cyow5
cyow5 Reader
7/27/23 10:33 a.m.

Alright, I am going on two weeks now, and e90post.com has been worth its weight in sewage. Maybe that's generous because I imagine there's a market for sewage. Anyways, here's to hoping y'all can offer some insight. 

Starting about two or three weeks ago, my E91 with N52 (328i manual) would take a long time to start. Cranking speed was okay, just no light-off. Then it would fire up, accompanied by a DSC light and eventually a CEL. Given the car's high mileage (140k) and I only had it for 20k miles, I wanted to knock out some of the known wear items first before really diagnosing since some of these could be culprits. I changed the battery (9 years old), water pump, thermostat, oil pan gasket, and VANOS solenoids. Slow starting persisted after these changes, and now I had a VANOS error. One of the solenoids did not go in nicely, and I could've damaged it with all the wiggling and prodding, so I swapped it with a used but good solenoid. Still slow to fire. 

Next, I swapped the crank sensor with a used but good sensor, and I swapped the cam sensors at the same time (same donor motor for all used parts). Slow start has not since returned, but now it traded in a harsh stumble at around 3k rpm. I am pretty sure the stumble was new after the crank sensor, but it *has* always had a stumble off idle when engaging the clutch and shortly after. It could either be I created the problem when doing the crank sensor, or the problem got worse when the other issues were sorted and this stumble is now center stage. Prior to doing all the work, it could stall when pulling in my driveway, and this is still the case. It is not driver error. I thought it could be that the first time, but now it has consistently stalled or tried to stall when pulling in my driveway after experiencing the stumble. This is my main evidence that I did not create a new problem when doing the crank sensor. After pulling in the garage after experiencing the near-stall and stumbling, it idles like it always has. Then, out of nowhere, revs will flair (2k) and die. 

When doing the crank sensor, I noticed the grain strap near the crank sensor was horribly frayed, so I swapped it. No change. I also removed the crank sensor, checked the plug and harness, cleaned it, and reinstalled it. Still stumbles. 

I moved on to the valve cover gasket which had been leaking, and I replaced the valvetronic gasket, fuel injectors and rail (donor), eccentric shaft sensor (donor motor), and its gasket. Cams all look good, chain looked good, no sludge or buildup. All in all, it looked great for even half the miles. Put everything back together, and the valve cover leaks seem to be gone, but the stumble is still very present. 

I took out the MAF, and it looked fine. Wires are a bit tighter than I'd like because of the AEM intake, but everything seems intact. I cleaned it anyway, no change. I drove with it unplugged, and it drove all sorts of awful. Reconnected the sensor, still stumbles and idle still dies. 

Next, I removed the VANOS check valves / filters. They were clean, which I expected from the clean camshafts. I sprayed them out anyway with carb cleaner and am letting them dry. That's where I am currently. Any thoughts? One friend who is a BMW guru suggested it may be a fueling issue. I did noticed that I had no rail pressure when I pressed the relief valve on the rail, but it did dribble out of the hose when I disconnected that. A poor filter or pump could explain a lot but not the flair before dying at idle. That is, unless the ECU is doing something strange in response to a lean condition, but it is not reporting any fueling faults (or any other faults for that matter now). It would also be a remarkable coincidence to have such a major fuel deficiency after all these changes when fueling was never an issue before. I frequently floor it to redline, and it has never missed a beat under load. 

VANOS adaptations can do strange things, according to the internet. I have not tried resetting those yet, but that's pretty much the only theory I have left that could fit this whole chain of events without relying too heavily on bad luck / coincidence. 

cyow5
cyow5 Reader
7/27/23 2:20 p.m.

Bump - could use some help here. Cleaning the check valves did nothing, which I expected since there were no obvious issues. The first test drive seemed a little better, but the second one was easily no different than normal. One difference is now I have a misfire code for cylinders 5 and 6, and the idle was obviously misfiring. It seemed to have normal torque until the light for the misfire, and then it had absolutely no torque at all. 

enginenerd
enginenerd HalfDork
7/27/23 2:51 p.m.

Sounds like you've addressed all the normal culprits. Having worked on mine plenty I think you're going to have a pretty tough time narrowing it down without diagnostic software. Something generic could work if you can see adaptations and sensor outputs but the BMW ISTA software is around and cheap OBD cables can be found on Amazon.

