Troubleshooting odd engine surging, what else is left?

[m.olennick]

I finally got the O2 sensor replaced this afternoon. I'm up in the air with the results.

Firstly, the mid-range and overall engine performance "feels" smoother -- surprisingly I also "feel" an improved pull in the mid-range RPM.

The surging under a load is gone, however it still feels as though there is a small bit of surging when under partial acceleration i.e. getting up to street speeds ~35mph.

Definitely an improvement, dunno what to look for next.

All seems positive. For your sake I hope it smooths out more and you don't have to go to the extent I have in diagnosing and replacing parts. Also, since I'm on the subject; how did you get your injectors out of the manifold or off the fuel rail? I'm in the middle of replacing mine and those bastards won't budge. It's tough to get any leverage and I don't want to bend the rail.

This relates to the M 20 engine , but the super eta fuel rail is a minor PITA to remove. You have to unbolt it from its fastenings. I found I got a better grip at the front of the rail. I had best success removing the injectors and the rail from the head as a unit.

That is how I went about it as well.

There are four (if I am correct) bolt fixing the rail to the manifold. After removing them, and removing all the fuel lines and draining as much as I could into a shop rag -- it was merely a game of persistence. The rubber o-rings really do swell up inside the fittings so I pulled the rail/injectors as a unit as that was the "easiest."

The injectors I attacked one by one on the ground and had them out nicely.

Replacing the O2 sensor really is a 20/30 minute job being extremely careful and putting all the zip-ties back in, otherwise it should only take a couple minutes of actual work.

[dauer]

Well finally got the old injectors to budge with some coaxing courtesy of some big channel locks and the valve cover as a fulcrum. Installed rebuilt design III injectors and let me tell you, they go in a lot easier than the old ones came out and aside from losing an injector connector clip, all went well. The downside is it didn't cure my partial throttle surging, although it wasn't for naught because the old ones were leaking and had to go anyhow. One thing worth noting is that the rebuilt, cleaned injectors were ticking loudly. Is this common or should I be concerned that the supposedly rebuilt and cleaned injectors are rubbish?

[m.olennick]

Well never mind, my problem is back. Except when cruising on the highway (oddly enough.)

Additionally, I was playing with the throttle body and noticed that when holding the flap at a certain angle of attack, the rpm would climb, hold steady, drop approximately ~100-200 and then proceed to climb up ~100-150 (below what the initial rpm holding point was.) I did this from both the TB and while in the cabin yielding the same results each time.

[dauer]

I'm no expert, but you might want to check the throttle position sensor. Not sure what would cause it to move, although it wouldn't hurt. At the very least you'll rule it out.

[m.olennick]

I'm no expert, but you might want to check the throttle position sensor. Not sure what would cause it to move, although it wouldn't hurt. At the very least you'll rule it out.

Nope, replaced it already.

[m.olennick]

Any other input? This is really maddening.

[dauer]

Any other input? This is really maddening.

Likewise, I'm at my wits end and running out of options.

[m.olennick]

Any other input? This is really maddening.

Likewise, I'm at my wits end and running out of options.

I believe my problem is getting worse. I'm wondering if this may be a sign of a failing ECM/ECU unit?

Can you describe the feeling you get from your engine? Like your's, my problem is around 10% throttle. The engine jerks and does feel like a misfire-ish. If I'm cruising at low speeds, I can aggravate the car by keeping the rpm at ~1200 in second and give the usual 10% throttle, the car will buck and jerk extensively.

[dauer]

I believe my problem is getting worse. I'm wondering if this may be a sign of a failing ECM/ECU unit?

Can you describe the feeling you get from your engine? Like your's, my problem is around 10% throttle. The engine jerks and does feel like a misfire-ish. If I'm cruising at low speeds, I can aggravate the car by keeping the rpm at ~1200 in second and give the usual 10% throttle, the car will buck and jerk extensively.

You just described my symptoms exactly. The surging seems to be directly linked to throttle input. Therefore if I'm experiencing surging at say a 10% throttle if I dip my foot into it a bit more the engine smooths right out and pulls hard. My engine will surge at any RPM if I'm maintaining or accelerating slightly, however it's more pronounced at lower speed/RPM because I don't have momentum to help mask the surging.

