fuel pump/non-start issues - been doing lots of research, still NEED HELP... please >

[B318M42W]

so if i get your situation right... you DO have spark and you DO have gas. based on the fire triangle, air is the only thing left... I,m not a mechanic or anything, so i might be waaayyy off...

here's what i'm thinking: either some of your injectors are stuck open (too rich to burn), or maybe your engine isn't breathing right. My 318 had a little idle issue on start-up, turned out that the MAF was starting to go bad. (and the DME didn't give me any codes...)  maybe take a look at your MAF, just to rule it out? and also check your air filter?   easy stuff, shouldn't be too long.

good luck

[colin86325]

Cobra Jet--thanks for the detailed analysis.  It's good that members take the trouble to follow up and even do a sensor dissection!  
In the dissection, the outer layer serves as a shield for the internal signal wires.  It's typically grounded at the connector end, and left unattached (floating) at the sensor end.
I believe the thin intertwined wires are used to control inductance in the two core wires, especially the way you described them as being woven. Inductance can distort a signal that varies in frequency (rpm).  The same principles are used in high-end audio connectors.

[Cobra Jet]

so if i get your situation right... you DO have spark and you DO have gas. based on the fire triangle, air is the only thing left... I,m not a mechanic or anything, so i might be waaayyy off...

here's what i'm thinking: either some of your injectors are stuck open (too rich to burn), or maybe your engine isn't breathing right. My 318 had a little idle issue on start-up, turned out that the MAF was starting to go bad. (and the DME didn't give me any codes...)  maybe take a look at your MAF, just to rule it out? and also check your air filter?   easy stuff, shouldn't be too long.

good luck

Yes, he's good w/ fuel, spark & operatinal DME, so the fuel pump, DME, relays for fuel pump & DME & CPS can be ruled out at this point.

As you mentioned, the engine getting or needing air is important.  The ICV and the AFM control the air intake for the engine.  I would thoroughly clean out the ICV, because just by looking into open ends of it, they normally won't look "dirty" from that vantage point.  The problematic area is within the actual cylinder, where you can't see anything.  The internal parts of the cylinder contain a spring & plunger mechanism, which when powered up opens and closes the ICV valve and this regulates the air going into the intake/engine.  If the ICV cannot regulate the air, you will experience the idle flucations, high idle, low idle, etc.  In the event that the internals of the ICV are stuck, the car will be hard to start and/or you won't be able to start the car at all.  As soon as the key is turned to the "on" position, the ICV powers up and will always have a slight "buzz" to it so you'll know at least it is getting power.

When my ICV was acting up, I could not start the car and when touching the ICV or listening, the ICV was not operational at all - no buzzing.  When I tested the resistance on it (ohms), it was out of spec too.  So, I removed it, soaked the internals w/ carb cleaner (brake cleaner will work too) and after it dried out, I also shot WD40 inside to lubricate the plunger/spring mechanism.  I then tapped around the entire outer casing w/ a hammer hard enough to dislodge the internals but not so hard to damage the casing.  After performing those steps, I then retested the resistance and it was very close to spec, so I reinstalled the ICV.  The car started right up after the cleaning and drove fine.

Now, the AFM also controls air intake and if the AFM is dirty, it too will either be completely non-functional OR partially functional.  I did not have to clean mine, as when I removed it and looked everything over, it was really clean to begin with as M42 Member bmwpower had cleaned it before I bought the car from him.  There are some threads on this site that discuss cleaning the AFM, I'd search for those threads if you do want to clean it.

As for the camshaft sensor - the camshaft sensor is relays info back to the DME for sequential fuel injection and knock control.  

Here's a good thread for you!  Another M42 Member started this thread about finding a faulty camshaft sensor.  In it, he also states his CMP sensor read "0" when the resistance spec should be 1250 ohms (do read the thread, because it does have valuable info about that sensor):

http://www.m42club.com/forums/showthread.php?t=5376&highlight=camshaft+sensor

I think you're almost there to getting your M42 back on the road!

[318i91]

Ok. The ICV reads about 8 ohms on multimeter, in the correct range. The valve did activate when charged with 12Volt, but there is some black soot in there I will clean out.

