1996 BMW E36 328i Stalling When Coasting / Stopping
Discussion
My 1996 E36 328i has a strange issue where it will intermittently decide to stall when coming to a stop / coasting. I.e. Clutch pedal depressed. After putting the clutch in the revs will fall gradually to 0 and the car will turn off (Power Steering / Servo off). This is not a hard shut off, but rather like the car doesn't seem to catch idle RPM after coming off load to a stop. The car will re-start immediately on key when this happens.
This issue is completely intermittent and some days you can drive the car with 0 problems what so ever. This will only happen once the car is warm, it does not happen when stone cold.
The car has no loss of power, flat spots and pulls cleanly. If you start the car and do not drive it (no load) the car will idle indefinitely at ~ 650 RPM. It drives flawlessly other than this, seemingly random, stalling issue when coming to a stop / coasting now and again.
The following has been done (In order):
Previous owner had even swapped the DME unit with a known working one and the car was still stalling.
Pulling the dipstick / oil filler cover while running there is some suction but not loads to indicate that the CCV has failed. The crankcase breather hose looks fine with no visible cracking.
Just today I drove it to work with the MAF sensor unplugged and it still stalled out - This proves that the MAF is not the cause...?
Once the car is warm, it will burble on over-run in the exhaust - I'm presuming it's running slightly rich. I desperately need to get this car scanned, I have the 20 pin OBD1 adapter which I tried with a cheap bluetooth scanner (Couldn't connect to the DME). I think I will purchase the cables and try to get it hooked up to INPA.
Has anyone else had this issue before and successfully solved it? I have read almost every thread on this and a lot of them never have a conclusion.
Issue Resolution - Vacuum Leak
I had already suspected the CCV system was leaking and because I had limited history about the part fitted to the car, I decided to replace all items marked in yellow. The main CCV hose, the CCV itself, intake plenum grommet and the return line to the dipstick.
The previous CCV hose I replaced (1) was not leaking (So I thought), but I replaced it anyway. After smoke testing the system, there was a tiny leak coming from the connection point (Red arrow on figure). This was actually from the sealing O-RING on the CCV pipe. Each end has an O-RING inside a channel which seats against the inside diameter of the housing.
This was my vacuum leak and the cause of my problems. I would never have found this if I had not done a smoke test as visually the pipe was seated correctly and no other areas of the intake had cracks or leaks. Get your intake smoke tested! (I used a DIY soldering iron smoke machine).
I would suggest replacing this pipe with a piece of 1/2 inch diameter hosing with jubilee clips on either end as the O-RINGs are a pain to source and difficult to ensure they are seated correctly.
This issue is completely intermittent and some days you can drive the car with 0 problems what so ever. This will only happen once the car is warm, it does not happen when stone cold.
The car has no loss of power, flat spots and pulls cleanly. If you start the car and do not drive it (no load) the car will idle indefinitely at ~ 650 RPM. It drives flawlessly other than this, seemingly random, stalling issue when coming to a stop / coasting now and again.
The following has been done (In order):
- Checked for split hoses / lines - No cracks or splits found. Even sprayed brake cleaner everywhere whilst the engine was running - No change to engine idle RPM
- MAF Cleaned
- Throttle Body Cleaned
- ICV cleaned and checked (Flap is moving freely)
- Crankshaft Position Sensor (OEM BMW)
- Camshaft Position Sensor (OEM BMW)
- Intake Temperature Sensor Cleaned
- CCV, hoses & intake plenum grommet replaced
Previous owner had even swapped the DME unit with a known working one and the car was still stalling.
Pulling the dipstick / oil filler cover while running there is some suction but not loads to indicate that the CCV has failed. The crankcase breather hose looks fine with no visible cracking.
Just today I drove it to work with the MAF sensor unplugged and it still stalled out - This proves that the MAF is not the cause...?
Once the car is warm, it will burble on over-run in the exhaust - I'm presuming it's running slightly rich. I desperately need to get this car scanned, I have the 20 pin OBD1 adapter which I tried with a cheap bluetooth scanner (Couldn't connect to the DME). I think I will purchase the cables and try to get it hooked up to INPA.
Has anyone else had this issue before and successfully solved it? I have read almost every thread on this and a lot of them never have a conclusion.
Issue Resolution - Vacuum Leak
I had already suspected the CCV system was leaking and because I had limited history about the part fitted to the car, I decided to replace all items marked in yellow. The main CCV hose, the CCV itself, intake plenum grommet and the return line to the dipstick.
The previous CCV hose I replaced (1) was not leaking (So I thought), but I replaced it anyway. After smoke testing the system, there was a tiny leak coming from the connection point (Red arrow on figure). This was actually from the sealing O-RING on the CCV pipe. Each end has an O-RING inside a channel which seats against the inside diameter of the housing.