[m.olennick]

I believe my problem is getting worse. I'm wondering if this may be a sign of a failing ECM/ECU unit?

Can you describe the feeling you get from your engine? Like your's, my problem is around 10% throttle. The engine jerks and does feel like a misfire-ish. If I'm cruising at low speeds, I can aggravate the car by keeping the rpm at ~1200 in second and give the usual 10% throttle, the car will buck and jerk extensively.

You just described my symptoms exactly. The surging seems to be directly linked to throttle input. Therefore if I'm experiencing surging at say a 10% throttle if I dip my foot into it a bit more the engine smooths right out and pulls hard. My engine will surge at any RPM if I'm maintaining or accelerating slightly, however it's more pronounced at lower speed/RPM because I don't have momentum to help mask the surging.

I'm meeting up with a local tuner. Said it might be the ECU/ECM (whatever they are called on this chassis.) We are going to play around and maybe try a different box.

[dauer]

I'm meeting up with a local tuner. Said it might be the ECU/ECM (whatever they are called on this chassis.) We are going to play around and maybe try a different box.

When the issue initially reared it's ugly head I thought it was the ecu then I replaced the crank reference sensors because I got some funky readouts from them, after that the ecu kind've fell by the wayside. I'm eager to know what you find out keep us posted.

[m.olennick]

I'm meeting up with a local tuner. Said it might be the ECU/ECM (whatever they are called on this chassis.) We are going to play around and maybe try a different box.

When the issue initially reared it's ugly head I thought it was the ecu then I replaced the crank reference sensors because I got some funky readouts from them, after that the ecu kind've fell by the wayside. I'm eager to know what you find out keep us posted.

Ok, after a ride along my tuner thinks a couple things "too many variables." Lol.

Primarily he is concerned that the ECU has just failed. Seeing as you and I both are experiencing the same problem and we have different air metering systems (MAF vs. AFM.) He is wondering about my idle, noting a possible mis-fire. He briefly mentioned a back-fire but wasn't very confident in that assertion.

I'll be trying a new ecu here soon. If that doesn't work, he wants me to check the injectors, I don't really want to so I would have the flow benched.

Lastly to check is the timing and verify that I infact am not one-tooth off. I don't know how this could be as the car is extremely powerful compared to the first time I got in it. The mystery machine indeed.

[steve 87 535is]

I've been watching this thread for a sloution, as I have it on my 85 e23 735i automatic. It only happens when I let off the accelerator when approaching a stop sign, or driving in a parking lot etc. I too can sometimes make it go away with a little accelerator work-adding just a touch.
The only time it got any better is when I adjusted my throttle cables alittle bit. It lasted about a year, then came back this past summer. I absolutely agree that this is very maddening and annoying.
On another thread, someone suggested that it could be just the injectors shutting down at a certain RPM and the electronics getting alittle mixed signal causing the surge, surge, surge, smooth.
I took the car off the road before I could play with the throttle again, but that's my first thing to adjust in the spring.

Some points on my car:
Original:about 190,000 miles
02 sensor
injectors
ECU swap didn't help
AFM swap, may have helped for a while but I did that around the same time of the first throttle adjustment. Both AFM's are unopened and not tampered with.
TPS- adjusted to specs
Tranny sensors

Usual tuneup stuff has been done: plugs, cap, rotor, valve adjust, fuel filter, both pumps, coolant sensors etc etc..

Have either of you played with the throttle or AFM?
I just thought of a question: my original cat section has cracks in it, the cats case itself, not the pipes. I've been brazing them up as best as possible, but some have opened up again. Could this be a cause also?

[Manny D.]

The idle is rock solid and throttle response is good. The problem is when in gear and maintaining a steady up to 10% throttle my engine exhibits a surging behavior. If I get on it a bit more it's fine. I checked for an intake leaks per Rich's advise and none were found. I also triple checked for vacuum leaks under the intake and found none. How many vacuum lines should there be? I only saw one connected under the manifold near the throttle body.