The air filter is fine. The temperature sensor in the AFM is clean and the duct is clean, too. There are 5 labeled pins on the connector but can some one tell me which pins I should check with multimeter and what is the correct ohm value?

With the ICV and AFM off, I tried to crank over. Still won't start; but for a few cranks it felt like the engine was almost trying to start for a split second but couldn't. After that, the DME did return the correct faulty code 1262 (ICV problem b/c I disconnected it)

If the no-start was due to no air getting in, why won't it start with open air ducts?

Cobra Jet, if the camshaft sensor helps the DME on sequential fuel injection control, could a faulty cam sensor mess up the fuel injection or spark timing and cause no-start? I am asking b/c I read from the thread (
http://www.m42club.com/forums/showth...amshaft+sensor) that engine should run with bad cam sensor. (I posted the question there, too)

Thank you folks for giving lots of help on this. I am going to get some carb cleaner now.

[RED IS 91]

If you tested your camshaft position sensor  and it was faulty why haven't you replaced it ? :confused:
These sensors all work together through the computer.

[318i91]

If you tested your camshaft position sensor  and it was faulty why haven't you replaced it ? :confused:
These sensors all work together through the computer.

I didn't know the cam sensor was bad until last Thurs. I did google search and learned that a faulty cam sensor (also called cylinder identification sensor) cause some issues, but was not convinced that a bad one would prevent an engine from starting. See this article: http://www.diagnostic-assistance.co.uk/cid.htm
Besides, I have been as puzzled with my DME code as Cobra Jet with his b/c the it never returned a faulty code 1244 (for cam sensor problem)1444 code, even with the cam sensor disconnected!

So I am not sure the cam sensor is the culprit for the no-start problem.

That said, I am placing an order for a new one and hope it will solve the problem. Will keep you updated when I receive it.

The ICV is now cleaned and functions well. Thanks to Cobra Jet's advice.

[RED IS 91]

If the cam sensor is faulty the computer won't be able to tell the injectors when to squirt fuel into the cylinder.

[Colo318IS]

thanks to everyone for documenting these non-start issues, testing, and remedies...I unfortunately too am having some very similar issues...intermittent starting/running for a couple days/no sound from the fuel pump/etc...I too will be going through the same steps/tests documented here, but had a couple specific questions....

One, where exactly is the crankshaft position sensor located?  Any pictures would be greatly helpful (I'm a visual person)...also, where is the camshaft position sensor located?...

Regarding changing the fuel pump (if it gets to that)...does all the fuel need to be out of the tank to do this or can it be changed with fuel in the tank?  (I have about a 1/2 tank currently)...

Lastly, in the fusebox, there are (oh about 6) 1 inch by 1 inch orange blocks...very stupid question but what are those?  Are they related to any issues with the fuel pump...I had thought one of those might be the fuel pump relay, but after seeing the photo above, I see that that is not the case...I now know where that is at and will look into that tonight...at any rate, I'd be grateful for any insight.

Jay

[Cobra Jet]

One, where exactly is the crankshaft position sensor located?  Any pictures would be greatly helpful (I'm a visual person)...also, where is the camshaft position sensor located?...

A) CPS (crank sensor) - if looking at & standing in front of the engine, the CPS is located at the very bottom to the LEFT of the crank.  The crank balancer is the last "pulley" at the bottom of the engine and has teeth on it (the CPS reads the teeth).  The CPS is secured to the engine via a small metal bracket.  You can see the CPS much better if you are looking under the vehicle (if it's on ramps or a lift).  It's hard to see from the top side.

B) CMP (cam sensor) - if looking at & standing in front of the engine, the CMP is located on the front of the head, right above and slightly to the left of the thermostat housing (above water pump).  The sensor is secured to the head via a small Torx screw.  You can visibly see this sensor when the hood is open.

CPS has a long black wire and is the bottom plug on the black junction box found under the intake on the driver's side of the engine.  CMS has a short black wire and will be the top plug on the same black junction box on the driver's side of the engine.

Regarding changing the fuel pump (if it gets to that)...does all the fuel need to be out of the tank to do this or can it be changed with fuel in the tank?  (I have about a 1/2 tank currently)...

No, the fuel tank does not need to be emptied at all.  

