BMW M5 Forum and M6 Forums banner

1 - 20 of 29 Posts

·
Registered
Joined
·
23 Posts
Discussion Starter · #1 · (Edited)
I worked quite hard over the winter on our newly acquired E39 M5 on a good list of items. Just after Easter the car resurrected. With a couple of residual issues:
1. Tire pressure control inactive, and I cannot fix that.
2. Hard start.

It takes long cranking to start the car. Initially it seems that the long cranking was related to cold start. I also noted that there was an odour of gas just under the driver seat when the car finally starts after a couple seconds of cranking.

So FPR replaced. Before replacement, I tested and noted that the FPR was in fact not holding pressure, and there was a leak from the diaphragm as well. So I thought problem fixed. Nope. The gasoline odour was gone, but not the hard start.

read about adaptation of new CPS sensors on this forum. So I disconnected battery overnight. Next morning, car cranks and started. but it did not last. problem recurred.

Next up, battery. a load test on the battery seemed to suggest that maybe the battery was borderline. When cranking the battery went down to ~9.5V. So there, I found the problem. H8 AGM out. H9 AGM in. Problem fixed! car started without delay twice. Then the long crank came back.

Nowadays it seems the long crank happens more or less sporadically. some morning when I crank, it starts within a couple cranks. Other times when engine can be warm (or cold) and I walk out of the grocery store 10 minutes later, i have to crank for up to 2-3 seconds. it always eventually starts. And when it starts, it runs, no stalls and it runs well.

No codes from INPA/Peake/Autel. other than a couple of misfire when warming up after a long long crank

CAM position sensors and throttle position sensors all new

plugs a couple years new.

new Bremi coils

injectors cleaned with good and matching performance. when testing new FPR I clamped all hoses to test pressure to injectors, and pressure was holding. No gasoline smell in engine bay. Injectors inspected (as much as the video revealed) with boroscope without any overt evidence of gasoline.

Any suggestions?

starter? what are symptoms of a failing starter? Would it give me sporadic long cranks? how to test? Just throw a new one in? Curious why it started just fine before I worked on it, and it doesn't work anymore after...

CPS? if that is it, why doesn't it throw me any codes? I would expect CPS to cause trouble not only when cranking...

I am at a bit of a loss.

Input greatly appreciated.
 

·
Registered
Joined
·
10,454 Posts
when testing new FPR I clamped all hoses to test pressure to injectors, and pressure was holding.
Sort off sounds like a fuel problem, but there are lots of parts to check and I don't want to write it all out at once. You have an idea about the test I quoted so lets get you doing that correctly. At the end I will give you another test that is simple, actually I will explain it now with another thread. It will be helpful for you to know this part so you don't have to start the car during the next tests.
I will give you this link but all you are really concerned with is how you jump the fuel pump relay with a paper clip( I would suggest a larger diameter wire) so you can run the fuel pump and pressurize the fuel system. Do all the tests if you want, can't hurt to be thorough.
DIY: - Suddenly stranded? Testing your fuel tank levels...
So have a read of that. First thing I would do is jump the fuel pump listen that the pump is running, leave your jumper in walk to the seat count to ten and try and start the car. Even if the car does not start does not rule out a fuel issue, but it does rule out most but one or two. My guess is it will fire right up unless the back flow valve at the fuel pump has failed, that adds a twist. BTW that is the only time I mean to leave the jumper in, take it out after the system is charged for the rest of the tests. Get back to the jumper quick and pull it because they can get hot if you use too small a diameter, but that is clear in the other thread.
I will ask now if you have ever changed the fuel filter? If no now would be the perfect time.
Next after that is to look at the fuel system for internal leaks.
You charge up the fuel pressure with your gauge attached and the car not running by jumping the fuel pump relay. Then just watch the gauge and see if it falls. If it does then charge the system again and once at pressure pinch off the line coming to the fuel filter from the tank, that tests the back flow at the pump. If the pressure falls the same as it did. Charge it back and pinch the return line from the FPR and watch. Last is the line to injectors. Sometimes you have to pinch off 2 lines because you have two leaks.
Just because it does not fall right away my not be an indicator. Sometimes you have to leave it for 1/2 and hour and come back and check especially if the leak is at an injector. A bad spring in an injector can be overridden by the pressurized system, but at a certain point the pressure will drop like a rock.
If it does not fall at all check it in the morning but don't leave the gauge on it while waiting they do cause tiny leaks and will give you a false positive. Even if the gauge says 0 in the morning that can still be acceptable, so look for even the slightest pressure by pushing on the pin where the gauge hooks up and listen closely also. My bet is you will still have a small squirt of fuel, but you might hear a gurgle or air noise.
There are lots of other things to look at but fuel is important to rule out first, the very first test will rule out some electric issues also.
Let me give you a run down on how a Can car fuel pump works, you might have a US car and it might work different, but might be the same. Reports on the board show they all don't do the same thing.
Year seems to change things also, but these are the basics for a Can car. If the car is dead cold and the key has been out for hours the fuel pump will run for a couple of seconds to charge the system, then it runs for another couple of seconds while cranking but not constantly. So on a failed start if because of fuel it is a faster start if you stop and try again rather than cranking. At some point however the fuel pump has had a few cycles the DME will not activate it. The DME control will never stop the car from starting it is just smart enough to not flood the car.
Now if the car is warm, the DME may not fire the fuel pump until the engine does or at least until the DME detects accel in the flywheel. It will run the pump if you have some number of failed tries. So again holding the key is not the best approach, turning the key back and trying again is a better approach.
 