This was my vacuum leak and the cause of my problems. I would never have found this if I had not done a smoke test as visually the pipe was seated correctly and no other areas of the intake had cracks or leaks. Get your intake smoke tested! (I used a DIY soldering iron smoke machine).
I would suggest replacing this pipe with a piece of 1/2 inch diameter hosing with jubilee clips on either end as the O-RINGs are a pain to source and difficult to ensure they are seated correctly.
Edited by Sin Of Greed on Saturday 27th November 16:17
Edited by Sin Of Greed on Sunday 2nd January 14:02
Sin Of Greed said:
My 1996 E36 328i has a strange issue where it will intermittently decide to stall when coming to a stop / coasting. I.e. Clutch pedal depressed. After putting the clutch in the revs will fall gradually to 0 and the car will turn off (Power Steering / Servo off). This is not a hard shut off, but rather like the car doesn't seem to catch idle RPM after coming off load to a stop. The car will re-start immediately on key when this happens.
This issue is completely intermittent and some days you can drive the car with 0 problems what so ever. This will only happen once the car is warm, it does not happen when stone cold.
The car has no loss of power, flat spots and pulls cleanly. If you start the car and do not drive it (no load) the car will idle indefinitely at ~ 650 RPM. It drives flawlessly other than this, seemingly random, stalling issue when coming to a stop / coasting now and again.
The following has been done (In order):
Previous owner had even swapped the DME unit with a known working one and the car was still stalling.
Pulling the dipstick / oil filler cover while running there is some suction but not loads to indicate that the CCV has failed. The crankcase breather hose looks fine with no visible cracking.
Just today I drove it to work with the MAF sensor unplugged and it still stalled out - This proves that the MAF is not the cause...?
Once the car is warm, it will burble on over-run in the exhaust - I'm presuming it's running slightly rich. I desperately need to get this car scanned, I have the 20 pin OBD1 adapter which I tried with a cheap bluetooth scanner (Couldn't connect to the DME). I think I will purchase the cables and try to get it hooked up to INPA.
Has anyone else had this issue before and successfully solved it? I have read almost every thread on this and a lot of them never have a conclusion.
I have the exact same problem on my 1998 328i, forgot about it until a couple of days ago when it nearly caught me out at a roundabout.This issue is completely intermittent and some days you can drive the car with 0 problems what so ever. This will only happen once the car is warm, it does not happen when stone cold.
The car has no loss of power, flat spots and pulls cleanly. If you start the car and do not drive it (no load) the car will idle indefinitely at ~ 650 RPM. It drives flawlessly other than this, seemingly random, stalling issue when coming to a stop / coasting now and again.
The following has been done (In order):
- Checked for split hoses / lines - No cracks or splits found. Even sprayed brake cleaner everywhere whilst the engine was running - No change to engine idle RPM
- Coolant Temperature Sensor Replaced
- MAF Cleaned
- Throttle Body Cleaned
- ICV cleaned and checked (Flap is moving freely)
- Crankshaft Position Sensor (OEM BMW)
- Camshaft Position Sensor (OEM BMW)
Previous owner had even swapped the DME unit with a known working one and the car was still stalling.
Pulling the dipstick / oil filler cover while running there is some suction but not loads to indicate that the CCV has failed. The crankcase breather hose looks fine with no visible cracking.
Just today I drove it to work with the MAF sensor unplugged and it still stalled out - This proves that the MAF is not the cause...?
Once the car is warm, it will burble on over-run in the exhaust - I'm presuming it's running slightly rich. I desperately need to get this car scanned, I have the 20 pin OBD1 adapter which I tried with a cheap bluetooth scanner (Couldn't connect to the DME). I think I will purchase the cables and try to get it hooked up to INPA.
Has anyone else had this issue before and successfully solved it? I have read almost every thread on this and a lot of them never have a conclusion.
For mine it only seems to do it when it is very cold outside which is why I had forgotten about it as not happened for about 6-9 months. I have not tried to diagnose it as I only drive about 5 miles to work really but will be keeping my eye on this for a potential fix if it gets worse.
E36Lover said:
I have the exact same problem on my 1998 328i, forgot about it until a couple of days ago when it nearly caught me out at a roundabout.
For mine it only seems to do it when it is very cold outside which is why I had forgotten about it as not happened for about 6-9 months. I have not tried to diagnose it as I only drive about 5 miles to work really but will be keeping my eye on this for a potential fix if it gets worse.
That's interesting as mine appears to be very similar and is much worse when it is very cold outside as well! I have recently cleaned the Air Intake Temperature Sensor (Just inside the intake manifold on the left hand side when you take off the throttle body). I will be driving it tomorrow to work and will see if it stalls. For mine it only seems to do it when it is very cold outside which is why I had forgotten about it as not happened for about 6-9 months. I have not tried to diagnose it as I only drive about 5 miles to work really but will be keeping my eye on this for a potential fix if it gets worse.
