BMW E46 Fanatics Forum banner

1 - 20 of 32 Posts

·
Registered
Joined
·
151 Posts
Discussion Starter #1
Hi Guys,

I'm trying to debug a cold hard start situation with my 325Ci first from the fuel delivery perspective as I have absolutely zero codes.

Does anyone have data on how the residual fuel pressure at the rail should be behaving after certain amounts of time?
I have a strong 50 PSI as soon as the fuel pump is primed, then after 40 minutes it's dropping to 45 PSI. Now I know these meet the specs. However after two days of standing the residual fuel pressure drops to 8-10 PSI. Is such a big drop over a long period normal, I haven't found any documentation on this? (my fuel pump and filter are practically new: VDO and Mahle) Then with a single prime on the fuel pump it's back at 48-50 PSI. With the engine cranking or running at idle speed I have a steady 52 PSI.

I then went on and pulled the fuel rail to check the injectors.Mines are Siemens 1427240, the measured resistance on the coils are between 15-16 Ohm all around. I then tested the injectors one-by-one and they all spray strong and the same way. Finally I re-pressured the rail and kept checking the injectors for leak. After 1 hour the residual fuel pressure is still at 45 PSI with no visible drops. I'm now waiting for the long-term leak test results, but I doubt some of them just begin to leak after a given time.

Thanks for the input.
 

·
Registered
Joined
·
10,232 Posts
I doubt your cold start issue is fuel pressure related as you you get 50psi as soon as the pump primes. Please describe the cold start issue.
 

·
Registered
E30M3 Race F10 535 R1150Rt M Coupe
Joined
·
1,531 Posts
I don't have the spec's in front of me now. However the numbers you cite seem very respectable.
Having the fuel pump energize (the way VW did it for years) when you turn the key solved most cold start issues related to fuel pressure.

You might try some liqui moly fuel injector cleaner. Injectors behave slightly different (drip/leakwise) hot and installed in a cyl head vs: cool and sitting on top of a piece of cardboard. IOW the begginings of injector leakdown usually don't show up on paper towels or cardboard.

If that's what you're chasing, hook up the gauge, idle engine until fully heat soaked (thermostat has opened at least twice) then shut off.
Monitor the gauge. Likely you'll see the pressure climb after a while, owing to the heat increasing the fuel expansion in the rail. Then as the engine cools it returns to normal.
Trick is to try to determine if/when that increased pressure is happening, does that make an injector or two drip?

Watch the gauge and see if you observe negative spiking, indicating dripping.

After all that it's just easier to proactively treat the fuel once in a while and never have to think about it.
 

·
Registered
Joined
·
151 Posts
Discussion Starter #4 (Edited)
Thank you guys for the suggestions so far. I will definitely give this Liqui Moly injector cleaner a try.

So the background on my problem is the following:
One day the car starts fine cold/warm, next day it begins to start hard when cold, which means that if the car has been off for a night, then it does not start on the first or second attempt. It cranks and I see the tacho go up to 500-700 RPM, then as soon as I release the starter it stalls. (seeing the tacho go up makes me think that it might not be related to the crank position sensor) On the third attempt it fires up and runs for a couple seconds, then it usually stalls. From the fourth attempt it stays running and it may or may not run a bit rough for the first 10-15 seconds. From that point the car may be restarted without problem, even if you shut it off for an hour.
When it finally starts running, the exhaust smells gas for the first minute or so.

So this is the background. No codes reading with INPA. Battery and fuel pump have been swapped with good used ones, no change. Then I checked the fuel delivery system for proper pressures, I began to think that I might have one or more leaking injectors that create a flood situation and therefore preventing some cylinders to fire. I have white paper towels under the pressurized injectors now and will check and report back if there are leaks. So my question: if the injectors don't leak and my check valve functions well, then the residual fuel pressure should not drop to <10 PSI, right? I don't seem to have an external leak. This absolute leakdown bogs my mind.

Anyway some additional data:
  • Upstream O2s have been replaced within 3000 km with BOSCH direct fitments. Banks go closed loop pretty soon. Voltage graphs look ok for both banks.
  • Spark plugs were also replaced within 3000 km with NGKs.
  • Coils are original Berus from 2001
  • MAF is Siemens also from 2001
  • Air filter practically new MANN
  • Throttle body and ICV have been cleaned, ICV moves freely
  • LTFT for both banks 0%, STFT ~ minus 1-3% during idle -> slight injector leakage?
I also pulled my coils and spark plugs. The boots are intact and the plugs visually look good as well to me. I plan to buy a spark checker next week and see if my coils fire on the first start attempts.
 

·
Registered
Joined
·
151 Posts
Discussion Starter #5
Update after 24 hours:
Residual pressure is down at 17 PSI and injector 6 is definitely leaking. Others are solid.

