BMW M5 Forum and M6 Forums banner

321 - 340 of 383 Posts

·
Registered
Joined
·
4,488 Posts
Why only Oz and UK cars? Maybe the tune in Oz and UK in the early 2000's was not so anal about misfires and emissions as the US, so when the CKPS signal gets weak, the DME uses the CMPS signal instead to identify firing TDC and fire plugs and injectors. Because the CKPS signal isn't necessary to do those basic functions. So the DME watches the CKPS signal for one crank, decides the signal is unreliable, and uses the CMPS signal on the next crank and the engine starts.

As I recall from that long thread, this tends to be a cold start, first thing in the morning phenomenon, so why is the CKPS signal weak just under these circumstances? I don't know, but that has never stopped me making wild-*** guesses.

The CKPS is a pure inductive sensor, a coil around a permanent magnet. When a flywheel tooth passes the sensor, the magnetic field in the sensor ramps up and generates a voltage in the coil.

It seems this type of magnetic circuit is sensitive to air gap, in this case the gap between the sensor and the flywheel. The bigger the gap the weaker the signal. And the effect is not linear.

Say the gap was a bit too large on a cold engine, so the signal to the DME is weak. First crank fails. Next crank, the DME uses the CMPS signal. Then when the engine has warmed up, the gap between the CKPS and the flywheel decreases and the signal improves, satisfying the DME every time.

I must admit I struggle with my own logic here because the flywheel is steel and the bellhousing where the CKPS is mounted is aluminum. Aluminum has a coefficient of thermal expansion twice that of steel, so if the flywheel and the bellhousing heated up to the same temperature, the air gap would increase even further. But say the bellhousing is cooled by the air flow over it, whereas the flywheel is so heated up by conduction from the engine and clutch operation that the gap closed?

So in summary, get the shop to check if the CKPS to flywheel gap is in spec. :)

Malcolm
 

·
Registered
Joined
·
351 Posts
Thanks Malcom and Sailor

The shop said that the sensor could just fail or have a problem and it wouldn't be a problem related to the work he has done, i'm sitting on the fence here. He said that the sensor was the first thing he removed and was extremely careful and the no crank first start can be a number of things. I'm uncertain what to do at this moment in time but my first course of action is resolving the acceleration issues I have. I've created another thread and my next step is replacing the TB vacuum line as another fellow Oz beast owner did a number of things before resolving his similar issues with replacing this line.

Once I have resolved this issue and if it is still not cranking first time then I will go back to the mechanic. We have very strict laws in Oz for professional car repairs protecting consumers very well.
 

·
Registered
Joined
·
374 Posts
Hi Guys,

Well it's been a while since I've had much time to play with the M5, but this issue seems to have been very inconclusive to resolve.

The only person who I assume solved this (other than those who's fuel pump was at fault) was Soarbank with a new Charcoal Canister, but have not heard if the fix was permanent. GreenM5 - very interesting that your service history shows charcoal canister replaced but the problem persisting...

So in summary, the issue some of us with (predominantly) Australian cars face is that when the car has sat overnight (or longer), the first start results in an initial stumble as though the car is about to start normally, followed by an extended crank as long as you are prepared to sit there, with no further stumbles. In some cases the problem can occur a second or even third attempt, followed by a normal start. It seems the ECU detects something it doesn't like, and shuts down that single start up attempt.

No amount of priming or re-priming of the fuel pump prior to the start attempt makes a difference, neither does loosening the gas cap overnight after shutdown (to avoid vacuum pressure buildup in tank). Starting with or without MAFS connected yields identical results. My car has no codes, and has had new OEM fuel pump, fuel filter, spark plugs, 4 x CPS, 4 x O2 sensors, thermostat, engine temp sender (on thermostat), MAFS, full vanos board maintenance, and no disturbance of the crank position sensor in my ownership (8 years and currently 180,000 kms).