I have also ordered a CCV kit and will be replacing that over the weekend if the weather permits!
I'll post back with my results.
geeks said:
Almost any garage should be able to code read it for you for a small fee otherwise you are just going to keep chucking parts at it. If you are based near Milton Keynes I would be happy to lend you my lead and bluetooth odb reader which work fine on my E36.
Thank you for the offer! 
I'm sadly a bit too far from there (Based in Woking, Surrey). If the the CCV change does not fix it, my next steps will be to try and get the codes read - I've been back and forth with someone that knows someone with a scanner haha - Failing that, I'll get an INPA cable.Locknut said:
I think your problem is the idle speed. 650 RPM is too low. See if you can get it re-set to 800 RPM, that should solve all your problems.
From what I have seen online, a typical idle RPM is 650 - 750 RPM - So mine seems correct. --
Today I replaced the CCV, hoses and the intake plenum grommet. No change to the stalling. (At least the CCV is some preventative maintenance).
Additional - In my original post I had listed that the Coolant Temperature Sensor (13621703993) had been changed. It hasn't. Upon further inspection and re-reading the comments from the previous owner. The readings were "checked" via live data from a scan tool. I'm starting to wonder whether that because it was checked during the warmer months of the year the readings appeared to be correct in relation to the ambient temperature. Now that we have ambient temperatures that are much lower, there may be a greater chance of this sensor reporting dodgy readings to the ECU.
Next Steps
- INPA cable has been ordered and is on the way. Should be with me in a couple of weeks. I can then begin to read fault codes / monitor some live data.
- While I wait, for the sake of £12 - I may purchase a Meyle Coolant Temperature Sensor and fit it next weekend to see if there is any change.
Edited by Sin Of Greed on Saturday 27th November 16:19
Update
Stalled again on the way to work, no surprises there!
On the journey home, there was more traffic than usual - Bumper to bumper, stop and start. The car was idling fairly low, just above the first notch 500 ish, but never stalled. This is not new either, as it only appears to stall if I've sustained a slightly higher RPM for a while (Like 3rd gear 30MPH coming up to a junction).
Car definitely smells like it is running rich as well. These all must be linked, surely. Any more help or feedback would be great.
New coolant temperature sensor has also been ordered. I will be fitting it this weekend. Just for peace of mind, I will also remove and inspect all of the ignition coils for cracks / tears in the boots and any potential arcing signs.
Stalled again on the way to work, no surprises there!
On the journey home, there was more traffic than usual - Bumper to bumper, stop and start. The car was idling fairly low, just above the first notch 500 ish, but never stalled. This is not new either, as it only appears to stall if I've sustained a slightly higher RPM for a while (Like 3rd gear 30MPH coming up to a junction).
Car definitely smells like it is running rich as well. These all must be linked, surely. Any more help or feedback would be great.
New coolant temperature sensor has also been ordered. I will be fitting it this weekend. Just for peace of mind, I will also remove and inspect all of the ignition coils for cracks / tears in the boots and any potential arcing signs.
I strongly suspect the coolant temp sensor isn't to blame here. If the symptoms are as described its air/fuel related, who changed the cam and crank sensors? when where the plugs and coil packs last done? has the exhaust been messed with at all? has the lambda sensor been checked? you can disconnect that as well and it will run to a default map. Ultimately though this is all guess work until you get those codes read, I wouldn't wait for an INPA cable and buggering around with that just get the codes read at a garage and go from there. Given you are in Woking Camberley isn't far, pop and see Will at TWG Automotive, tell him Dan sent you and get the codes read and him to give it a once over, you'll get your answer without throwing the parts bin at it
geeks said:
I strongly suspect the coolant temp sensor isn't to blame here. If the symptoms are as described its air/fuel related, who changed the cam and crank sensors? when where the plugs and coil packs last done? has the exhaust been messed with at all? has the lambda sensor been checked? you can disconnect that as well and it will run to a default map. Ultimately though this is all guess work until you get those codes read, I wouldn't wait for an INPA cable and buggering around with that just get the codes read at a garage and go from there. Given you are in Woking Camberley isn't far, pop and see Will at TWG Automotive, tell him Dan sent you and get the codes read and him to give it a once over, you'll get your answer without throwing the parts bin at it
Cheers, I'll give them a call if there is no progress this weekend. My line of thought on the Coolant Temp Sensor is that it is also responsible for fueling of the engine and this issue seems to be ambient temperature related as it is definitely worse when the outside temperature is lower (Could be a red herring).All done this year starting from around March:
- Cooling system overhaul (Waterpump, thermostat & housing, belt, coolant etc)
- Replaced commonly failing air intake pipework (Boot after MAF, ICV air pipe etc)
- Crank Sensor - Previous Owner (Audi Technician) - OEM BMW Part
- Spark Plugs - NGK BKR6EK - Previous Owner
- Rocker Cover Gasket Replace - Previous Owner
- New Battery - Previous Owner (Yuasa) - Terminals and battery snug
- Exhaust - New catalytic converter fitted in September this year. Issue was happening prior to this being fitted.