·
Registered
Joined
·
23 Posts
Discussion Starter · #3 ·
Thank you Sailor24.

1. yes, fuel filter is new, so are injector o-rings, after injectors are cleaned
2. no I did not do a relay check. that is because as soon as I crank key to second position, fuel pressure goes to 70 instantaneously, predictably. I have not observed a low pressure when key is turned
3. I did pressure decay test (over 7 min) with old FPR and noted a significant drop in pressure, then I proceeded to change the FPR and retested. To isolate all possible leaks, I clamp 3 lines at a time (4 total including the vacuum line, just so there is no doubt) with the new FPR in place, and noted the pressure decay over 7 minutes. Please see attached table.

So, with these in mind, should I check the relay? Is it possible that the relay is dead, but when I turn the key fuel pressure goes to 70? I didn't think that was possible, so I did not do the test. I would have thought that a dead relay would give me more trouble than just a slow start.

Yes, I did note that the pump does not run again if there is a delay between turn key and crank. And your point of not cranking for a long time but rather crank a second time is well taken. most of time time, it starts before i would give it a second go. and that is almost always less than 2 seconds (which is a long time vs. before my winter project)

Injector spring: didn't think of that. I was focused on o-rings. However, if the injectors are the problem, should I not see a bigger pressure drop when the system is open only to the injectors? The drop is similar across the board it seems, whether it is open to supply, return, or injector. In fact, when I clamp all 4 lines as a control test, the drop was the same.

What do you think?
 

Attachments

·
Registered
Joined
·
10,454 Posts
So, with these in mind, should I check the relay? Is it possible that the relay is dead, but when I turn the key fuel pressure goes to 70?
It is bed time, I will have to get back to you later. No I don't suspect the relay. I am trying to get you to test it the way I would so I will understand the results. Remember I am not there and if you do it different I may miss something. The reason you are jumping it is for procedure. How do you know it jumps to 70?, with a gauge on the valve under the car or do you have some fuel pressure gauge attached, if so where, take a pic.
Explain what you mean when you say you clamped all three? One at a time or all together. & min does not tell you anything 1/2 hour minimum. I explain later how far the pressure has to fall for the injector to start leaking and what happens if they don't stay shut if you are leaking back to the tank. But think about it, with the injector at the top above the fuel level if fuel is flowing back to the tank air can get in and all the fuel drains out. Then the line has a ton of air in it. Pressurize the system and the injector closes with all that air still trapped.
Do the test the way I said so I can understand and listen closely on the morning after test for air, it needs to be very quiet.
 

·
Registered
Joined
·
23 Posts
Discussion Starter · #5 ·
How do I know pressure: with a pressure gauge attached to the FPR housing Schrader

Clamping 3 lines means clamping 3 out of 4 lines, leaving only one open. One of: vacuum line, supply line, return line, or line to engine. You mention in your post that sometimes I have to clamp 2 lines. That's what I did, I just included the vacuum line as well, so I clamped 3.