Two questions:

Is it realistic that one flooded cylinder causes a stall on cold start? I mean can the crankshaft lose that much of momentum or one flooded intake path can poison the neighbours as well?

Can anything be done with the leak or it needs to be replaced?

Thanks.
 

·
Registered
Joined
·
10,232 Posts
Could be. Your crankshaft position sensor is fine. If the tach needle didn’t move during starting then I’d suspect this sensor. As far as the bad injector, try the LM cleaner and see what happens. Btw, injector cleaners are hard in the oil so after you’ve run the 2 bottles through the system and you’ve run through a couple tanks of gas only, change the oil and filter.

Now, suppose the injector still leaks,you can replace it with a new identical injector or you can find a reputable injector cleaning company and have all 6 injectors cleaned ultrasonically. What brands of gas/petrol do you buy? I ask as some brands put proper injector cleaner in the fuel. In the western hemisphere we have top tier gas. These are brands that sell gas that meets the cleaning standards established by BMW and other major car makers. Go to www.toptiergas.com to learn more.
 

·
Registered
Joined
·
151 Posts
Discussion Starter #8
Could be. Your crankshaft position sensor is fine. If the tach needle didn’t move during starting then I’d suspect this sensor. As far as the bad injector, try the LM cleaner and see what happens. Btw, injector cleaners are hard in the oil so after you’ve run the 2 bottles through the system and you’ve run through a couple tanks of gas only, change the oil and filter.

Now, suppose the injector still leaks,you can replace it with a new identical injector or you can find a reputable injector cleaning company and have all 6 injectors cleaned ultrasonically. What brands of gas/petrol do you buy? I ask as some brands put proper injector cleaner in the fuel. In the western hemisphere we have top tier gas. These are brands that sell gas that meets the cleaning standards established by BMW and other major car makers. Go to www.toptiergas.com to learn more.
Because of the aftermath of the pandemic situation I'm working long term from home, so this car will be down for some time, I don't see burning 140 liter of petrol anytime soon.I will go down the replacement path this time, but will keep this LM fuel injector cleaner in relevance and run it yearly.

I'm using the best petrol money can buy here in Europe, which is E5 RON 100+ with plenty additives. As I mentioned earlier the spray pattern is a nice V shape on all 6 injectors, so I think they should be in decent shape. Now regarding the leak on number six, does an ultrasonic cleaning make sense in this case? I mean a leak from the spray nozzle isn't a result of a wear down?

An injector leaking externally, wet paper towel, doesn’t mean you’re getting extra fuel in the intake. Usually means pinched or worn o-rings.
Seems you have good fuel pressure at cold start.

You reported 0% LTFTs, but what are they at hot idle?
Aren't the o-rings there to air tightly secure the injectors into the manifold and onto the rail? I mean in my case the fuel is definitely leaking out of the spray nozzle.

My LTFTs are 0% for both banks hot or cold. My STFTs seem to be slightly negative at hot idle, like a couple %.
 

·
Registered
Joined
·
10,232 Posts
A good injector shop will first inspect the injector for external leaks. It’ll turn ultrasonically clean the injector. Lastly, it will test the injector for flow pattern and volume. In reality you should have all six injectors done at the same time as the shop will ensure that all injectors are sending the same volume of fuel. It’s called injector balancing or matching.

The upper oring provides a tight fit with the fuel rail. The lower oring ensure an tight air seal.
 

·
Registered
Joined
·
151 Posts
Discussion Starter #10
Ok, thanks for the answer. I will have to decide now whether I should be sending all six for inspection and cleaning or getting a set of used ones that look promising by the pictures.
 

·
Registered
Joined
·
151 Posts
Discussion Starter #11
Update on my slow progress to debug this:
I had contradicting results with my second injector leak test, because none of them seemed to leak onto the paper that time, although the paper was literally touching the tip of the injectors. That's very weird, because number 6 showed a clear sign. For my second test the outlet of the fuel filter was pinched, still I was down at 20 PSI after a day.
I have just set up a third test to check once again the injectors for leak. This time I did not pinch the fuel line.

In the meantime I pulled the spark plugs and all of them seemed to be somewhat fouled and the threads are oily: picture, picture 2, but it's definitely not coming from the valve cover gasket as I replaced that not long ago and the spark holes are clean. I then went on and started investigating the CCV system. I pulled the pipe that goes from the CCV back into the distributor that's mounted on top of the intake manifold. Clean oil was literally dripping out of it as I detached it. Now I guess this is for crankcase gases only, absolutely now oil should be present here, I mean in liquid form at least. I then tore down the whole CCV system, the hoses are all OK. The CCV is still the original one from 2001, but it's not clogged up and looks normal on the inside. On the other hand the diaphragm inside is dead with two small tears around the edge. I guess that's normal for a 19 year old diaphagm, my DISA diaphragm was in similar condition at the time I rebuilt it.

