It’s the pre-cat sensors that influence the mixture but even then only under light and steady state throttle. Under full throttle the lambda is ignored and mixture is calculated from the map (i.e injector duration is matched to airflow meter signal). So at full throttle you are essentially running open loop. Steady state throttle and idle is closed loop. If you disconnect the afm and run full throttle it must use some default value not part of the mapping.
UPDATED::: Z4M S54 Engine Hesitation - POLL
- Thread starter bertman
- Start date
Buzzard88 said:
My understanding is that when the MAF is disconnected the ecu ignores all the adaptions/fuel trims. I think its a bad sensor that has pulled the adaptions to a point where the hesitation exists. Resetting the adaptions normally fixes the problem for a while until the adaptions have moved to the point that its a problem again. That in turn makes it difficult to identify the cause of the problem in the first place because everytime you change something you need to clear that adaptions which in turn make the problem go away for a while.
Ok thanks to all who have responded. Your time is appreciated and valued to me.
Just confirming my only problem seems to be during idle. If the car is idling normal then if I stab the throttle quickly then it's so unresponsive and almost seems like or is a misfire.
If this happens when I am stationary and releasing the clutch then the car nearly stalls and definitely would sometimes if I were not managing it with my two feet. What it basically means is that I have to bring the revs on excessively early, so it has it's little hissy fit, prior to releasing the clutch. After this and at higher revs the car is super awesome as it should be!
This also makes it such a lucky dip when trying to match the revs when downshifting.
I can have the car parked in the driveway with the engine running and in neutral, stab the throttle and burrrr (misfire or whatever it is). Then I can immediately disconnect the MAF sensor and it responds nicely and sharply to the throttle stab with no sign of burrr. While typing all this again my brain is saying ICV once again but I gave that a thorough clean and it wasn't too dirty at all. Perhaps I should try a new ICV. Soon I'll have enough parts to build a new car!
I am not totally convinced this is a misfire though. I actually think it is just a really bad fuel / air preparation cycle.
So my understanding is that the DME is operating in closed loop when my problem occurring.
Just confirming my only problem seems to be during idle. If the car is idling normal then if I stab the throttle quickly then it's so unresponsive and almost seems like or is a misfire.
If this happens when I am stationary and releasing the clutch then the car nearly stalls and definitely would sometimes if I were not managing it with my two feet. What it basically means is that I have to bring the revs on excessively early, so it has it's little hissy fit, prior to releasing the clutch. After this and at higher revs the car is super awesome as it should be!
This also makes it such a lucky dip when trying to match the revs when downshifting.
I can have the car parked in the driveway with the engine running and in neutral, stab the throttle and burrrr (misfire or whatever it is). Then I can immediately disconnect the MAF sensor and it responds nicely and sharply to the throttle stab with no sign of burrr. While typing all this again my brain is saying ICV once again but I gave that a thorough clean and it wasn't too dirty at all. Perhaps I should try a new ICV. Soon I'll have enough parts to build a new car!
I am not totally convinced this is a misfire though. I actually think it is just a really bad fuel / air preparation cycle.
So my understanding is that the DME is operating in closed loop when my problem occurring.
Buzzard88 said:
After looking for the post cat sensors I couldn't even find the cats. Now I feel pretty stupid as Australian cars are the Euro spec which apparently don't have cats and therefore no sensors. So if anybody needs a new pair let me know.
Back to the drawing board on this one..
Euro cars just have the main set of catalytic converters. US cars have an extra set on the exhaust manifolds.
Euro: http://www.realoem.com/bmw/enUS/partgrp?id=DU92-EUR-10-2006-E86-BMW-Z4_M32&mg=18
US: http://www.realoem.com/bmw/enUS/partgrp?id=DU93-USA-08-2006-E86-BMW-Z4_M32&mg=18
ga41 said:Buzzard88 said:
Back to the drawing board on this one..
Euro cars just have the main set of catalytic converters. US cars have an extra set on the exhaust manifolds.