Why 7 min? Because I saw a significant drop in pressure with the old FPR, and it happened to be 7 min. So I standardized the test to 7m.

I gather you are suggesting that fact that it does not loose pressure in 7 min does not mean it will not in 30min...

What is inconsistent is that sometimes I can leave the car overnight, and it will start right up the next morning. Then I drive it, leave it for a short time, and it cranks before it start in 1-2 second.

What I have trouble understanding is:

1. Even if there is a leak, if the fuel pump pressurizes the lines to 65, and it does not start, I would think it is not a fuel pressure problem. I seem to gather the contrary from you. Is that what you are saying?

2. I have caught a hard start with pressure of 60+.

3. You said that if the car is warm, DME would not kick in the fuel pump, and if pressure is low, it follows that car will not start. This car stays warm for a very long time. How does the DME decide whether to turn the fuel pump on or not, after 2 hours of engine off but engine temp still warm?

Thanks.


Sent from my SM-N975W using Tapatalk
 

·
Registered
Joined
·
10,454 Posts
To start you are clamping too many lines at the same time. Second there is no sense in clamping any line until unless you have identified that there is a leak and when that is happening.
Third is back to the fuel pump relay. The reason we do it that way is so the key does not need to be cycled, which makes a whole bunch of electrical stuff do it's things. Leave them out of the test by not using the key.
The object is to isolate the fuel system, we do this by only bringing it live. Next step identify if there is a leak, all cars will get to 0psi if left long enough. If the time passes so that the car goes into cold start sequence that is long enough to rule out the fuel system. Many cars will be at 0 psi 12 hours after the pump has been cycled.
First step is cycle the pump to full pressure and read the gauge. Watch it for a bit but really check it again in 15 mins and write down the reading then again at 30 mins then an hour, then two, writing down the numbers each time.
That is your base line that you will compare everything to. Every M5 I have worked on will leak back to the tank via the pump back flow valve some more than others. Even totally failed ones will not cause issues if there are no other faults. You can have multiple small leaks and things can still work but a wrong combo of leaks can create hard starts.
Ignore the vacuum line, my method never creates vacuum, just deal with the three lines one at a time.
A little more on what goes wrong with air in the system. If you have fuel draining back to the tank eventually the pressure goes low enough that the pintle in the injector will not be held closed. We can't test if an injector is not staying shut, all we can test is if it leaks under pressure. We can't introduce a vacuum or that drops the threshold of the mechanism holding it closed past it's design. Gravity will drag the fuel to the same level but only if air can replace the fuel.

Get your base line then test one hose at a time, depending on the results you might have to pinch two. That is so much later and only after the results of the other tests say there is a difference in these two tests.
When you pinch the line from the valve to the injectors that will test for air getting back into the system.
Some other things that might be going on are engine shutdown, and a bad unloader relay. On the modern car it is very hard to hear or feel, if it is a fuel problem or spark problem, they just don't start.
The unloader relay can't easily be tested and is not much money so is just easier to replace that to test. Again short cranks and a key cycle can point to them pretty quickly.
The last which is much harder to test is - did the engine shut down correctly? Sort of sounds like your warm 2 second start, but I would have to be there so lets just start with a procedure change for shutting down the engine. On a V engine you have two cyls at the top at the same time. An engine has to know where it is when it shuts down, if it does not then it has a strategy to restart the engine, but that strategy takes a little time to run. If your vanos is not 100% they will not stop where they are suppose to. So lets get you in a shutdown habit and see if you notice anything, blindly chasing stuff is not a good approach. Always return to idle for a few seconds, in neutral with the clutch pedal released and the park brake on so the brakes are not drawing vacuum, so no foot on the brake, then turn the key off. After doing that many times and knowing how the car restarts try skipping the park brake and use the brake pedal to add load to the engine, see if there is any difference. If you park on a slope set your park brake so that the engine can't move from being in gear. We have never figured out the restart strategy or what the car looks for when restarting so you just have to do it all. It is a BMW so maybe if the crank sensor or cam sensor is not in the exact spot it goes through the restart procedure, the only slight hint here is if you have a program that reads shadow codes you might see code 10, crank sensor fault only in the shadow codes not in the pending code list. Even that is still a guess so don't read too much into it. On that note you have no pending codes and you have checked? Have you had the tranny out, did the previous owner have the tranny out? You might need to confirm the crank sensor is shimmed correctly but I would expect other faults if it was not.
I have no idea what is going on with your car but likely one of these tests will show up a little difference. That little difference will likely point you in the direction of what is going on.
 