- Cleaned throttle body & ICV (Already clean) (ICV Flaps around freely and buzzes when ignition on). Throttle body also snaps shut as it should and has full range of motion
- Cleaned the MAF and driven with MAF disconnected, car still stalled and was noticeably more rough when warm
- Air Intake Temperature Sensor Cleaned - Not tested the resistance or readings
- Cam Sensor - Me - OEM BMW Part
- CCV & Related Hoses - Me
Lambda Sensor - Not checked yet, I'll try running the car with them disconnected. Shall I do both at the same time, or one at a time?
Update
Went out to change the Coolant Temperature Sensor today - I don't have a deep 22mm socket and I cannot seem to find a 22 spanner. Will have to order some additional tools
Will give it another go next weekend.
Ignition Coils - Took all of these out today and inspected the wiring and boots. No cracking in the boots or signs of arcing. All of them are BOSCH units. The spark plug wells are clean with no oil in them.
Lambda Sensors - Got home from a journey (Car fully warm) and switched it off. Unplugged both sensors and restarted the car. Idle appeared to be very slightly more consistent and less jumpy while stationary. I will leave these unplugged for a few journeys and see if there is any improvement with the idle and stalling
Fuel Economy - Run a few calculations on my MPG (Computer reports 25MPG) and I have been averaging around 16 / 17 MPG. My work commute is fairly short and I do mostly do 'town' driving so this may be a contributing factor. Although, the low MPG makes a lot of sense seeing as how rich the car smells when idling / stationary.
Injectors - For the sake of a couple of quid, I've dumped a bottle of fuel system cleaner into the tank to see if there is any difference to be made there.
INPA - INPA Cable should be with me by next week so I can begin scanning and reading some live data, along with resetting the adaptations.
Went out to change the Coolant Temperature Sensor today - I don't have a deep 22mm socket and I cannot seem to find a 22 spanner. Will have to order some additional tools
Will give it another go next weekend.Ignition Coils - Took all of these out today and inspected the wiring and boots. No cracking in the boots or signs of arcing. All of them are BOSCH units. The spark plug wells are clean with no oil in them.
Lambda Sensors - Got home from a journey (Car fully warm) and switched it off. Unplugged both sensors and restarted the car. Idle appeared to be very slightly more consistent and less jumpy while stationary. I will leave these unplugged for a few journeys and see if there is any improvement with the idle and stalling
Fuel Economy - Run a few calculations on my MPG (Computer reports 25MPG) and I have been averaging around 16 / 17 MPG. My work commute is fairly short and I do mostly do 'town' driving so this may be a contributing factor. Although, the low MPG makes a lot of sense seeing as how rich the car smells when idling / stationary.
Injectors - For the sake of a couple of quid, I've dumped a bottle of fuel system cleaner into the tank to see if there is any difference to be made there.
INPA - INPA Cable should be with me by next week so I can begin scanning and reading some live data, along with resetting the adaptations.
Edited by Sin Of Greed on Saturday 4th December 14:52
ian332isport said:
How about the idle control valve?
Idle Control Valve has been cleaned out twice. The flap rattles around very freely inside and it can be heard buzzing when the car's ignition is switched on.I did actually take a video of the valve after I had cleaned it for the second time. Watch / Listen here - https://imgur.com/a/zPOWvqi
E-bmw said:
From what you describe, I would be thinking unmetered air leak, but you say you have covered that.
I guess it's not completely ruled out with 100% certainty as I have not done a full smoke test. However, all the large hoses and vacuum lines around the intake manifold that I can get to have been checked and are ok. Including some of them being replaced with new ones (ICV hose, MAF - Throttle Body boot). I would have also thought that an unmetered air leak would cause the car to run lean as there would be far too much air to fuel in the mixture - This car is definitely running rich (I will be able to confirm that via the fuel trims when I get the diagnostics cable).
I also don't experience the common idle surge / hunting that an air leak would cause.
Update
DCAN cable came today for INPA. Got the drivers and everything setup and successfully connect to the DME.
Some very interesting change of events!
Couple of fault codes, mainly O2 sensors (1 x MAF, 2 x O2 code are 'Interruption' errors - Most likely where I have unplugged them with the car on):
If anyone could help me understand these codes, that would be great.
I have cleared the codes and will wait to see what comes back. I also decided to clear the Adaptations. This is where things get interesting. I started the car afterwards and the cold idle shot up to 2000 RPM and the warm idle is now around 1500RPM. Naturally, this meant the car did not stall whatsoever when I took it out for a drive. Every junction I got to and with the clutch in the idle was around 1250 - 1500RPM. I got the car back home and sat on the driveway for 30 seconds and the idle came back down to 800RPM... I slightly tapped the throttle and then the idle RPM shot back up to 1500RPM and was hunting slightly from 1250 - 1500RPM again.