Remove lower rear seat cushion.  On the passenger side (if standing in front of the vehicle), the fuel pump access panel will be on your LEFT after removing lower rear seat cushion.  Remove the (4) Phillips head screws from the metal access cover.  Remove metal access panel cover.  Loosen the  clamps holding the section of rubber fuel line that goes from the metal body fuel line to the top of the fuel pump (it's only a single rubber hose and easy to identify).  Pull rubber line off fuel pump outlet after loosening clamps.  Unplug electrical fuel pump connector from fuel pump.  Unplug fuel tank level sender connector from level sensor.  Grasp the fuel pump and turn it (counterclockwise I believe).  This will release the fuel pump from the top of the tank.  Slowly pull the fuel pump out of the tank.  Be careful when removing the fuel pump from the tank, as there is a black rubber O-ring that is between it and the top of the tank - DON'T LOOSE THE O-RING, you will need to use it again when reinstalling the pump.  Install is the reverse of removal.

Lastly, in the fusebox, there are (oh about 6) 1 inch by 1 inch orange blocks...very stupid question but what are those?  Are they related to any issues with the fuel pump...I had thought one of those might be the fuel pump relay, but after seeing the photo above, I see that that is not the case...  Jay

No, they (orange relays) are not related to the fuel pump in that fuse box.  The only fuse you want to be concerned w/ in that engine bay fuse box will be Fuse location # 11 - this is the fuel pump fuse.  The orange relays in that sealed fuse box are for the headlights, high beams, etc etc etc...  

The actual RELAYS you do need to be concerned w/ are found underneath of the black plastic covering located on the firewall in the engine bay.  This will be located directly below the windshield wipers at the very top of the firewall in the engine bay.  You just pull the cover carefully and it will come completely off (there are clips on the far right and far left that you have to unfasten underneath of the cover to let it come loose).  Once that cover is removed, you will see the (3) relays on the DRIVER'S side of the firewall (DME relay, Fuel Pump relay & the O2 relay).  These are the problematic relays on some vehicles.

~~

If you have an "intermittent" problem, I would definitely start with ohming out the CPS - the CPS can or will be intermittent when is starts to fail or it can (and will) just fail completely.

If a fuse fails, it fails completely and is dead (blown), there are no intermittent signs prior to a fuse failure, it just dies instantly and anything linked to it will be inoperable as well.  A relay on the other hand *could* be an intermittent problem if the relay contacts inside are sticking.  I've heard of other people taking apart relays and cleaning them w/ Eletrical Contact cleaner, but I have never tried to do this myself.

Hopefully this thread and the above will help you out.  Do post back any other questions and/or results!

[Colo318IS]

Thanks so much for your assistance/response...I will report back after I've gone through the tests...

Jay

[RED IS 91]

CAREFUL
remember,If your changing the fuel pump be very careful about sparks and ventilation (fumes)while removing.
Have both doors open ,sunroof too if you have one .have a fire extinguisher near by.

[318i91]

If the cam sensor is faulty the computer won't be able to tell the injectors when to squirt fuel into the cylinder.

I replaced the cam position sensor yesterday. The bad news is that the engine still won't start. So let me recap:

1.  There is spark
2.  Fuel pump relay and fuel pump works and the engine should be getting fuel. I could smell gas in the tail pipe after crank-overs
3.  According to Cobra Jet, if there is spark the CPS and DME should be working. The CPS did read a resistance in the spec (could a faulty CPS still fire up the sparks but mess up the spark/injector timing?)
4. The ICV should be working and I cleaned it.

So why the engine cranks normally but won't start? Did I miss anything?

I am going to follow B318M42W's word and test the compression next. But my gut feeling tells me the problem lies somewhere else. I am saying so b/c the car was operating normally the day before the no-start. Can all 4 pistons or the valves be damaged all at once without warning signs?

This must be among the most mystorious no-start problems. Thanks to you guys for stay with me and I still need advice.

BTW, Looks like there is continued interest on this topic. Colo318IS, good luck.

[Cobra Jet]

318i91 -

How old are or how many miles are on your timing components or do you know if any of the timing components have ever been replaced?  The reason I am asking is another type of non-start issue that could occur IF you do have spark is that the timing components are no longer within spec or alignment.  If somehow the timing chain skipped, the chain broke or the timing sprockets have become misaligned, then the engine timing will be affected which could create non-start issues.  