I need to re-read this thread, but I'm sure Soarbank found there was still spark during the start attempt. However, I have now tested and found there is no fuel supply - I assume injectors are disabled. To do this, during an aborted morning start attempt, we continued to crank the car and sprayed some fuel spray into the plenum. Result was the initial expected kick, but the difference was that there were further stumbles as I sprayed the fuel mixture. These usually do not happen. Therefore the engine must usually be missing fuel, and spraying the mixture allows the engine to stumble.

As you will see the re-start is normal - obviously the injectors are re-activated by the ECU. The question is WHY??!

Would be keen to hear you thoughts...

Nick

https://www.youtube.com/watch?v=LkGdbL_QaWc
 

·
Registered
Joined
·
4,488 Posts
Nick, You must be getting desperate to fix this so I suggest you buy or borrow an automotive oscilloscope so you can do some high tech testing. ebay has Chinese ones that are PC-based for about $100 that work quite well.

Removing the ebox lid then allows you to probe the wiring from the DME to the coils, injectors, CMPS's and CKPS quite easily. That way you can check for a good 12V firing signal to each coil, whether you truly are getting no injector signal, and whether there is anything to my theory above about the CKPS being weak.

You need to do more in-depth testing to get to the bottom if this.

Malcolm
 

·
Registered
Joined
·
374 Posts
Thanks for your replies & interest guys...

Question is, if CKPS signal is weak, why is there still a correctly timed ongoing spark? I could hear the car stumble & try to start with some of my fuel sprays that were drawn into the throttle bodies, so there HAS to be a spark being timed from the CKPS signal during the initial crank cycle. The only missing element must be fuel (which I manually spray in). And it can't be a "fuel supply" issue with a crank that long, surely it can only be deactivated injectors that then re-activate immediately on the second attempt?

Really weird, and I know an oscilloscope would prove it which I can try to organise. But I'm almost certain from my test that everything is working except the injectors, that DO begin to work (hence initial stumble) and then get cut for the remainder of the crank.

68FB, as a first step I could change the CKPS to eliminate that?

Nick
 

·
Registered
Joined
·
374 Posts
One clarification - in the video I was not spaying fuel when the initial stumble occurred - that is normal and occurs whenever this problem happens.

But in normal circumstances, there are definitely no further stumbles, and the extended crank is completely void of any sign of the engine attempting to fire. So the video shows unusual further coughs from the engine that can only be related to the fuel spray. Hope that makes sense?

Nick.
 

·
Registered
Joined
·
4,488 Posts
Yes, it does make sense, but I still have some doubts about if it is spark, fuel or CKPS. You haven't confirmed you have a good continuous spark during cranking, have you? If I read you correctly, you are assuming you have a good spark based only on the fact you get a stumble on the first crank. That's a bit tenuous. You need to confirm that, either by pulling a few coils and plugs, and grounding them (very important) and watching for spark, or the oscilloscope option.

My doubt is compounded by the fact you got so little firing with the ether. In my experience a car will fire right up and run for a bit with a good shot of ether. Maybe you didn't spray enough in. It has a long way to go to fill the plenum and then get through the IACV and all that hosing. Maybe try again with a bigger shot. Or pull the IACV inlet hose off the plenum and fire right into the IACV. Anyway, the poor firing with ether makes me still suspect spark.

As far as the CKPS, see my 28 Nov post above. A bit convoluted, I know, but you have an intractable problem here. I think it's risky to just assume the CKPS is good.

Malcolm
 

·
Registered
Joined
·
10,181 Posts
Sorry but I don't think you have confirmed anything other than the car won't start. It sounds like a car out of gas. I don't think you can assume that there is spark or that the injectors are not firing with what I saw. I wonder if it is trying to start on the wrong firing order? I don't think this V8 has this problem because of the CPS, but I don't know. With only the CKPS a V8 does not know if it is on the exhaust stroke or the firing stroke, but manufacturers have good strategies for that.
The other question I don't have an answer for is if there is no acceleration of a piston/flywheel what does the start strategy do?

My experience is if there is not enough gas or no gas the ether will not ignite on a gas engine. On a diesel yes but ether likes to ignite on heat not spark, but it will ignite on spark.