Euro: http://www.realoem.com/bmw/enUS/partgrp?id=DU92-EUR-10-2006-E86-BMW-Z4_M32&mg=18
US: http://www.realoem.com/bmw/enUS/partgrp?id=DU93-USA-08-2006-E86-BMW-Z4_M32&mg=18
Thanks ga41. I was just editing my post to correct it but you beat me to it. Going out to do them now.
Buzzard88 said:
Yes, what you are essentially doing by stabbing the throttle from idle is transitioning from closed loop control (a/f ratio controlled by lambda sensors) to open loop control (a/f ratio controlled by ECU map against air flow meter signal). I'm just wondering that if by disconnecting the air flow meter the ECU never runs in the open loop state (because it's has no air flow information to allow it to do so), so you aren't going from one state to the other. I'm betting it stays in closed loop control all the time with the AFM disconnected.
Muckinonthesofa
Member
ph001 said:
Interesting. Excuse my ignorance but is it safe to run without the AFM connected?
When you disconnect the MAF the car reverts to the Alpha-N map, which I believe estimates air volume instead using the engine speed and TPS. Do some Googling on it but IMO disconnecting the MAF doesn't prove anything conclusive as it doesn't really help you narrow down if a sensor is giving you an issue as the mapping has changed.
Have you tried INPA / DIS yet, see if you can run some tests and check for error codes. I fear your going to end up replacing a lot of expensive parts for no reason. For example, the ICV replacement is pointless, it really just works or doesn't it's a simple valve that opens and closes as you would have seen. Cleaning it is good but I wouldn't waste money replacing it.
It's safe to run without the MAF connected but absolutely no reason to do so long term unless you've specifically had Alpha N tuned for your modifications, such as running a CSL airbox where you have no MAF for example.
Oh, INPA would be useful for checking your long term trim's/adaptations, see if there's anything untoward there
Have you tried INPA / DIS yet, see if you can run some tests and check for error codes. I fear your going to end up replacing a lot of expensive parts for no reason. For example, the ICV replacement is pointless, it really just works or doesn't it's a simple valve that opens and closes as you would have seen. Cleaning it is good but I wouldn't waste money replacing it.
It's safe to run without the MAF connected but absolutely no reason to do so long term unless you've specifically had Alpha N tuned for your modifications, such as running a CSL airbox where you have no MAF for example.
Oh, INPA would be useful for checking your long term trim's/adaptations, see if there's anything untoward there
ph001 said:
This is a very interesting lead indeed and one worth pursuing a little. With the two TPS sensors replaced and no fault codes though I would have thought the DME would know when it wanted to go into open loop mode and would throw codes if there was drama doing so. But I am going to somehow look closer at this idea.
AndyBeech said:
An interesting option to maybe get an Alpha-N map tune. So do you know if the closed loop mode is running with the Alpha-N map, the same Alpha-N map as when the MAF sensor is disconnected?
I Have not yet tried INPA or DIS as I am still waiting on my cable to be delivered, but very keen to do so.
Buzzard88 said:
Well I wouldn't go the Alpha N route personally, unless you've got specific mods that require it and can take advantage, not a cheap route either. Alpha N doesn't adjust fuel trims to compensate for normal engine/sensor wear. It just runs best case scenario as its tuned to the optimal workings of the parts/engine whereas the base map will compensate for readings that variate from the factory defaults to achieve (I believe) lamba 1 or as close as. As a road car, the base map is fine but it does come with it's quirks, especially on for some reason our DME for the S54 which is slightly different to the M3 variation. I'm not sure anyone has worked out exactly why this is because the problem is fairly common as demonstrated on this thread. Alpha N still utilises the lamba sensors as normal as far as I am aware.
Sadly, trying to fix it can also be a money pit as there usually is no obvious reason, fault codes or otherwise. It could be many things related across the whole intake/fuelling chain.
