Weird idle issue after disconnecting battery overnight

[madbum90]

So I disconnected my battery overnight because I was clearing ecu codes. When I connected the battery back up the car would start fine and idle fine. But when I drive off and come to a stop, the idle goes up to 1500 rpms for a quick 2-3 seconds and then drops to normal. It does this wether it's hot or cold. And every time I start up the car it idles perfectly fine. It's just after driving and coming to a stop. I've also noticed when I shift that sometimes the Rpms will hang there and sometimes they don't. Anyone have this issue before?

[bmwdirtracer]

90 %:You've got an intake leak. A smoke test with a professional smoke machine is how to find it.
10%: Idle Control Valve is sticking. To fix this, remove it, fill with carb cleaner and shake vigorously. Repeat until carb cleaner comes out clean.

Oh, and a "battery reset" can be done in 60 seconds, by disconnecting both battery cables and holding them together. This accomplishes exactly the same thing as disconnecting the battery overnight. However, while this "reset" can reset the CEL, it does not clear codes. Clearing codes requires a scan-tool, or even a Peake device, or until a certain number of drive cycles are completed.

[madbum90]

90 %:You've got an intake leak. A smoke test with a professional smoke machine is how to find it.
10%: Idle Control Valve is sticking. To fix this, remove it, fill with carb cleaner and shake vigorously. Repeat until carb cleaner comes out clean.

Oh, and a "battery reset" can be done in 60 seconds, by disconnecting both battery cables and holding them together. This accomplishes exactly the same thing as disconnecting the battery overnight. However, while this "reset" can reset the CEL, it does not clear codes. Clearing codes requires a scan-tool, or even a Peake device, or until a certain number of drive cycles are completed.

thank you for the info! It seams odd that it could be either those 2 since it was running flawlessly and all I did was disconnect the battery and let it sit overnight. I thought maybe the icv had to relearn or something along those lines

[bmwdirtracer]

Always remember coincidence. Some things don't happen because of what you did.......

.....and some things happen because of a related matter. Like, your computer had developed a set of adaptations, to deal with a large intake leak. Unplugging the battery (on a '95 car) might just have reset adaptations, and made your engine run worse.

What codes were you trying to erase?

I'd still start with a smoke test.

[madbum90]

Always remember coincidence. Some things don't happen because of what you did.......

.....and some things happen because of a related matter. Like, your computer had developed a set of adaptations, to deal with a large intake leak. Unplugging the battery (on a '95 car) might just have reset adaptations, and made your engine run worse.

What codes were you trying to erase?

I'd still start with a smoke test.

That is very true, I have a 1221 code for o2 sensor. O2 sensor is brand new and the light came back on. But I'm also not running cats so maybe that's why the lights on.

[bmwdirtracer]

A large intake leak will set a code for the Oxygen sensor, because the DME can't correlate the data from the MAF with the mixture readings from the O2 sensor.

[shadowpuck]

just to add - on a '95 m3, the oxygen sensor is prior to the cats anyway. running a cat delete wouldn't trigger (or i wouldn't expect it to trigger) an o2 code.
i have had a custom exhaust trigger both a 13 and 201 (13 being o2 sensor and 201 being o2 sensor control in bmw speak on this car). are you running stock headers?

like chris says - i'd look for the basics first.

[madbum90]

A large intake leak will set a code for the Oxygen sensor, because the DME can't correlate the data from the MAF with the mixture readings from the O2 sensor.

Alright I'll look for a leak, could it also be that I'm running a 3.5" maf with 24lb injectors and no chip for the set up?

- - - Updated - - -

just to add - on a '95 m3, the oxygen sensor is prior to the cats anyway. running a cat delete wouldn't trigger (or i wouldn't expect it to trigger) an o2 code.
i have had a custom exhaust trigger both a 13 and 201 (13 being o2 sensor and 201 being o2 sensor control in bmw speak on this car). are you running stock headers?

like chris says - i'd look for the basics first.