Maybe this has been the issue all along and the DME had adapted to reduce the idle which was causing the stalling. I'll keep digging.
Any feedback / pointers or advice would be great!
DCAN cable came today for INPA. Got the drivers and everything setup and successfully connect to the DME.
Some very interesting change of events!
Couple of fault codes, mainly O2 sensors (1 x MAF, 2 x O2 code are 'Interruption' errors - Most likely where I have unplugged them with the car on):
203 Lambda controller limit bank 2
Error frequency: 21
Logistic counter: 38
Engine speed 672.00 1 / min
Engine temperature 99.18 degrees C
Probe heating 75.85%
L probe voltage 2 4.84 volts
Short circuit to U-Batt
Exhaust-related error
Error saved after debouncing
Error does not exist at the moment
sporadic error
-------------------------------------------------- -----------
202 Lambda controller limit bank 1
Error frequency: 20
Logistic counter: 38
Engine speed 672.00 1 / min
Engine temperature 99.18 degrees C
Probe heating 67.25%
L-probe voltage 1 4.88 volts
Short circuit to U-Batt
Exhaust-related error
Error saved after debouncing
Error does not exist at the moment
sporadic error
-------------------------------------------------- -----------
8 air mass meters
Error frequency: 7
Logistic counter: 31
Engine speed 736.00 1 / min
Throttle valve angle 0.00 degrees DK
Idle speed actuator 14.84%
LMM voltage value 0.10 volts
Interruption
Exhaust-related error
Error saved after debouncing
Error does not exist at the moment
sporadic error
-------------------------------------------------- -----------
55 Lambda probe heating probe 2
Error frequency: 1
Logistic counter: 37
Motor speed 0.00 1 / min
Load 0.00 mg / stroke
L-probe voltage 2 4.90 volts
Engine temperature 87.97 degrees C
Interruption
Exhaust-related error
Error saved after debouncing
Error does not exist at the moment
sporadic error
-------------------------------------------------- -----------
25 Lambda probe heating probe 1
Error frequency: 1
Logistic counter: 37
Motor speed 0.00 1 / min
Load 0.00 mg / stroke
L probe voltage 1 4.90 volts
Engine temperature 87.97 degrees C
Interruption
Exhaust-related error
Error saved after debouncing
Error does not exist at the moment
sporadic error
============================================================================
If anyone could help me understand these codes, that would be great.
I have cleared the codes and will wait to see what comes back. I also decided to clear the Adaptations. This is where things get interesting. I started the car afterwards and the cold idle shot up to 2000 RPM and the warm idle is now around 1500RPM. Naturally, this meant the car did not stall whatsoever when I took it out for a drive. Every junction I got to and with the clutch in the idle was around 1250 - 1500RPM. I got the car back home and sat on the driveway for 30 seconds and the idle came back down to 800RPM... I slightly tapped the throttle and then the idle RPM shot back up to 1500RPM and was hunting slightly from 1250 - 1500RPM again.
Maybe this has been the issue all along and the DME had adapted to reduce the idle which was causing the stalling. I'll keep digging.
Any feedback / pointers or advice would be great!
Update
Since yesterday, I have driven the car a couple more times. On the last journey the high warm idle has already corrected itself and it's back to stalling when coming to a stop.
No other error codes have been thrown yet. I have taken some snippets of the live data at Warm Idle at after getting home (20 minute drive):
! - INPA had an exclamation mark in front of these values
Is there anything to be gathered from these values?
Since yesterday, I have driven the car a couple more times. On the last journey the high warm idle has already corrected itself and it's back to stalling when coming to a stop.
No other error codes have been thrown yet. I have taken some snippets of the live data at Warm Idle at after getting home (20 minute drive):
! - INPA had an exclamation mark in front of these values
| Sensor | Value |
|---|---|
| Battery Voltage | 13.74V |
| Engine Speed | 731 RPM (Normally Lower) |
| Intake Air Temp | 47.06 Degrees C |
| Coolant Temp | 80.55 Degrees C |
| TP Position | 0.7 Volts (4.6 Volts @ WOT) |
| Engine Load | !97.57 (Mg/Stroke) |
| Air Use - Mass Air Flow | 12.00 (Kg/h) |
| Mass Air Flow Voltage | 0.64V |
| Air Use - Idle Control Valve | !12.00 (Kg/h) |
| Idle Speed Integrator | -6.76% |
| Camshaft Position | 18.73 Degrees KW |
| Oxygen Sensor Heater 1 | 23.44% |
| Oxygen Sensor Heater 2 | 23.44% |
| Oxygen Sensor Voltage 1 | 4.83V |
| Oxygen Sensor Voltage 2 | 4.84V |
| Knock Sensor Signal 1 | 0.12V |
| Knock Sensor Signal 2 | 0.10V |
Is there anything to be gathered from these values?