·
Registered
Joined
·
23 Posts
Discussion Starter · #7 ·
I am sitting by the car frustrated.


1. Check engine light came on today while driving. 175 on INPA AF from Peake. Not sure if that's related to the hard start about 5 minute prior to the CEL...


2. Electronic error. Fuel gauge all of a sudden went way down to almost empty when I know it was half full. This is at the beginning of my fuel pressure test that I noticed it. Hidden menu #6 indicate 2L on left 31L on right. Tire pressure inactive came on without me pressing on the button, and STAYED on with ignition off and key out! I have also noticed recently that fasten seat belt stays on most of the time, but goes away sometimes, without fastening the belt. Coincidence?


3. Fuel pressure. Pressure dropping like stone! This is a drastic difference from when I last tested it a month ago. It went from 65 to 35 in 30 minutes. Fine. Then I clamp ALL lines. 4 lines. Same drop! Leaky fuel pressure regulator housing? I even got a new (but the same) fuel pressure gauge. Same thing. A 3rd gauge?


Not making sense to me. I changed the fuel filter last year, put the housing back in. Car started just fine. Change the FPR a month ago because I smell gas, and fuel pressure was dropping with the old regulator. I even confirmed that the new regulator held pressure, and it did! Gas odour is gone, although the Schrader screw cap looked wet inside.


Between the clamps and the Schrader there are 3 hose connectors, 1 vacuum connector, a FPR, and a housing to filter connection. I suppose any of these 6 things are suspect. The Schrader stem may be leaky, but all that would be insider the pressure tubing, so for the sake of this test, it should not matter.

At 45min now it is down to 10 psi.

It is also a problem that I don't seem to be able to find standard decay values for fuel pressure. So I have nothing to compare with.

4. Relay: there is a slight click at position 1. Not position 2.

5. Pump works when terminals jumped. I can also hear pump works when at position 2 sometimes. I assume the DME is being selective from time to time.

Maybe I just need some sleep and come back with a fresher mind.
 

Attachments

·
Registered
Joined
·
2,588 Posts
1.) I believe AF and AE are lean codes (one for each bank). Sorry for my poor memory. When was the last time you had the plenum off? There are plenty of suspects under there (the other end of the vac line for the pressure regulator for instance). How old are your MAF's? Does it hard start with them unplugged?

2.) I think it's time to focus on your electrical issues. The ignition switch may be faulty.

3.) I think you need to ignore this for now. IMO over analyzing it. If you're getting full pressure when you turn the key on, the system is primed to run. How old is the pump? Have you inspected the connections on the pump? Look closely for a crack in the plastic line, particularly around where it clamps on.

4/5) Correct. The pump does not run every single time you turn the key to "on". It will restart/run the pump once you bump the key to turn over the engine.
 

·
Registered
Joined
·
10,454 Posts
This is at the beginning of my fuel pressure test that I noticed it. Hidden menu #6 indicate 2L on left 31L on right.
How long had the car been sitting without being run?
Why your results are what they were is not really known, but they are more like others have reported in the past. I would be curious to know the pressure at 3 hours which is when the car likely will be dead cold and is most likely in the full cold start mode which almost guarantees full pressure.

There is no single thread with your electrical problems, but there are a few that have had the hard start with a bad unloader relay and it could allow power to other places because it should only be active when those other things are powered. I like ignition switch idea also, for me that has been common to replace/fix part.
AF is adaption code for bank 2 which could be caused by your tests maybe, but most frequently is when the DME has reached it's max adjustment, sorry don't remember exactly and I don't have notes on this computer, but I think it is when it leans out the fuel to the max. I think fuel control is adding fuel, but maybe I have this backwards, my memory is less than unreliable sometimes. That frequently means leaky injector.
Let me get one thing clear that I am still not clear on how often does the car have issues starting warm and how extended are those starts? Are they always the same?
Keep this in the back of your head maybe you have several problems going on and they are compounding things causing you start problems. It seems to me that a major change in the fuel system "fixed" the problem for a short time, then the same when you made a major change in the electric system. Maybe the real issues are still hidden but slowly getting worse which brings the problem out again.
 