I'm running tms shorty headers

[328 Power 04]

Alright I'll look for a leak, could it also be that I'm running a 3.5" maf with 24lb injectors and no chip for the set up?

Ohh ya. You're actually losing power this way.

- - - Updated - - -

That was the single most important piece of information you omitted.

[shadowpuck]

well, now, that's a critical piece of information.....

where is the o2 sensor located?
a 3.5 maf and 24lb injectors without a tune i would expect would make the computer quite unhappy.

[madbum90]

well, now, that's a critical piece of information.....

where is the o2 sensor located?
a 3.5 maf and 24lb injectors without a tune i would expect would make the computer quite unhappy.

So yea the o2 is before the cat. And yea I could see that just thought it was weird it didn't idle bad prior to battery change. I've ordered up the chip so while I wait for it in the mail illlookfor vac leaks. But you're thinking it's possibly just in need of a chip to idle fine?

[328 Power 04]

Yes, the MAF sensor needs to be defined correctly in the tune for proper idle.... also the same exact thing with injectors.

[shadowpuck]

just curious, how are you reading codes right now?

yes, knowing the proper calibration for the maf and the injectors is critical to how the tune develops its responses, etc.
from personal experience, i can tell you that on an OBD I M3, getting the o2 sensor in the right spot on a custom exhaust is important.

it's never a bad idea to smoke the intake system to check for leaks.

[madbum90]

Yes, the MAF sensor needs to be defined correctly in the tune for proper idle.... also the same exact thing with injectors.

Alright cool I look forward to getting the chip then and seeing the outcome

- - - Updated - - -

just curious, how are you reading codes right now?

yes, knowing the proper calibration for the maf and the injectors is critical to how the tune develops its responses, etc.
from personal experience, i can tell you that on an OBD I M3, getting the o2 sensor in the right spot on a custom exhaust is important.

it's never a bad idea to smoke the intake system to check for leaks.

I'm doing the pedal stomp method. How do tell where the right spot is?

[shadowpuck]

i'd try it in the stock spot first - if that's an option. that's where we started. we did have to move mine closer to the engine....i'm not running shorty headers, but i'm also not running a stock exhaust.

look into picking up a code reader, at the very least, they are not expensive at all - peake makes one that would be just fine for your purposes.
the other option, if you want a small project, would be to build a INPA laptop with a 20 pin ADS adapter.
carly (a bmw diagnostic app) is supposed to work with OBD 1/ADS equipped BMWs - i've verified that the software can talk to such a car, but i've not gone any further.

or, check into your local bmw cca chapter - often times members or chapters will have diagnostic gear and they'll let you use it for the price of a beer or two!

if you don't have one already, also a bentley service manual is real nice to have if you are going to work on this car.

basically - it'll serve you well to not have to rely upon battery disconnects and stomp methods....

[bmwdirtracer]

I absolutely agree with my wonderful and erudite friends, shadowpuck and 328 Power 04, that you need a custom chip to go with your custom equipment.

That said, there's no question in my mind that a smoke test is where I'd start, because an INTACT intake system is the first parameter to establish, when you have a fuel / air mixture issue.

[shadowpuck]

yup, also agreed on that point, chris! you have to know the system is capable of running right before you can determine what's wrong.

[madbum90]

I absolutely agree with my wonderful and erudite friends, shadowpuck and 328 Power 04, that you need a custom chip to go with your custom equipment.

That said, there's no question in my mind that a smoke test is where I'd start, because an INTACT intake system is the first parameter to establish, when you have a fuel / air mixture issue.

So I got in contact with the previous owner that I just bought the car off of. He found the chip that it had previously. It's a turner chip and he said he took it off because when he had it on, the car would have a intermittent no start. Any ideas to that?