All the sensors/working parts along this chain could be causing the issue, but probably the most important are:
VANOS
MAF
Lamba sensors
Air leaks/split pipes/cylinder issues (including head gasket)
TPS (including throttle pedal)
Throttle Actuator
Fuel system (including injectors/filter (which you can't replace)
Cam timing
CATS
Cam sensors
There's more but you'd hope you would get a error code for failures, though as mentioned, the base map will compensate for sensors not giving expected values within pre-programmed variations without giving error codes but potentially small issues with performance. As mentioned, the DME is looking to achieve lambda 1 and will adapt fuel trims across it programmed tables, long term and short to achieve this. If the lambda sensors (usually pre cat) are off, which is quite common on the S54 because of wear on these sensors then it messes up the whole chain to a degree so if you haven't already, deffo look at replacing the pre cat sensors.
After that, it becomes tricky. The VANOS controls the cam timing constantly (and as such the A/F ratio) and this in itself works by the readings from other sensors so it can be directly wrongly controlled by 'off' sensor readings or not adjusting correctly to correct sensor readings if its slightly faulty but not enough to give an error code. It's a proper minefield as one thing usually dictates another along the chain and our DME generally won't give error codes for these 'minor' faults as they aren't severe enough to warrant it with it's inbuilt 'ok' deviations.
Sorry if a lot of that doesn't actually help but hopefully it gives you a bit of an idea as to the confusing nature of the issue. If I was you, I would clean what you can which it sounds like you've done. Get the intake plenum off and clean the TB's, look for split hoses, ICV etc.
Then your looking at replacing parts, the obvious one's:
MAF
Lambda's
VANOS system
Check cam timing (can easily be disturbed and you want someone competent to check it/set it, will be done with any VANOS work so 2 birds with one stone)
Get it plugged in at BMW, there is a software update for our DME's and it may not have been updated. It was some time ago the update was released but mine only updated when I asked about it at BMW in 2016. It mostly addressed cold running issues but still worth doing.
After that, as you've already experienced, your into replacing parts which may or may not be fine, and on these engines, it's not a cheap venture, especially when you might be replacing parts which are perfectly fine.
It's a tricky one with no definitive answer sadly. Aim for the bigger/preventive parts that will benefit the resale value if you can otherwise you could be spending a lot of cash without benefit or perceived benefit when you come to sell. For example, a refurbished VANOS might help and it will also be a good thing for the future if you see what I mean as worthwhile preventative maintenance.
To add, when you get DIS working, do the VANOS test and see what the readings come back as. It will give a +/- 5 variations to factory defaults and if you are a fair bit outside these it could indicate a problem there, and wouldn't necessarily give an error code either.
AndyBeech has put some great info up there, I perhaps would add to that the DME allows 20% long term fuel trim deviation before spitting out it’s dummy and pushing an error code. Thats a pretty big error imho.
Secondly, a lambda of 1 (ie perfect stoichiometric) is only good for one thing and that is emissions st steady state throttle and idle. For the car to make maximum torque and power, the mixture is way more rich than stoich. Hence any kind of serious throttle input is a mile away from a lambda of 1. That is why the o2 sensors are ignored during acceleration.
Secondly, a lambda of 1 (ie perfect stoichiometric) is only good for one thing and that is emissions st steady state throttle and idle. For the car to make maximum torque and power, the mixture is way more rich than stoich. Hence any kind of serious throttle input is a mile away from a lambda of 1. That is why the o2 sensors are ignored during acceleration.
I'm raising the dead here, I know, but I read all 27 pages of this thread and I just had to make a post about my experience.
I had hesitation between 1500rpm and 3000rpm and the car would stutter when I went WOT between this rev range before giving me full power. Blipping the throttle at idle would result in misfire-like effects, where the throttle response felt delayed and there would be backfire-like sounds from the muffler.
I tried several suggestions from this thread, one at a time, until I isolated the issue to faulty Throttle Position Sensors. There are two:
1. One on the end of the throttle body assembly
2. One on the throttle actuator
They have 3 pins. You can measure if yours are bad by using a multimeter and setting it to Ohms. Put one probe on pin 1 and the other on pin 2. Rotate the sensor and the Ohm value should range between 1000 ohm and 4000 Ohm. If the upper number is larger than 4500 ohm, your TPS is going bad and should be replaced. This rule only applies to TPS with part numbers 13637840383 and 13631402143. I hope this helps someone from the future.