·
Registered
Joined
·
23 Posts
Discussion Starter · #10 ·
Thanks to all the input! I appreciate it!

Let me try to answer some of the questions:

Tim,
1. Yeah, many suspects, but I am not planning on attacking it right now. Who knows, with all the hard starts, I suppose I can trigger a rich code?

2. I am suspicious of the switch as well! Maybe I'll throw a new one in?
I looked up menu 14, and got these:
BF 88 000
D7 81 000
90 84 000
11 84 000
C7 81 000

Now according to one explanation of interpretation of these codes by Sailor24, 000 means it has been at least hundreds of starts ago that these codes appeared. If that's the case, these are likely inconsequential for the time being.

Having said that, one of these codes is a terminal 15 code, and I understand terminal 15 is associated with the ignition switch... and then the other 2 from fuel level sender... so maybe these codes are pointing me in the right direction?

My INPA is working on modules, and attached are codes from the cluster. I think they are the same as the menu 14 codes. they are in the memory, but at the same time it says frequency is 0!?

3. so yesterday, I figure that when I turn to position 2, the pump comes on for a couple seconds then stop. it still doesn't start well. the DME inhibit the pump from coming on again if i turn the key to off then to position 2 immediately. but if I wait just a little longer and turn the key again to position 2, the pump comes on again. This is what I did tonight: made sure the pump came on twice, and the car started right up: 24 hours from last start. i'll try that and see if i can consistently start the car after 2x pump prime.

It is as if it take 2 brief pump runs to reprime the fuel hoses

Sailor24, see my response:

how long has the car been sitting without running: from December to April when I was doing a list of refresh. April first start, with a piece of something stuck in one VANOS solenoid, throwing a code, and fixed. detailed in


as well as pinched and broken o-ring from the thermostat housing connector tubes. then car running, and since then I experienced the hard start. We drive the car a few times every week currently.

"Let me get one thing clear that I am still not clear on how often does the car have issues starting warm and how extended are those starts? Are they always the same?"

I think I am starting to be able to reproduce the problem. in the past few days we were up to 30C here, and I left car off for say 15-30min, and I get a hard start. I should use a stopwatch to accurately document how long it takes to crank to start. But if I can count 1 mississippi 2 mississippi properly, I would say it cranks for > 1, and up to 2 seconds.

I should not say that they always require the same amount of time of cranking. For example, if i turn the car off, and restart it, there is barely a crank and it starts. the longest is about 2 seconds, and I have experienced somewhere in between.

Yes, there may be a couple things going on at the same time.

You asked about 3 hour pressure: forget 3 hours, fuel gauge showed just about 15 PSI at 31 min and 10 at 45min. this is with nothing clamped. clamping does not seem to change anything, pressure drop is basically going at the same rate.

I am getting a different fuel pressure gauge.

sprinkling powder to locate leak?

Thank you!
 

Attachments

·
Registered
Joined
·
10,454 Posts
Lets just start with the codes you pulled off the cluster. There are a bunch of codes that come and the first time they set they have a specific number. I forget the number but I think it is in the thread that deals with the secret menu. Not all of the codes have that specific number behind them. None the less whatever the number is it decreases by 1 each time you start the car, you can also do it by cycling the key a certain way but that is moot. Bottom line if you are getting 000 those codes have not been present in many starts. IIRC the lowest number is 200 and something but you would have to check the thread because my memory sucks.

By your 3 hr reading I would say your hard starts probably are not fuel related, but it is still hard to know that a leaking injector is not giving you hard starts warm. If an injector is leaking it floods the car, but if you still have pressure 3 hrs later it can't be a really bad leak.