 

cyow5
cyow5 Reader
7/27/23 2:57 p.m.
enginenerd said:

Sounds like you've addressed all the normal culprits. Having worked on mine plenty I think you're going to have a pretty tough time narrowing it down without diagnostic software. Something generic could work if you can see adaptations and sensor outputs but the BMW ISTA software is around and cheap OBD cables can be found on Amazon.

 

Yeah, I am using Carly and DashCmd. Currently shopping for a new laptop, so that's delayed getting any proper software. I do have a friend with ISTA, so that's on the docket. Or just tossing it to a dealer and cutting a blank check. Not really wanting to do that, but I have sunk way too much time into this so far with next to nothing to show for it. 

enginenerd
enginenerd HalfDork
7/28/23 12:28 p.m.

Wish I had a better direction to point you in...if it were most other BMW engines of the era I would say to take a look at the intake valves for deposit build up but the N52 wasn't DI. Could be a possible control issue (behavior changed with new sensors) but it could just be the combination of existing adaptations and new sensors, or more likely they are just still trying to correct for an underlying issue. 

I think at this point what I would do is:

  • Use ISTA or similar to reset all adaptations then keep a close eye on anything related to fuel trim feedback loops, injector and ignition characteristics, and VANOS behavior. Most likely looking for an imbalance between banks or a single cylinder.
  • Smoke test to verify no possibility of air leaks
  • Check spark plugs, replace if not recently done. Check coils if original to car or aftermarket
  • Check injectors and clean (saw you replaced these but not sure if new or from another car)

*Not professional advice...just from a guy that has owned too many old BMWs

enginenerd
enginenerd HalfDork
7/28/23 12:41 p.m.

Also, more than likely unrelated but with the different variants of N52, how many (if at all) DISA valves does yours have?

cyow5
cyow5 Reader
7/28/23 12:45 p.m.
enginenerd said:

Wish I had a better direction to point you in...if it were most other BMW engines of the era I would say to take a look at the intake valves for deposit build up but the N52 wasn't DI. Could be a possible control issue (behavior changed with new sensors) but it could just be the combination of existing adaptations and new sensors, or more likely they are just still trying to correct for an underlying issue. 

I think at this point what I would do is:

  • Use ISTA or similar to reset all adaptations then keep a close eye on anything related to fuel trim feedback loops, injector and ignition characteristics, and VANOS behavior. Most likely looking for an imbalance between banks or a single cylinder.
  • Smoke test to verify no possibility of air leaks
  • Check spark plugs, replace if not recently done. Check coils if original to car or aftermarket
  • Check injectors and clean (saw you replaced these but not sure if new or from another car)

*Not professional advice...just from a guy that has owned too many old BMWs

I agree with your suggestions and really appreciate the input. I dropped the car off at a mechanic this morning with instructions to reset VANOS adaptations as well as fuel and everything else, so here's to hoping it is that simple. I'm just tired of messing with it at this point and wasted enough time on it. 

If a reset does not help, compression is next. Plugs and coils are new-ish (<20,000 miles) and OE-spec,  but sh!t happens. Honestly, this motor looks really good on the inside. It has the usual BMW leak issues, but no sludge or excessive wear or things like that. Injector nozzles looked good. A little crud had gotten past the o-rings, but the nozzles themselves were clear of debris. I have not had them flowed yet though. After that, fuel pressure to see how the pump and filter are doing. 

cyow5
cyow5 Reader
7/28/23 4:53 p.m.

Plot twist - it was the crank sensor. Well, specifically, it was the extra o-ring. The old o-ring was still in the block when I removed the old sensor, and I didn't notice. When I put in the new sensor with its old o-ring, there was one o-ring too many. 

Note to self, always remove the old parts *before* adding new ones...

enginenerd
enginenerd HalfDork
7/28/23 5:38 p.m.

In reply to cyow5 :

Glad you got it sorted! I missed the part about it being used but it makes sense the behavior changed after it was swapped. Funny how a single o-ring can make such a difference...

cyow5
cyow5 Reader
7/28/23 5:42 p.m.

In reply to enginenerd :

Yeah, even with a new one I would've made the same mistake since a new one would also have a new gasket. It's a bit of a dumb mistake, but it is a good reminder when doing diagnostics to never assume any step was done correctly. 

You'll need to log in to post.

Our Preferred Partners
xkxK2Vr8UnhPXQCYFS89s497jIc9kDxIy827xasv48g1Pv69HYLKnc9uCHvSL4GD