Edited by Sin Of Greed on Sunday 12th December 13:08
Update
These error codes have returned in INPA:
My understanding of these codes is that both Oxygen sensors are reporting that they are at lean limit. Today I have spent an hour or so thoroughly smoke testing the intake system (DIY Smoke tester with a soldering iron and mineral oil). There was a slight leak at the new CCV crankcase hose where it connects into the unit. The new ORINGs supplied on the new hose were too big, so I swapped over the old ORINGS - These are leaking. I will replace these ORINGS with new ones to see if there is any difference.
There were 0 other leaks which has stumped me a little, I genuinely thought I must have had an inconspicuous leak somewhere.
Next Steps
Fuel pressure test - I'll be buying a gauge and hooking it up to the rail to test the pressure. Potentially the Fuel Pressure Regulator is on its way out.
These error codes have returned in INPA:
202 Lambda controller limit bank 1
Error frequency: 1
Logistic counter: 40
Engine speed 704.00 rpm
Engine temperature 87.97 degrees C
Probe heating 68.82%
L-probe voltage 1 4.86 volts
Short circuit to U-Batt
Exhaust-related error
Error saved after debouncing
Error present at the moment
static error
-------------------------------------------------- -----------
203 Lambda controller limit bank 2
Error frequency: 1
Logistic counter: 40
Engine speed 704.00 rpm
Engine temperature 87.97 degrees C
Probe heating 75.85%
L probe voltage 2 4.82 volts
Short circuit to U-Batt
Exhaust-related error
Error saved after debouncing
Error present at the moment
static error
============================================================================ ===========
My understanding of these codes is that both Oxygen sensors are reporting that they are at lean limit. Today I have spent an hour or so thoroughly smoke testing the intake system (DIY Smoke tester with a soldering iron and mineral oil). There was a slight leak at the new CCV crankcase hose where it connects into the unit. The new ORINGs supplied on the new hose were too big, so I swapped over the old ORINGS - These are leaking. I will replace these ORINGS with new ones to see if there is any difference.
There were 0 other leaks which has stumped me a little, I genuinely thought I must have had an inconspicuous leak somewhere.
Next Steps
Fuel pressure test - I'll be buying a gauge and hooking it up to the rail to test the pressure. Potentially the Fuel Pressure Regulator is on its way out.
Happy new year all!
I've been waiting a while to post this as I did not want to tempt fate on the issue. After a few weeks of driving the car around, my issue is resolved.
It turns out the apparently "small" vacuum leak I had previously mentioned above, from the CCV, was the indeed the issue. I guess it wasn't such a small leak with the car running due to the increased pressure
The rest of this post is going to be quite long as I want to ensure that anyone else that comes across this issue in the future has a good chance of solving it. Most of the threads I have read in the past are extremely inconclusive and devoid of useful information.
My Leak
I had already suspected the CCV system was leaking and because I had limited history about the part fitted to the car, I decided to replace all items marked in yellow. The main CCV hose, the CCV itself, intake plenum grommet and the return line to the dipstick.
The previous CCV hose I replaced (1) was not leaking (So I thought), but I replaced it anyway. After smoke testing the system, there was a tiny leak coming from the connection point (Red arrow on figure). This was actually from the sealing O-RING on the CCV pipe. Each end has an O-RING inside a channel which seats against the inside diameter of the housing.
This was my vacuum leak and the cause of my problems. I would never have found this if I had not done a smoke test as visually the pipe was seated correctly and no other areas of the intake had cracks or leaks. Get your intake smoke tested! (I used a DIY soldering iron smoke machine).
I would suggest replacing this pipe with a piece of 1/2 inch diameter hosing with jubilee clips on either end as the O-RINGs are a pain to source and difficult to ensure they are seated correctly.
Engine Warm (Closed Loop - Computer starts reading sensors)
When the engine is cold the DME is in Open Loop mode and ignores a lot of sensors. It fuels the engine from the pre-defined tables stored within the DME. As the engine temperature starts to rise the Coolant Temperature Sensor sends a signal to the DME to say that the engine is warm and the DME enters 'Closed Loop' and begins to receive information from the remaining sensors. It's during this stage that the O2 sensor sends signals to the DME that there is unburnt oxygen within the exhaust - Lean Mixture. To compensate, the DME increases the Pulse Width of the injectors to allow them to spray more fuel into the combustion chamber (Black Spark Plugs).
The DME commanding the Pulse Width increase in the short term creates the positive STFT (Short Term Fuel Trim - Adaptive in INPA), after some time; the DME instead takes the average STFT value and turns it into a LTFT (Long Term Fuel Trim - Multiplicative in INPA). This LTFT then becomes the new datum / baseline figure that the STFT begins to work from. This is why it is important to reset the adaptations after troubleshooting or implementing fixes.