Lots of people have reported the same sort of thing as you when clamping lines they get no difference. It probably has something to do with the line clamps. I use a C clamp and two dowels and make them tight until I wonder if I am breaking things. I don't really like clamping lines so I bought a set of taps that I put in line.

You wrote so much to explain how long the car was sitting, so I feel bad when I say that is not what I meant. What I was looking at was just a detail and I don't think it could be related to your start problem so it was just an extra.
What caught my attention was the fuel level in the two tanks. If the tanks were reading 2 and 31 that sort of told me you had run the car before you did the test. If the car was dead cold and had not been run then those two readings should have been the same. If the car has not been run the tanks seek one level so 16.5 on both sides or there abouts.
I was just trying to confirm how you did the test. There was a reason I did not want you to activate the car before you did the test. Not important, because I think you might have ruled out fuel.
Just to be sure though since you seem to have some success cycling the key lets rule that out.
I will type it out one more time. Dead cold car, been sitting for at least 5 hours. Key in your hand, don't put it in the ignition until after you have jumped the fuel pump. Jump the fuel pump let run and get to the seat put the key in and go to 2 count to 5 or 10 and try and start the car, the jumper should still be in place.
This is the second part of the test, which I did not mention, because I already typed too much.
Maybe a day later try the same thing but this time first cycle the key many times going 0 to 2 don't rush leave it fo 10 secs when you go to 2. Don't engage the starter. Take the key out and install your jumper wire for the pump and try and start the car. What you are looking for is any difference, or even if it will not start until you cycle the key again.
Might seem this is a fuel test but it is not it is an electrical test. The reason we jump the fuel pump is to make sure that there is fuel in the system and that that is the exact same as the next test. We are over riding that feature in the car that charges the fuel system, because it may charge the system slightly different because of the way it works.
 

·
Registered
Joined
·
23 Posts
Discussion Starter · #12 · (Edited)
Hi,

Once again, your on-going help is much appreciated!

Busy with work the last few days.

Tried first part of the test: car not started for 2-3 days. Shorting the relay to run the fuel pump for about 10s then start car without holding in position 2.

Starts immediately! see below:


Let me clarify one thing: you mention "judging by my 3 hour pressure... Not fuel problem..."

I did not have pressure per my last post. I see barely any pressure at 30 minutes!

However, it made no sense to me that the pressure dropped like stone with all lines clamped. There has to be a leak somewhere, unless as you said, Sailor24, that the clamp was not on tight.

So today I went back to it, and I screw on the schrader hard. I also wiggled all fuel line connections, in case they are the source of the leak.

Interestingly, I had better results. After 120 minutes, no clamp, pressure was ~20psi. This is a major improvement from last time, getting less than 20psi in 30 minutes. All I did was screw in the schrader tighter and wiggled the fuel line connectors.

then I turned to position 2. let the pump run for the 2s and off, and started the car. It starts ok, though it feels like it struggled just a touch, compared to when I jumped the relay a couple hours ago.

While I was waiting for the pressure decay, I used a boroscope to check injectors. It's not 100%, but there was no gross fuel leak. I understand that it can leak inside, however.


Question: How long is the fuel pump supposed to run when key is turned? 2 seconds, like in my video? is that normal?

I am going to bed now, and as of 2351h pressure was 60psi. Will see how well or not well the car starts tomorrow.

By the way, the tank volumes today reads 12.5L and 24L, and the gauge is still showing less than a quarter...

Peake code 5C and AF unchanged, no new codes.
 

Attachments

·
Registered
Joined
·
23 Posts
Discussion Starter · #13 ·
morning start just now:
fuel pressure last night 60psi
fuel pressure at start ~5psi
fuel pump activated for about 2s or less. Is that normal?
crank x 1 second with not the smoothest start, but better than how it behaved over the last while.

plan next step: disconnect fuel pressure gauge and see if behaviour improves. If not: Schrader valve stem?