If you are indeed getting spark then the car should fire up providing you are also getting fuel AND air to the engine...  You have said you are getting spark for sure, so it would seem that your ignition coils are productive and the CPS may be functioning properly (or at least within spec).  Also, from what I was told in this very thread, if you see the check engine light when you turn the key to the "on" position and are able to get it into the code extraction mode, then the DME Relay is good, the DME is getting power and supposedly is working.  I suppose a DME could have other issues which may affect it's internal data operations, but I don't know how to diagnose DME internal failures.

What was your CPS readings?  Did you take a reading w/ the CPS removed from the car and also take a reading w/ it installed?  

Just curious too - when you removed the old CMP, what was the reading w/ it removed from the car?  If you still have it, see if it still tested out at "0" as it did when you initially tested it.

What is puzzling is you are saying you have spark...  Is the AFM and ICV within spec too (both components regulate air into the engine)?

If the fuel pump is working is fuel definitely getting to the injectors is each injector firing or spraying fuel as well?  Have you tested the fuel pressure?

[Colo318IS]

thanks Red IS 91 for the cautionary notes...yes, definitely will be careful when messing around with the fuel pump (I think I've read and re-read this thread I dunno how many times and probably a couple more times:))...good luck to you to 318i91 as well, sorry to hear you're still havin issues...I'm hopin to get some time this weekend in to work on it...I'll stay in touch!

[318i91]

318i91 -
How old are or how many miles are on your timing components or do you know if any of the timing components have ever been replaced?  The reason I am asking is another type of non-start issue that could occur IF you do have spark is that the timing components are no longer within spec or alignment.  If somehow the timing chain skipped, the chain broke or the timing sprockets have become misaligned, then the engine timing will be affected which could create non-start issues.  

155K on the engine and I don't think the timing chain has ever been replaced. I wish the timing chain did not skip b/c I may lack the skills and equipment to fix it.

- If you are indeed getting spark then the car should fire up providing you are also getting fuel AND air to the engine...  You have said you are getting spark for sure, so it would seem that your ignition coils are productive and the CPS may be functioning properly (or at least within spec).  Also, from what I was told in this very thread, if you see the check engine light when you turn the key to the "on" position and are able to get it into the code extraction mode, then the DME Relay is good, the DME is getting power and supposedly is working.  I suppose a DME could have other issues which may affect it's internal data operations, but I don't know how to diagnose DME internal failures.

I am crossing my fingers on that.

What was your CPS readings?  Did you take a reading w/ the CPS removed from the car and also take a reading w/ it installed?  
Just curious too - when you removed the old CMP, what was the reading w/ it removed from the car?  If you still have it, see if it still tested out at "0" as it did when you initially tested it.

The CPS has 520-ohm resistance. I did not remove it as it's not easy to access. The OLD cam sensor didn't have a reading when it was in. Then I removed it and actually could read 1260 ohms which is in spec. After careful check, I found a crack right where the cable going in the sensor head. I think the the cable has poor connection at the crack so the sensor did not read, but once removed the cable could flex to better connection and read ohms. The NEW cam sensor reads 1420 ohms, really at or off the upper boundary of the spec.

What is puzzling is you are saying you have spark...  Is the AFM and ICV within spec too (both components regulate air into the engine)?

When I took it out of the car, The ICV initially read 8 ohm which was in spec. After soaking it in cab-cleaner for 30mins(not sure If I damaged the coil by doing this), it read 2-3 ohms but the valve still operated when 12V was applied from a power adapter. Now, the ICV's  does NOT move when I put it back to the car and turn the ignition to "ON" position. I could hear very faint buzzing only with ears directly on it. but the metal valve does not move at all (looking from the openning). The cable to the ICV valve does have 12V reading on it.

There are 5 pins in the AFM. Which pins should I check?

If the fuel pump is working is fuel definitely getting to the injectors is each injector firing or spraying fuel as well?  Have you tested the fuel pressure?

How do you check if the injectors are working?

Going to buy a compression tester now and check compression tomorrow after work.
Thanks