My inclination would be to pull the top of the plenum and use a squirt bottle and spray gas, not ether into the ICV. I think I would go farther and remove the whole plenum and hook up a volt meter to the injector or use a node light and spray the gas into the loose hose for the ICV.
 

·
Registered
Joined
·
374 Posts
Maybe you didn't spray enough in. It has a long way to go to fill the plenum and then get through the IACV and all that hosing. Maybe try again with a bigger shot. Or pull the IACV inlet hose off the plenum and fire right into the IACV.
I think this is pretty spot on - that can of ether is over 20 years old & 2/3 empty (back from when I had a car with a carby!) and not spraying with much velocity. Plus as you say it's not a direct spray into the IACV. But physically being there and hearing, and being very aware of the usual symptoms, I'm 100% convinced that when the ether spray did manage to make it's way through (mainly from the first few sprays) it produced a reaction from the motor that is certainly not usually there.

But will pull the plenum for the next experiments! Problem is that at least 2 out of 3 times the car starts normally with around with a 24 hour delay between tests, as once it starts once it's perfect until it sits overnight...
 

·
Registered
Joined
·
351 Posts
No disturbance of the crankshaft sensor? I replaced mine and the condition between the two was quite different. I haven't had any cold start issues yet.

Another difference on the sensors was a tiny wedge of metal making the metal end not completely round with a flat spot.

I will report back but my opinion was that it was disturbed when I got the clutch/DMF replaced as I hadn't any cold start issues before the work.
 

·
Registered
Joined
·
374 Posts
No disturbance of the crankshaft sensor?
Mine has been doing this for two years.....:nono: I didn't worry about it too much as I expected the problem to disappear as I worked through replacing all the sensors and other electrical components as part of the 'restoration' of my car. However, the problem remains. This is where I am so far;

Replaced in last 2000 Kim

Inlet and exhaust cam sensors
Crank sensor
Fuel pump
Fuel filter
Pressure regulator
Fuel injectors serviced and leak tested

Replaced in last 5000 Kim
Battery
Ignition electrical switch
Coolant (DME) sensor
Coolant outlet sensor
Oxygen Sensors (pre-cat only) :1:

In last 10,000 Kim
Complete set ignition coils
Spark Plugs
Pair of MAF's
I've never had cause to touch my crank position sensor in 8 years of ownership, but I guess there may be an issue with it. However re-reading I notice others in this thread have replaced crank position sensor without resolving this particular issue. However yours was recently disturbed so finger's crossed this is the problem and let us know results...

Edit: PS I've changed all of the above also expecting problem to go (except crank sensor, coils & ignition switch).

Nick
 

·
Registered
Joined
·
10,181 Posts
Don't assume that two problems can't have the same symptoms. Although many have changed the CKPS and nothing happened some have and it worked. Since yours does not respond to the vapor issues and others did.... maybe your problem is different. Frequency of your problem seems to be less.
 

·
Registered
Joined
·
351 Posts
I've had the car off for two full days now and will keep it off for another day and go for a first crank start tomorrow. Watch this space. I'm not overly optimistic.
 

·
Registered
Joined
·
351 Posts
I have done three cold starts now with one in the morning outside of the city and parked out in the elements. I haven't had any first time start issues since replacing CKPS.

Just to make sure I was correctly diagnosing the CKPS as a possible culprit, I changed this separate to the CPS/TPS. I did CPS/TPS first and I had the first time start issue as expected, I then fired up as normal on the second crank. After I proceeded to change the CKPS and since then I have not had any first start issues.

When I get a chance I will post a pic of my old CKPS.
 
  • Like
Reactions: xrviz and M5 Sydney

·
Registered
Joined
·
351 Posts
Still no cold start issues. Last week I left it for 5 days before starting her up and no problems.

Definitely the new crankshaft position sensor was the cure for me.
 

·
Registered
Joined
·
351 Posts
You had your bios set to boot from a detected USB port - just a little off topic!!
 
321 - 340 of 383 Posts
Top