Worse When Ambient Temperature Is Low
I first thought this was a coincidence as I believed I had a knackered sensor. When realising it was a vacuum leak, it made sense. When cold, air is more dense as there is less energy for the oxygen particles to disperse. This means more oxygen per cubic meter of air (Roughly 21% of the molecules per unit of volume). More oxygen in the engine for the same volume of air that is unmonitored - Lean condition is exacerbated (Fuel Deficit)
Example:
Air at 1ATM (Standard Atmosphere) and 5 degrees Celsius weighs 1.268g per cubic meter
Air at 1ATM and 25 degrees Celsius weighs 1.184g per cubic meter
Thus the air at 25 degrees contains ~7% less Oxygen than the air at 5 degrees.
Issue Only Pronounced At Idle
The volume of air entering the engine at idle is far lower (Controlled By IACV) than when on throttle. When throttle is applied, the butterfly valve opens and allows a much greater volume of air into the intake manifold for combustion. During this process, the additional air/oxygen from the leak is non-apparent because it becomes a tiny proportion of the air available to the engine for combustion.
Resetting Adaptations Caused High Idle
Theory - Stepper motor inside the IACV gets reset to a default position (I believe there are 150 steps). Let's say it resets to being wide open at 150 steps - The DME knows that when the IACV is wide open, the volume of air should nearly match the flow rate past the MAF. When it then sees the O2 sensors complaining of a lean condition it begins to close the IACV to reduce the volume of air at idle. This gets worse and worse over driving cycles.
Fuel Economy
My fuel economy had nose dived terrifically. I was averaging about 14 MPG and would frequently hear pops and bangs from the exhaust during over-run. This again was a pointer towards it being a vacuum leak. The additional fueling caused in the short term by the oxygen sensors was destroying my fuel economy.
Diagnosis
I, including many others, would dismiss a vacuum leak after checking all the hoses and intake system after testing with carb cleaner and removing to inspect hoses. But the car was telling me otherwise, and it was correct. The biggest take away I have from this, is that smoke testing is an invaluable diagnostic tool.
INPA reported 2 codes: 202, 203 - These are for the oxygen sensors maxing out the short term fuel trims. For me, this was the confirmation of a vacuum leak that led me to perform a smoke test.
This was easy to confirm with live diagnostics - The STFT was increasing until I increased the RPM, once doing so, the STFT began to decrease, hence, indicating a vacuum leak. If the STFT does not decrease whilst increasing the RPM, it is either the fuel system (Fuel Pressure Regulator or Pump) or the MAF sensor.
This is a list of things I had checked and / or read up on trying to diagnose this issue and it may be valuable to someone else:
I've been waiting a while to post this as I did not want to tempt fate on the issue. After a few weeks of driving the car around, my issue is resolved.
It turns out the apparently "small" vacuum leak I had previously mentioned above, from the CCV, was the indeed the issue. I guess it wasn't such a small leak with the car running due to the increased pressure
The rest of this post is going to be quite long as I want to ensure that anyone else that comes across this issue in the future has a good chance of solving it. Most of the threads I have read in the past are extremely inconclusive and devoid of useful information.
My Leak
I had already suspected the CCV system was leaking and because I had limited history about the part fitted to the car, I decided to replace all items marked in yellow. The main CCV hose, the CCV itself, intake plenum grommet and the return line to the dipstick.
The previous CCV hose I replaced (1) was not leaking (So I thought), but I replaced it anyway. After smoke testing the system, there was a tiny leak coming from the connection point (Red arrow on figure). This was actually from the sealing O-RING on the CCV pipe. Each end has an O-RING inside a channel which seats against the inside diameter of the housing.
This was my vacuum leak and the cause of my problems. I would never have found this if I had not done a smoke test as visually the pipe was seated correctly and no other areas of the intake had cracks or leaks. Get your intake smoke tested! (I used a DIY soldering iron smoke machine).
I would suggest replacing this pipe with a piece of 1/2 inch diameter hosing with jubilee clips on either end as the O-RINGs are a pain to source and difficult to ensure they are seated correctly.
Symptoms Experienced
Engine Warm (Closed Loop - Computer starts reading sensors)
When the engine is cold the DME is in Open Loop mode and ignores a lot of sensors. It fuels the engine from the pre-defined tables stored within the DME. As the engine temperature starts to rise the Coolant Temperature Sensor sends a signal to the DME to say that the engine is warm and the DME enters 'Closed Loop' and begins to receive information from the remaining sensors. It's during this stage that the O2 sensor sends signals to the DME that there is unburnt oxygen within the exhaust - Lean Mixture. To compensate, the DME increases the Pulse Width of the injectors to allow them to spray more fuel into the combustion chamber (Black Spark Plugs).