My car does the same thing.. me and the guy at the bmw shop have been trying to figure it out as his 535i does the verry same thing.. mine usually happens around 50-70 mph and 2,000-2,500rpm crusing with little throttle and it surges, Give it just a little more and its gone.. hope we figure out this annoying problem.. I was woundering if it could be a bad fuel pump?

[demetk]

I'd like to add my experience to this thread. I had the same problem with my 528e (533 engine, 059 ecu) - at around 30 mph, if I let off the gas the engine would surge repeatedly until the car dropped in speed sufficiently where the idle system would kick in.

I tried the following adjustments/tests with no luck:

Changed O2 sensor - just in case.
Checked for vacuum leaks with a propane cylinder - none found.
Checked temp sensor - ok.
Checked AFM temp sensor - ok.
Checked AFM carbon track for voltage output continuity - no dropouts.
Disconnected purge system - just in case.
Checked plugs, wires, cap & rotor - all fine.
Valves were adjusted very recently.
Adjusted throttle opening and TPS to specs - one more time.

I was about to try another ECU until I reasoned that the surging was very similar to what others have described as lean surging. If that was the case here, then what I needed to do was richen the mixture. Well the AFM mixture screw affects mixture at idle so that wouldn't do. But what I also learned from these boards is that the AFM spring affects mixture throughout the rpm band, so I popped the AFM cap with a shot of compressed air and loosened the AFM spring (the thick type) 2 notches rich. The adjustment was still within the ecu's ability to control mixture as verified by the O2 sensor voltage rapidly jumping low & high.

So I took it for a spin and wouldn't you know it the surging has almost vanished. I'm planning on one more click rich on the AFM spring.

Hope this helps.

[BRRV]

This seems like an epidemic problem - everybody can relate to the issue, as they seem to have a little bit of it. The throttle plate is adjusted at the factory, and I think is subject to wear or needs adjustment at high mileage. I loosened and adjusted mine (set screw on the throttle body, next to the return springs on the shaft) and it doesn't do it anymore. YMMV.

The other thing I was thinking of (and this would be just about impossible to diagnose) was that, with high mileage cars, the bores in the throttle body can wear slightly, at certain positions, or the shaft can go out of round. This could introduce 'false air' and could raise an idle slightly, especially off-throttle. This used to be a common problem on SU carbs, which have a similar 'side draft' throttle arrangement. NO, I am not comparing an SU side-draft to a BMW throttle body, merely suggesting that the same arrangement might contribute to an idle issue -it always did with the SU's. The fix was to have the throttle bores replaced and re-bushed.

For cars that have a very pronounced surge off-throttle, when decelerating, I'd check the AFM closely, making sure the flapper (swing arm) is free and not hanging at all.

[m.olennick]

This seems like an epidemic problem - everybody can relate to the issue, as they seem to have a little bit of it. The throttle plate is adjusted at the factory, and I think is subject to wear or needs adjustment at high mileage. I loosened and adjusted mine (set screw on the throttle body, next to the return springs on the shaft) and it doesn't do it anymore. YMMV.

The other thing I was thinking of (and this would be just about impossible to diagnose) was that, with high mileage cars, the bores in the throttle body can wear slightly, at certain positions, or the shaft can go out of round. This could introduce 'false air' and could raise an idle slightly, especially off-throttle. This used to be a common problem on SU carbs, which have a similar 'side draft' throttle arrangement. NO, I am not comparing an SU side-draft to a BMW throttle body, merely suggesting that the same arrangement might contribute to an idle issue -it always did with the SU's. The fix was to have the throttle bores replaced and re-bushed.

For cars that have a very pronounced surge off-throttle, when decelerating, I'd check the AFM closely, making sure the flapper (swing arm) is free and not hanging at all.

This has been a looming thought in the back of my mind. I thoroughly cleaned my throttle body when I had my engine nearly fully disassembled over my winter break. I ended up cleaning a ton of buildup off the valve and may have been a bit forceful trying to wipe things up.

I wonder if this could be a problem?!

Wow, fuck that -- a new throttle body would be a little spendy ~ $350 my cost. Might have to try a junker for some trouble shooting fun!

[moengel]

Surging idles or hunting idles are often caused by vaccum leaks. I would have checked this first. I'm surprised no one mentioned it. The backfire could have easily created the leak.