I had hesitation between 1500rpm and 3000rpm and the car would stutter when I went WOT between this rev range before giving me full power. Blipping the throttle at idle would result in misfire-like effects, where the throttle response felt delayed and there would be backfire-like sounds from the muffler.
I tried several suggestions from this thread, one at a time, until I isolated the issue to faulty Throttle Position Sensors. There are two:
1. One on the end of the throttle body assembly
2. One on the throttle actuator
They have 3 pins. You can measure if yours are bad by using a multimeter and setting it to Ohms. Put one probe on pin 1 and the other on pin 2. Rotate the sensor and the Ohm value should range between 1000 ohm and 4000 Ohm. If the upper number is larger than 4500 ohm, your TPS is going bad and should be replaced. This rule only applies to TPS with part numbers 13637840383 and 13631402143. I hope this helps someone from the future.
Simon 3.2M
Senior member
To add my 2 penneth worth, my roadster dipped when the clutch was depressed and it had the usual hesitation. I carried out my usual regular servicing and changed the oil and plugs - the cars pulls more strongly and the revs no longer dip. Maybe there is a correlation, I don't know but it may assist.
just thought I'd add some observations. maybe worthwhile, maybe rubbish. either way...
the S54 is EXTREMELY sensitive to back pressure.. while i was developing my car, i ran it for quite some time ending at the Xpipe (no pipes backwards from the xpipe). now you also have to know that i have no cats and straight through X pipe from Supersprint, as well as the Section 1 from supersprint.. , along with stepped headers from the same mob. Either way, so theres no back pressure.. and that revved like it was going outa fashion.. also other mods, but we wont go into that here..
When it came time for me to get the rear section put on, (3 inch going into 4 inch), and i was away while the guys at the garage welded up the rear pipes, (which are also empty), though they welded up the cans with 2 90 degree bends in it.. At the time i commented on that that would produce too much back pressure, but the guy assured me it would not!! yeah right.. This DID produce a certain amount of loud pedal lag due to a lot of back presusure.. I drove it like that for a few laps.. pulled in and at that point decided to get them to re-weld the rear 3 inch to 4 inch section (i have pics somewhere) section, with only 1 flowing bend in it.. A world of difference.. revs so freely now, like it did with no rear exhaust on it at all.
This was from idle to wot.. sometimes to the point where the car would nearly stall.. absolutely zero issues now.. revs like a MF.. :evil:
so adding to that list you might wanna see if the exhaust system has been modified. :roll:
just my 2 dirhams.
the S54 is EXTREMELY sensitive to back pressure.. while i was developing my car, i ran it for quite some time ending at the Xpipe (no pipes backwards from the xpipe). now you also have to know that i have no cats and straight through X pipe from Supersprint, as well as the Section 1 from supersprint.. , along with stepped headers from the same mob. Either way, so theres no back pressure.. and that revved like it was going outa fashion.. also other mods, but we wont go into that here..
When it came time for me to get the rear section put on, (3 inch going into 4 inch), and i was away while the guys at the garage welded up the rear pipes, (which are also empty), though they welded up the cans with 2 90 degree bends in it.. At the time i commented on that that would produce too much back pressure, but the guy assured me it would not!! yeah right.. This DID produce a certain amount of loud pedal lag due to a lot of back presusure.. I drove it like that for a few laps.. pulled in and at that point decided to get them to re-weld the rear 3 inch to 4 inch section (i have pics somewhere) section, with only 1 flowing bend in it.. A world of difference.. revs so freely now, like it did with no rear exhaust on it at all.
This was from idle to wot.. sometimes to the point where the car would nearly stall.. absolutely zero issues now.. revs like a MF.. :evil:
so adding to that list you might wanna see if the exhaust system has been modified. :roll:
just my 2 dirhams.