 

·
Registered
Joined
·
10,454 Posts
While I was waiting for the pressure decay, I used a boroscope to check injectors. It's not 100%, but there was no gross fuel leak. I understand that it can leak inside, however.
No that is about 10% chance of working at best. First, this is not a diesel the injectors don't go to the combustion chamber they sit above the valves. Leaking fuel will sit on top of the closed valve. The engine should stop in the same position every time you turn it off so really your only chance to see a leak is if the injector that is sitting over the open intake valve is the one leaking.
It is becoming more clear you have a leaking injector with the AF code, but let me look it up because my memory sucks. The way the terms of rich and lean are reported is sort of opposite of what is really happening. There are also two ways for that code to set, one being the max setting has been reached, but the other is a swing over a certain number and that can be the exact opposite.
The only way to look for leaking injectors is to pull the rails and put them on paper towel and run the fuel pump.
There is a simple test I would do at this point, before I jumped down the Rabbit hole completely. I would simply jump the relay for 30s then replace the relay and start the car. If it always starts the car as it should and is always the same I would dig deeper into the fuel system. That is for cold starts. For hot starts I would run the pump longer and would be looking for it to start worse as I lengthened the time I ran the pump. I might start with 10s and see how it starts then 20s the next time 30 and onward. The longer the pump runs the more fuel that is dumped on top of the valve, and the more the car gets flooded.
Your pump times seem normal but it has been a long time since I have even seen my M5 let alone start it. My car is 1000s of Kms away. I have not watched your vids for details just a quick browse so I will later.
 

·
Registered
Joined
·
10,454 Posts
crank x 1 second with not the smoothest start, but better than how it behaved over the last while.
Was the car warm for the second video start? Maybe warm is the wrong word, had it been started before? sort of sounded like a flooded start? What were you trying to show with the OBC in that video? You will only get equal levels if the car has sat for a long time, it takes time to level and the instant you run the fuel pump that returns the fuel to the tank which drives the transfer pump which will put more into the pass side tank. To look at levels after the car has been sitting then don't run the fuel pump. Does that change what you reported?
 

·
Registered
Joined
·
23 Posts
Discussion Starter · #16 ·
Post #13 is a cold start. First thing in the morning, 7 hours after shutdown.

Sent from my SM-N975W using Tapatalk
 

·
Registered
Joined
·
10,454 Posts
Post #13 is a cold start. First thing in the morning, 7 hours after shutdown.
It is very confusing? What is the sound in the background? It ends just before the video? The fuel in the tanks is not how I would expect. The fuel transfer rate is 2.5 L per minute, you only ran the car for 45seconds. With 35 L total that should have been ~ 17.5 L per tank, running it for 60 secs should have brought one to 15 and the other to 20. It is buffered which might explain why it is still dropping after the car was off, unless that noise I hear is the fuel pump being jumped? In that case why was the pressure 0 at the start?
There is also a spike and drop when you engaged the starter, not sure what that is about, and I don't have a car to look at, but my Audi does not do that, it remains the same until fuel is getting consumed.
Fluid can't compress, but air can. I know how an injector works in fluid, but that does not mean it works the same in air. The design of the pintle is to have drag so the fluid passing by the pintle is actually what closes it in between the pulses. Air would have much less drag, and also can be compressed which would change the behavior of the pintle? Like I said confusing?
Sorry to ask again but it is always in the details. The gauge was hooked up 7 hrs before and the fuel pump had not been triggered, a clean virgin start with nothing done prior to the start you taped? To ask again what is that sound? Is it the SAS air pump, sort of sounds like a vacuum cleaner.
 

·
Registered
Joined
·
23 Posts
Discussion Starter · #18 ·
Again, we are talking about post #13, to clarify:
1. noise in background: yes, secondary air pump
2. 7 hours of no intervention.
3. spike and drop: from the first time I tapped in a pressure gauge, it is ALWAYS like that. when the fuel pump engages, it spiked to 70psi, 2 seconds later when pump stops, pressure is down to 60s.
4. tank volumes: Yup, since you explained that letting the car sit would equalize the volume, I have been scratching my head. the car HAD BEEN sitting. and yet when I started again L side is almost empty and R side has most of the volume.
5. looked at the schrader core. Nothing obvious. screwed back tight, but hopefully not too tight. capped. Drove car last night. fuel tank volumes equalizes with drive, and by this morning fuel gauge seemed to recovered to indicate true volue (about 35L) when last week the gauge basically said tank was empty.