[MIKYZZ4]

I suspect he meant a hard start issue, probably due to an intake leak, as we all now suspect.
--------------------------------------------------------------------------------
We only know as much as you know, omitting vital info makes diagnosis difficult

[shadowpuck]

well, lots of possibilities there with little info....

also, people notoriously describe hard start when they mean no start, and vice versa, etc. so, we'd need to know exactly what the car is doing in that situation.

now, my educated guess (which, really, is no different than a guess covered in clever) is he may have been describing an actual hard start. the only reason i say that is that it sounds like this car has been modified, and given some of the other issues (including us now finding out about the possibility of swapped chips) makes me suspicious of the DME. However, we need to know the build date of the car before i can make any real suggestions. You can get the build date from the vin (simply type the last 7 of the VIN into realoem.com), or you can look at the build plate on the driver's side door jamb. of course, that's assuming that everything still matches - hard to say at this point!

what i've experienced is a car that will sometimes take long crank time before starting. we put a turner chip in tuned for the car's mods and the proper EWS (based on build date) and all worked fine after that. what was interesting was the car would still start while appearing to have some type of EWS issue. EWS is the drive-away protection system - or, in simple terms, the security system.

now, as i've said - depending upon your build date and what DME is in the car - none of that may apply at all.

in my example i had a 6/95 build (late june at that, too) with EWS II. once we got the right chip from Turner, all issues went away.

[madbum90]

well, lots of possibilities there with little info....

also, people notoriously describe hard start when they mean no start, and vice versa, etc. so, we'd need to know exactly what the car is doing in that situation.

now, my educated guess (which, really, is no different than a guess covered in clever) is he may have been describing an actual hard start. the only reason i say that is that it sounds like this car has been modified, and given some of the other issues (including us now finding out about the possibility of swapped chips) makes me suspicious of the DME. However, we need to know the build date of the car before i can make any real suggestions. You can get the build date from the vin (simply type the last 7 of the VIN into realoem.com), or you can look at the build plate on the driver's side door jamb. of course, that's assuming that everything still matches - hard to say at this point!

what i've experienced is a car that will sometimes take long crank time before starting. we put a turner chip in tuned for the car's mods and the proper EWS (based on build date) and all worked fine after that. what was interesting was the car would still start while appearing to have some type of EWS issue. EWS is the drive-away protection system - or, in simple terms, the security system.

now, as i've said - depending upon your build date and what DME is in the car - none of that may apply at all.

in my example i had a 6/95 build (late june at that, too) with EWS II. once we got the right chip from Turner, all issues went away.

The build date is 07/94. I've looked and looked for a vacuum leak and can't find anything. It's now starting to sputter after 1,500 rpms or so. My plan is to disconnect battery overnight, soak maf in maf cleaner, visually check spark plugs and install the chip and hope that brings the car back to life

[shadowpuck]

looking for a vacuum leak isn't going to gain you much....

you need to go back to basic troubleshooting (in my opinion, just my thoughts) and eliminate variables. we have too much going on here. we need to get data to understand the issue.

[MIKYZZ4]

Hook up a fuel pressure gauge to observe pressure during sputtering.
I would still perform a smoke test to confirm a baseline idle.
I think the chip is going to surprise us all, and introduce a new set of variables.

[madbum90]

looking for a vacuum leak isn't going to gain you much....

you need to go back to basic troubleshooting (in my opinion, just my thoughts) and eliminate variables. we have too much going on here. we need to get data to understand the issue.

I put the chip in and now when I turn the key to on position the rpm needles on the gauge cluster bounce around like crazy and when I try and start it this is what happens. Crank, stop, crank, stop and forever like that

[madbum90]

Update: so since my car is a 95 m3 build date 07/94 my edu should be bosch 506 from what i hear. but the ecu i have is a 413 red label (0 261 200 413) that edu thats in the car has a stock chip in it. when i put the turner chip in it, the rpm needles mum around like crazy and car cranks and stops cranks and stops. I have another red label 413 laying around so i figured ill throw the turner chip in that one and see if anything changes. the car starts up and wants to die. so I let it warm up and it runs flawlessly. but then after idling for a little bit the motor just dies. Doesn't hesitate just shuts off quick like the key was off. now it won't start period just cranks over