The DME commanding the Pulse Width increase in the short term creates the positive STFT (Short Term Fuel Trim - Adaptive in INPA), after some time; the DME instead takes the average STFT value and turns it into a LTFT (Long Term Fuel Trim - Multiplicative in INPA). This LTFT then becomes the new datum / baseline figure that the STFT begins to work from. This is why it is important to reset the adaptations after troubleshooting or implementing fixes.
Worse When Ambient Temperature Is Low
I first thought this was a coincidence as I believed I had a knackered sensor. When realising it was a vacuum leak, it made sense. When cold, air is more dense as there is less energy for the oxygen particles to disperse. This means more oxygen per cubic meter of air (Roughly 21% of the molecules per unit of volume). More oxygen in the engine for the same volume of air that is unmonitored - Lean condition is exacerbated (Fuel Deficit)
Example:
Air at 1ATM (Standard Atmosphere) and 5 degrees Celsius weighs 1.268g per cubic meter
Air at 1ATM and 25 degrees Celsius weighs 1.184g per cubic meter
Thus the air at 25 degrees contains ~7% less Oxygen than the air at 5 degrees.
Issue Only Pronounced At Idle
The volume of air entering the engine at idle is far lower (Controlled By IACV) than when on throttle. When throttle is applied, the butterfly valve opens and allows a much greater volume of air into the intake manifold for combustion. During this process, the additional air/oxygen from the leak is non-apparent because it becomes a tiny proportion of the air available to the engine for combustion.
Resetting Adaptations Caused High Idle
Theory - Stepper motor inside the IACV gets reset to a default position (I believe there are 150 steps). Let's say it resets to being wide open at 150 steps - The DME knows that when the IACV is wide open, the volume of air should nearly match the flow rate past the MAF. When it then sees the O2 sensors complaining of a lean condition it begins to close the IACV to reduce the volume of air at idle. This gets worse and worse over driving cycles.
Fuel Economy
My fuel economy had nose dived terrifically. I was averaging about 14 MPG and would frequently hear pops and bangs from the exhaust during over-run. This again was a pointer towards it being a vacuum leak. The additional fueling caused in the short term by the oxygen sensors was destroying my fuel economy.
Diagnosis
I, including many others, would dismiss a vacuum leak after checking all the hoses and intake system after testing with carb cleaner and removing to inspect hoses. But the car was telling me otherwise, and it was correct. The biggest take away I have from this, is that smoke testing is an invaluable diagnostic tool.
INPA reported 2 codes: 202, 203 - These are for the oxygen sensors maxing out the short term fuel trims. For me, this was the confirmation of a vacuum leak that led me to perform a smoke test.
This was easy to confirm with live diagnostics - The STFT was increasing until I increased the RPM, once doing so, the STFT began to decrease, hence, indicating a vacuum leak. If the STFT does not decrease whilst increasing the RPM, it is either the fuel system (Fuel Pressure Regulator or Pump) or the MAF sensor.
Other Things To Check If You Are Experiencing Intermittent Stalling
This is a list of things I had checked and / or read up on trying to diagnose this issue and it may be valuable to someone else:
- MAF Sensor - If the car performs better with the MAF sensor disconnected then it could be faulty or you have a vacuum leak. Use INPA / Diagnostic tool to measure the grams/s being reported by the MAF. If while driving it is not linearly increasing, the MAF is faulty. Do not buy a cheap aftermarket MAF for these cars. BMW part only.
- Throttle Body - Clean the throttle body and check for movement in the butterfly plate. Also check the throttle cable and ensure that it is returning to the fully closed position whilst the accelerator pedal is not in use. If there is a gap, this will induce a vacuum leak.
- Idle Control Valve - Remove and clean. It should sound like this (https://imgur.com/a/zPOWvqi). Test it is getting power and is operating. With the car switched off, put your hand on the ICV, get someone to switch on the accessory power (Key Position 2), you should be able to feel and hear buzzing.
- Camshaft & Crankshaft Position Sensors - Camshaft position sensor is also known to cause intermittent stalling, typically a tell tale symptom would be batch firing of the injectors. Crankshaft would be the car switches off (hard starting) and then after a while (allowing the car to cool down) it will let you restart the car. This is due to the sensor getting heat soaked and dying.
- Throttle Position Sensor - I've read about a lot of people replacing these to no effect. I think they rarely fail as they are a simple piece of equipment. You can check the voltage readings in INPA. At idle ~ 0.7 Volts, WOT ~ 4.6 Volts is good.
- Coolant Temperature Sensor - This can cause havoc with the closed / open loop states of the car. Can be checked in INPA, ensure that it is linearly increases as the car warms up and the readings are logical. This can cause stalling if faulty.
- Air Intake Temperature Sensor - Same as above.
Edited by Sin Of Greed on Sunday 2nd January 14:02
Edited by Sin Of Greed on Sunday 2nd January 14:03
Gassing Station | BMW General | Top of Page | What's New | My Stuff