Yes but the fuel pressure has to leak to zero fairly quickly. Then when the pressure is zero there still must be enough heat to force more out. Where something can get out, something can get in under the opposite force.

 

In Finnish M5 E39 forum there was a similar thread last autumn, the solution there was to change the fuel tank activated carbon container (#10).

RealOEM.com   BMW E39 M5 EXP.TANK/ACTIVATED CARBON CONTAINER

 

Thanks MaSak,

That's an interesting thought, do you have any more detail? Is there a test which can be performed? I wonder how the canister could cause this issue.

Any thoughts?

Cheers

 

Do you ever get a "broke the seal" noise when you open the gas cap. If the charcoal is plugged you basically have no vent to the fuel tank.

 

Not that I recall, although I suspect I may have noticed this if it was the case. I'm due for a fill-up so ill check it out tonight, we've got 30c here today so it'll be a good time to test.

Cranking; cranked it for almost 20 seconds this morning and no hint of a start, just turning over (quite well actually, battery is gooood!)...of course, second attempt fired immediately as usual.

I'd like to test if the injectors are being fired, and I have a noid light for this, but on the S62 this presents a fair amount of work - is this a worthwhile task to undertake?

Will report back on any pressure in the fuel tank.

Cheers

 

It will not be pressure likely vacuum in the tank. You could just try a start with the gas cap off. You will likely get a code if you leave it off to long so be prepared to clear them. B4 and B5

 

The problem can't be be back at the tank, can it it? Because soarbank is getting good pressure as soon as he turns the key on, so he has good pressure at the injectors. The only thing that would prevent normal injection then would be air getting into the fuel lines under the hood as suggested. That sounds like an unlikely scenario: pressure dropping off quickly to zero after hot shut down due to a line leak, bad injector(s), bad FPR, or bad check valve in the pump, then hot fuel in the lines cooling and shrinking enough to generate a vacuum, then an o-ring seal or similar that seals under pressure but lets air in under vacuum?

The best way to check for this is a prolonged first crank. If it is air it will eventually flush through and allow a start.

Edit: just read post 37.

OP, I suggest a spark test first just to rule it out, and its easier than an injector test, I would think.

It takes air, fuel, and spark to start. What about the IACV sticking closed when it cooled down? But why would it free up after one crank?

 

He did that long crank and nothing? I myself am out of ideas, grasping for straws so to speak. This beginning to be remind me of the Vanos problem on the E34 transplant. :Thumbdown: :viking2:

 

OK, will conduct spark test in the morning (and for once hope that it doesn't do the 1 in 10 'start first time' thing....) - will report back.

Had a quick listen to see if there were any sounds at the fuel tank cap when removed, no apparent vacuum (nor pressure for that matter). I was kind of hoping for something there. I will try a 'fuel cap-less' start the day after tomorrow.

I'll also fire up INPA and see if there are any codes, but I'll admit that I've only used it twice and I'm not sure if I'm looking in all the right places when looking for codes.

Cheers

 

Oh, you've got INPA. Well check out this thread.

http://www.m5board.com/vbulletin/e3...-m5-e52-z8-discussion/211022-my-m5-wont-start-just-only-sometimes-vid-inpa.html

OP and I posted some INPA cranking test screen captures but they were a bit inconclusive. In short, INPA may be able to monitor injector pulse width during a no-start crank to confirm injectors are pulsing.

Too bad the thread kind of peters out after detecting delayed coil supply.

 

I think I have it. Have you tried your other key? Do that first if you have one. If not then try these steps as diagnostics.Try (put it on your list) to put the key to position two but don't turn it over count to 20 then turn the key back then to start. It might take a bit more, not sure I fully grasp what I am reading it is in Russian. If that does nothing try just clicking the key to start and then try your second start if it fires right up it is a key problem. Has Something to do with rolling codes and how the coil works.

 

Have you tried your other key?.

Sailor - tried the other key this morning - no difference.

My plan is to systematically go through all the tests highlighted in this thread and report back, this will be on Saturday, hopefully the momentum of this thread will continue as I feel progress is being made.

Cheers

 

Sailor,

Sounds plausible...... Especially given I do seem to remember a complete no start once (no cranking) and it threw an EWS fault. This was only once though and as I say, no starter motor (as you'd expect with no key reading).

I'll try this tomorrow morning as I will not have enough time tomorrow morning to do the spark test.

68FB - I think I need to hookup INPA and watch for the injectors, confirm spark and come back with the results (if the spare key doesn't work), but this will not be until the weekend - I've got three 10 hour days ahead of me.