So my next plan is (besides waiting for some input from you):
1) get a new CCV and check the results from the third injector leak test
2) since I'm basically a couple bolts away from pulling the whole intake, I might do that as well and have a look underneath
 

·
Registered
Joined
·
151 Posts
Discussion Starter #12
Now that I've been thinking about the situation, I'm pretty positive that this whole hard cold start/cold stalling thing started happening kinda right after I replaced my CCV drain tube because it was broken open at the angled part, which caused some intermittent lean codes. Could it be that by replacing that tube I solved a vacuum leak, however because of the ruptured diaphragm inside the CCV I created a suction at the drain tube, which as a result started picking up oil from the pan and introduce it into the intake manifold? Now I don't know how engine oil exactly impacts combustion, but given that cold start is a pretty delicate state, it seems to be worth checking out.

Also before I removed everything, I conducted a CCV suction test at the oil filler cap. At idle I had some significant suction and by removing the cap the idle changed. The engine began to stumble with tacho needle slightly moving up and down, but did not stall.
 

·
Registered
Joined
·
2,347 Posts
With the intake manifold off I would consider an OFHG replacement as well as intake manifold gasket. Hard coolant pipes also. (If not done recently)


Sent from my iPhone using E46Fanatics
 

·
Registered
Joined
·
151 Posts
Discussion Starter #14
OFHG has been recently replaced. I don't seem to have any leak anywhere at the moment, which is kinda cool. I decided to leave the intake on for now and concentrate on the hard start issue.

So my update:
I've thoroughly cleaned my spark plugs and they went back. I also bought a brand new CCV, which is ready to be installed. I inspected the CCV pipes/hoses and they are all fine, so I'm reusing those. Regarding the injectors, a tiny (smaller than the tip of the injector) bit of yellowish stain is under 2 and 6 after one day. I hardly doubt that would create a flood situation. I repressurized the fuel rail and give it another day as I don't have too much time today to put everything back together. I plan to retest the injectors for proper spraying, but other than that my hope is in the CCV and the oil in the distributor that fouls the cold start. It still refuses to give me a code after dozens of failed cold starts.
 

·
Registered
Joined
·
151 Posts
Discussion Starter #15
Update:
Could not get the injectors to leak and they also spray nicely. So I just installed a new CCV and put everything back together. No change in cold start and still no codes in DME. The car restarts fine when warm or hot. After a couple hours the start is getting gradually worse. I created a video for cold start after 12 hours, 3rd attempt is with accelerator. Also logged this with OBD Fusion and included a log for hot start as well: video, log
Could someone look into these? I can only see the MAF start to report relatively late, like 3-4 seconds after the log shows RPM, but the could be normal. Thanks.
 

·
Registered
Joined
·
151 Posts
Discussion Starter #16
Also here's a video on fuel pressure figures during cold startup with no accelerator.
1st: dies instantly without revs going beyond 700
2nd: similarly as before, maybe RPM did go a bit higher
3rd: fires up but RPM does not go over 1000, which is weird, after a couple seconds it dies
4th: fires up and maintains stable idle

All in all even when the car start on its own, there is not really a raised idle RPM, barely go over 1000.
 

·
Registered
Joined
·
151 Posts
Discussion Starter #18
I sorta tested it. It wasn't a real first start in the morning as I was after a couple failed starts by then, but I unplugged it and it similarly died on the first start after that, then stayed running on its own on the second with a lumpy idle, EML and a fault code for MAF. It wasn't a conclusive evidence for me, so I considered it working and continued with other stuff.
 

·
Registered
Joined
·
38 Posts
Looking at your log file it seems the processes to be out of sync so it's hard to tell anything definitive. Might be because response is too slow while logging so many processes simultainously.

LTFT's seem to be some default value all the time. That's because the tool you're using can't read additive adaptive fuel trim value which is what BMW uses to control idle fuel trims. STFT's seem to cycle quite OK around this value. But it's impossible to tell whether this value is on the rich or lean side. You should connect Inpa and check what your additive values are to get proper picture of your idle fuel trims.
 

·
Registered
Joined
·
2,347 Posts
LTFT's seem to be some default value all the time. That's because the tool you're using can't read additive adaptive fuel trim value which is what BMW uses to control idle fuel trims. STFT's seem to cycle quite OK around this value. But it's impossible to tell whether this value is on the rich or lean side. You should connect Inpa and check what your additive values are to get proper picture of your idle fuel trims.
Don’t the LTFTs report zero until all is warmed up and in closed loop?


Sent from my iPhone using E46Fanatics mobile app
 
1 - 20 of 32 Posts
Top