

needle jolts up? needle jerks up?

I will try the crankshaft sensor first.

thx for the answer.

[Grantnz]

Because it started fine on Jan 7th when cold and then had problems when hot, I would consider O2 sensor. Can you try disconnection and see what happens? poor connection maybe?



I also would have wondered about a blocked cat converter, but the vacuum guage rules that out?

I wonder if the tacho irregularity is a cluster issue.

Try running with the cam sensor disconnected and note the symptoms.

I am surprised that as you are in Europe, there isnt a diagnostic service avaiable.


Good luck!

[bolloc]

Just a quick note, around 36 seconds when you say note the tach needle, the engine actually hesitates as well.

I would check the temp sensor and o2 sensor, if the dme doesn't sort it out.

(Too bad the LPG was installed before the problem was resolved)

[Mordan]

thx for bolloc and grantz for your messages.

I got my second hand DME. Put it in the car. No changes. Hard cold start and still same RPM hesitation after 3k, during the idle and idle take off.

I then decided to do the CPS. I'm doing one thing at a time to pin point the issue and learn.

Well the CPS replacement went awry. I couldn't remove the sensor. The wire is easy but the CPS is stuck. . Allen screwed is removed. Sprayed lots of WD40 but without success. Because of the Vanos, the cylinder head shape makes it impossible to use a drill! Obviously you cannot drill from the other side because of the toothed wheel side.

Has anyone succeed out of this awkward situation?

@grantz
I have been to the dealer and they saw nothing on the computer. No codes. I blame this on the OBD I MS40.1. I have heard that hall effect sensors may just "miss" but not be completely dead.

I did the cluster self test and all the needles work fine and smooth.

My take is that a blocked cat is very unlikely on a 55k miles car with a vacuum gauge showing no signs of a blocked exhaust.

@bolloc
well indeed. It buggers me as I cannot be sure the LPG thing is messing up my engine. Got to have faith in the guy to installed it.
I thought the issue was solved back then and it is only after I pre-paid the LPG thing that I noticed an increase in LPG in November.

LPG is twice cheaper than gasoline here in Europe so after 40k miles, the 3000 euros are paid back.

after the CPS, if I succeed removing it, I will do as you said: #1 disconnect Cam sensor and note symptoms. #2 disconnect O2 and note symptoms.

Just a quick note, around 36 seconds when you say note the tach needle, the engine actually hesitates as well.

you mean at idle acceleration take off? yea that's one of my symptoms. It is usually much worse, just as I depress the acceleration pedal, the engine sometimes wants to die before tach needle takes off. It is strange that it does not happen when clutch is engaged. I can go up to 3k RPM uphill without a problem.

You can feel the same kind of hesitation during a hot start, where around 500rpm on its way to 1000rpm, engine hesitates.

Here is a vid of just a hot start I made today.


[ame]http://www.youtube.com/watch?v=uWiuuzc0qJ4[/ame]

[shu328]

sub'd, this is good learning material

[bolloc]

My concern with the LPG is its just something else in the way during the diagnosis.

Have you ever done a compression test on this motor?

Good thought on the CPS, can't help you with removal tips though I have the M30....

[rotaryboots]

What kind of fuel pressure are you seeing?

[Mordan]

will test fuel pressure after CPS is done.

bad weather prevents me from working full time.

moreover fan clutch is stuck so working there is though

so my plan of action :

since my water pump has some play

remove radiator
remove vibration damper
unscrew water pump pulley
unscrew water pump and remove it together with the puller and the fan

then i will try a something with a strap or a vice. or just buy a new fan clutch.

replace water pump

remove stuck CPS! Since the damper will be out, I will be able to drill the CPS out, hammer it out. whatever

stuck stuck and stuck

cheers

[Mordan]

got it out.

removed radiator
removed fan
removed water pump

even then it was really annoying. A big vice grip was needed and lots of WD40

The new one doesn't want to go in. I'm not using force on the new sensor so tomorrow I will dremel the hole with some abrasive tool.

[WhoNew]

I had a very similar issue with a 318i. I know it's not an M50, but the fix was ... so easy it was stupid.

I had everything you are talking about. The "rev limit" was about 4500 RPM. I drove the car for about a year like this, as it was drivable and ran fine.

Trying to diagnose another issue, I pulled the cam sensor and crank sensor to replace the O-rings. Come to find out that the cam sensor had the wrong O-ring in it, and made the sensor sit up about an 1/8 of an inch. It could read the signal at low rmp, but as the rpm increased, that 1/8 inch made a huge difference. Replaced it with the proper O-ring, and it ran like a new car.

I hope your solution is that simple, and something to check or think about when diagnosing. Come to find out that BMW lists the Cam and Crank sensors as "pulse generators" and sells the O-rings for them listed as Pulse generator O-ring. Even though they are 2 VERY different O-rings.

[Mordan]

==

[Mordan]

Final Feedback for this thread

New CPS removed starting issue altogether.

New Cam sensor smoothed everything out.

Sensors were repleaced one at a time.

Lessons Learned:
Hall Effect sensors are CRAP. (car is only 55k miles)
Hall Effect sensors resistance tests mean NOTHING.
Hall Effect sensors are not completely dead.
Hall Effect sensors are disposable items.
Hall Effect sensors are not "lifetime" like sometimes advertised by BMW.
Never trust anything "lifetime" advertised by BMW.
Hall Effect sensors are more precise, give a better ignition, and are more expensive.

If you car has Hall Effect sensors, any running issue could be one of them. (rough idle, rpm hesitations, rought cold or hot starting (random))

I did some tests. Car would not start with bad CPS sensor connected. But it would start with CPS disconnected entirely, altough the cam sensor (new one) would take a while for starting the car.

PS: Don't trust the bloody computer. It is piece of useless trash.

[uk525td]

i see mentions of cam pos sensors in this thread, i thought m50's only had crank position sensors? (one in same???)
glad you sorted your problem ,i never thought a seemingly good crank pos could have so much bearing on running, its good to know for future

[Mordan]

Hello

@ the australian guy who contacted me by email. (reply got bounced so i put it here)

About the camshaft sensor reference:

I went to the dealer and they couldn't tell me either for sure. I had to
remove it or i managed to look up the BMW reference while it was attached
I don't remember.

So basically I would advice you to get the right reference by looking at
yours. Sorry I don't have it right here. The part invoice is 200km away
from where I'm right now

My DME is MS40.1 . It sucks because spare sensors are very expensive.

Bosch DMEs are always better in this regard.

[inm8]

Hello

@ the australian guy who contacted me by email. (reply got bounced so i put it here)

Thank you for getting back to me - I logged on today wondering why I hadn't heard from you and then realised that my email address was wrong. Gah!

Looks like I'll be pulling the crankshaft sensor out and going to the bmw shop. This silly MS40.1 is making things expensive...

When I get a part number I'll post back here for future reference.