Has anyone had any trouble with Deutschepart OEM ignition coils? I have eight of these in there at present...just a thought (although I'm conscious that I may just be clutching at straws now)

Cheers,

 

Hello Soarbank! The troubleshooting went on in the Finnish case(s) for months, they changed a fuel filter (that helped in one case), pressure regulator, spark plugs and a LOT of other stuff. Finally the guy took his car to BMW service and they told him it is the carbon filter. He changed that and the problem was solved. There was also another case that I just read about, somebody had a 15A fuse for the fuel pump, and the fuse should have been 25A. This might not have had anything to do with the problem, but at least it is an easy check. Mysterious problems, I hope you got it sorted soon. Good luck!

 

In researching my occurrence of this problem, I remember coming across a thread where they found an issue in the ignition switch. Something to do with a bad contact in the switch not enabling the spark circuits. I can't find it at the moment. I guess when you test the spark that will either be ruled in or out.

 

Gzig5, ignition switch has already been replaced - this was done to cure a terminal 15 fault code about 9 months ago. The 'second start' issue existed prior and post fitment of this switch. It was a genuine, new ignition switch.

Cheers

 

Still no codes?

 

There is one relay supplying all the coils. Could it have bad main contacts that need to be wiped clean by one cycle when cold?

The spark test may shed some light.

 

You stand a chance with that one maybe? Think it might be too consistent as second start always works, but we are running out of good bets so.... there is an interference cap at the end of that circuit anyway it robs all the power of the first try????

 

Yeah, the consistency of the second crank start really undermines all the random failure ideas.

 

Had to reinstall INPA last night (long story - why did IBM think it was a good idea to default to supervisor passwords on BIOS) onto my main laptop, it didn't go so well (64 bit Win 7) but should be good tonight. I'll dig around in there for codes and come straight back.

Yep! There's very little that is random about the behaviour, it's consistently 9 or so times out of ten, and I reckon if I recorded the exact environmental details when it does start first time I'd expect that there would be a consistent element there too. I think logic will prevail and through a systematic approach and the help of this board we'll identify the issue and have the information there for anyone in the future.

In the meantime however.......grrrrrrr

Where's the capacitor at the end of this circuit located? Might be worth a quick look?


Cheers

 

Search X6830 in the wds and you will find I01100 at the end of the string for bank one then I01101 for bank 2.

 

I'm in the same situation with my car - why does it seem to be so many Australian ones??

There's one thing I can add that may assist - I've noticed that mine often gives a little "cough" on the first crank, but after that it just cranks with no hint of a start and no more coughs. So I've always suspected that the DME won't let it start once it's had some sort of aborted attempt - hence if you crank for 20 seconds on the first attempt nothing, but starts right up on the second attempt. Have you noticed anything similar?

So my situation is:
- only happens when cold - ie overnight,
- usually starts on second attempt, somethimes third
- holding key so I can hear the fuel filter run (sometimes even 2 or 3 times) to try to build fuel pressure before trying to crank has no effect

I've changed in the last 10,000 kms all 4 CPS, vanos o-rings, ECU temp sensor, fuel pump and filter, spark plugs, thermostat, under plenum vaccum hoses, MAFS.

I have discovered a crack near the front o2 sensor bung, so was going to have this welded and change all 4 sensors to see if this made any impact before starting a thread on this issue, until I saw this one (I have ordered the sensors so will do this shortly and provide an update).

I'm at 170,000 kms and had no peake codes but recently got a pre cat o2 sensor code.

My only other thought was that I ran the car with a bad thermostat for a while, and also what turned out to be a bad ECU temp sender unit (on the thermostat), as once I changed that the car ran much better. SO I was wondering if it had been running really rich and fouled the 02 sensors - but I'm not sure if they would impact starting?

The charcoal cansiter may be an interesting theory - I may try to swap with another local M5 owner and see if it moves the symptoms to the other car...

Nick

 

I'm in the same situation with my car - why does it seem to be so many Australian ones??

There's one thing I can add that may assist - I've noticed that mine often gives a little "cough" on the first crank, but after that it just cranks with no hint of a start and no more coughs. So I've always suspected that the DME won't let it start once it's had some sort of aborted attempt - hence if you crank for 20 seconds on the first attempt nothing, but starts right up on the second attempt. Have you noticed anything similar?