This morning's start:
Condition at start--
1. clear a whole bunch of codes: engine AF code cleared, cluster codes, EWS codes, etc etc. Just to give it a clean start again. Any new code should represent current problem, I assume.
2. drove lastnight, car parked, no fuel pressure gauge
3. no intervention in between. no relay jumper
4. 7-8 hours later
5. hard start. cranked 1-2s, stopped crank. recrank for <1s
6. car start
7. hear triple gong / beep from cluster (what was that?)
8. no CEL
9. no Peake code
10. new DIAG code from cluster menu #14: 90 84 035. I bleieve that is INPA code 144 "Clamp 15", and in other posts "Terminal 15". All cluster codes were cleared last night.

Terminal 15? Is that Ignition switch? Can it be the reason for the start trouble?

My hunch is still on fuel pressure, getting closer to diagnosing an injector leak. Will start to look up diagnostic procedures.
 

·
Registered
Joined
·
10,454 Posts
10. new DIAG code from cluster menu #14: 90 84 035. I bleieve that is INPA code
That does not make sense that that is new with such a small number for start countdown. Check the thread(secret menu) maybe it has a low number.
Did not understand, if you read what was hear ignore, sometimes I don't read weel.
clear a whole bunch of codes: engine AF code cleared
Clear the codes, but write them down once they appear. I thought you were doing that already, that is sort of a basic. You can cause codes from testing. You look at your notes and then see if one or more is always coming back. They don't go by themselves.
Keep a frequent check on the levels something is wrong there maybe. That might be what is causing your start problem, but I don't have a clue what could cause that. Go back to the paper clip thread as it says it can test how the levels are working, but I have not tried that so don't know.
Sometimes you have to think outside the box and only you are going to notice a detail.
An example is the old E34 5 series. This was happening to M5s but more research showed it was also happening to 525s. Sort of rough running and other things. In tests the fuel pressure was high. No one could figure it out and by fluke one day I popped the return line and had pressure, should never be pressure there. Digging around and looking deeper the return fuel line was rusting on the inside and crud was building up where the line met the cup in the tank. At low fuel levels cars were running out of gas because the fuel was not returning but they also had very high fuel pressures limited only by what the pump could produce. At high fuel levels where the fuel could over flow the top of the cup they would not run out but ran with high fuel pressure. It was very confusing.
That is an example but it can't happen on these cars because the lines don't rust first off but the system is totally different, just an example of something you would never think of.
7. hear triple gong / beep from cluster (what was that?)
That might be the 1 minute warning before the top of the hour for news or something in euro or maybe the temp warning that that the outside temp is just a bit above freezing. Thought that was one beep but maybe? That could be significant because if the outside temp is wrong the car will have a rough start.
You are going to have to start finishing tests so that things can be ruled out. Lets just go back to the simple, put the jumper wire in the fuel pump for 30 seconds and then pull it and start the car. Does it always start perfect? If it does that will give us a direction to look. Do that for every cold start and doing it a few times before you draw any conclusions. One time does not tell us anything. If it behaves different then we can look somewhere else. Follow a thread if it appears and at least one time the car started perfect when you jumped the pump, lets see if that is consistent.
 

·
Registered
Joined
·
23 Posts
Discussion Starter · #20 ·
That does not make sense that that is new with such a small number for start countdown. Check the thread(secret menu) maybe it has a low number.
maybe it doesnt make sense, but that number went up with another hard start and down with ok start. so that's frequency, and something is not right.

now, 90 is "CLAMP 15" or "terminal 15". is that the ignition switcth at position 2?

Clear the codes, but write them down once they appear.
yes that's what I am doing. I was focused on engine/DME codes and not other module codes, but now I know my INPA is good for the modules as well I cleared everything last night.

Lets just go back to the simple, put the jumper wire in the fuel pump for 30 seconds and then pull it and start the car.
sounds good. I will jump x 30s then start. I will note if it's warm or cold, and time from last shutdown.

Quick question: can I hurt anything if i run pump x30s? If I am thinking straight: if there is an injector leak, I'll flood the cylinder and get a rich code? and if there is no leak, what happens to the pressure in the lines from the fuel pump?

can anyone point me to a good source to diagnose injector leak?

Thanks.
 
1 - 20 of 29 Posts
Top