So my situation is:
- only happens when cold - ie overnight,:hihi:
- usually starts on second attempt, somethimes third:hihi:
- holding key so I can hear the fuel filter run (sometimes even 2 or 3 times) to try to build fuel pressure before trying to crank has no effect:hihi:

I've changed in the last 10,000 kms all 4 CPS, vanos o-rings, ECU temp sensor, fuel pump and filter, spark plugs, thermostat, under plenum vaccum hoses, MAFS.


Nick

Nick' my symptoms EXACTLY, you have described my car perfectly....sometimes a little cough on first turn, but if it doesn't fire that the end until you turn to position 0 and try again, then it will always start second time. I've hard mine go to a third time only twice in two years though, it's a pretty rare occurrence..

...and yes, only if left overnight or for an extended period, mine also does it after sitting in the car park at work for 8+ hours. If I go to the car at lunchtime to head out, it'll start first time.

Thanks for joining the thread, your input can n,y serve to help us both.

Cheers,

 

Thanks Sailor, Again, appreciated.

I'm at my electrical knowledge limit now, to test these caps I'm suspecting I can send a voltage down the circuit and then measure that voltage decay over a short period? That's how I would measure a normal (circuit board) capacitor. A faulty capacitor would measure to zero almost instantly wouldn't it? ouich

Cheers,

 

Nick,

O2 sensors made no difference to mine at all. Also, the after cat O2 sensors just measure if the cats are doing their job, which in Australia will not be an issue or even throw a fault code. My understanding is that they do nothing other than this function.

If you've been running cool for a while and want to give yourself something to do on a Saturday that does make a nice difference to the car, pull off the intake plenum, detach the throttle linkages, rotate a throttle body 90 degrees and take a look at the carbon under there, it'll be pretty bad if its not been done before. Mine was horrific with a thick (min. 1mm) build up of carbon.

You'll need a tin of Liqui-moly throttle body cleaner otherwise its an eight hour job. The stuff is brilliant and only $25 a tin, which easily does all eight.

It doesn't solve the no start issue though!

Cheers

 

Nick,

If the Vanos wasn't returning to correct rest position it would throw a code so I reckon we can rule that one out, given that we've both replaced our Vanos seals too (I think I missed that off my list of jobs competed)...

I've replaced both DME coolant sensor (blue one on top of thermostat) and coolant outlet temp sensor (in bottom hose). :dunno:

Does yours start first time on really cold mornings or the morning after a very long (200klm+) run?

 

Nick,

If the Vanos wasn't returning to correct rest position it would throw a code so I reckon we can rule that one out, given that we've both replaced our Vanos seals too (I think I missed that off my list of jobs competed)...

That's true, but where does vanos move cams for initial start up? I notice all our cars are pretty early build, without the updated vanos accumulator. Mine starts up with a real vanos rattle before it settles down (despite new chain tensioner), so I wondered if it's leaking down too much oil pressure to get the cams where they need to be for first start up? Unless cold start position is the same as initial position? I also notice when mine does first fire it's a bit "lumpy" which I wondered if it was the cam timing sorting itself out as oil pressure built.

You have both done so much work on your cars and ruled out so much stuff, this suggestion is beginning to sound reasonable to me. Does the DME see something it dislikes so much early in the first crank that it prevents the start? It has a well-known ability to be reset by a simple shutdown/restart after it has gone into limp mode, so maybe that's what's happening here?

If it is preventing a start, it'll likely shut down both spark and fuel. The spark test and INPA injection time check will show if this is happening.

I'm almost certain this is the case. Also because if my car ever doesn't start on the second attempt, it's identical circumstances - ie there's one cough and then nothing on the first attempt, then one cough then nothing on the second (no matter how long you hold on the starter), then fires fine on the third. When I had temp sender unit problems, this could go on for three or four attempts, now it usually fires second go. Something's definitely making the DME abort the start attempt in my opinion.

I wonder if he will find a high frequency of shadow code 10? The engine is suppose to stop in a certain position with the cams in a position also.

Thanks Sailor, does this validate a vanos theory?

I'm in Sydney too, and I have the exact same problem.

I've replaced fuel pump, filter, pressure regulator, cam sensors, etc...

I'm going to look into the activated carbon canister, though for some reason I thought that Australian cars didn't have them.

Geez Chris, I didn't know you were in this same situation! I was working under the same misconception as Soarbank that once I finally changed all my sensors, something would sort this out, however not looking like this is the case. I think all cars have the activated charcoal canister (but looks like swapping with you won't tell us much :hihi.

Sailor & 68FB, thanks for getting involved in this - I think we need a bit of Canadian expertese over here!

 

i had this to, but when i repaced all mine both side cam en inlet sensors, After replacing my sensors it was gone.