Mark

[dauer]

Surging idles or hunting idles are often caused by vaccum leaks. I would have checked this first. I'm surprised no one mentioned it. The backfire could have easily created the leak.

Mark

This has been mentioned numerous times and as far as I'm concerned it was the first thing I checked. I'm still troubleshooting this problem and plan to replace the O2 sensor today. I'll keep you guys updated.

[moengel]

So I've been troubleshooting what can only be described as the most maddening engine surging problem on my 88' 535i. I decided to start a new thread in order to update and solicit some more input from you guys regarding my ongoing problem.
A brief description: it all started with a back-fire upon start up about 10 weeks ago, since then car has been exhibiting an odd surging at steady to 10% throttle. Lay off the throttle or put my foot in it a little more and it runs as it should. Idle is rock solid (for an M30) and it'll run to redline with no hesitation. Here's what I've R&R'd and has either been replaced or checked out fine, sometimes both:
ICV good
O2 sensor good
coolant temp sensor good but replaced anyway
FPR "
Fuel filter replaced
Have fuel pressure
AFM swapped out with unopened known good one
Coil good
Plugs, wires, cap, and rotor replaced
Both flywheel reference sensors replaced
Injectors show correct resistance
TPS good

I'm probably forgetting something but that's the gist of it. I'm really at my wits end. What else is left? The only thing I can think of that's left is getting reman'd injectors and performing a valve-adjustment, both of which I've never done myself, therefore I'm lacking a little confidence. Thanks in advance.

P.S. My MPG is down the tubes.

Sorry, you didn't mention it in your initial description and I, admitingly, was too lazy to read through the rest of the posts. My apologies. However, I still think you may have a vacumm leak. A vacuum leak on the exhaust side, before your O2 sensor, can cause similar problems and would contribute to your low MPG. It's also consistant with the problem occuring after your backfire. Check the exhaust manifod gasket for leaks.

And good luck. I know how annoying these things are.

Mark

[BRRV]

And, FWIW, when a "tech" says something like "I think your ECU is gone" and the car is still running within normal parameters, I'd walk away fast. When they fail, they fail, usually as a no start condition, sometimes as a full-rich condition, but it's almost always obvious after about ten minutes of troubleshooting with a meter. Everybody's knee jerk reaction to these old cars and odd driveability issues is "it's the ECU". It's always easier to replace a single box than to troubleshoot driveability issues properly. Swapping a box with a known working unit often doesn't solve the problem. This is not to say that a 20 year old component can't go wrong - I just think that it's more likely it's something else minor that's been overlooked. It doesn't take a genius to replace the ECU.

The false air issue (intake air leaks) has been covered in great detail previously in this thread. I for one think, given the price of a new throttle body, junk yards should be scavenged immediately for any that you can find. A lower mileage car just might yield a throttle body with less wear and it might be the cure. Can I prove this? Nope. Is it worth considering? Absolutely, IMO.

[demetk]

Check the exhaust manifod gasket for leaks.

How would you do that? Visual and/or sound?

[moengel]

Basic visual and sound test would be a start. If you partially block the tail pipe while the car is running, it should force some exhaust out the leak and make it easier to find. I've also heard of guys dripping ATF into the carb to get exhaust smoke but our cars don't have carbs, so I don't know if there is another way to do this or not.

Mark

[dauer]

Check the exhaust manifod gasket for leaks.

How would you do that? Visual and/or sound?

X2. Can't exactly spray WD-40 on a hot exhaust manifold. Although I guess it's common sense to conduct any tests while the exhaust's still cold, which I imagine isn't very long.

[moengel]

Can't exactly spray WD-40 on a hot exhaust manifold.

This wouldn't do you any good even if the exhaust manifold was cold the entire time the car was running.

[dauer]

Check for a blown intake manifold gasket. Sometimes when you have a significant backfire, it will blow a small chunk of the gasket out. You can do this by squirting WD-40 around the intake manifold / head interface when the engine is idling. If the engine RPM increases you've found a leak. You can do the same for anyplace you suspect a vacuum leak.

Rich

This is what maid me reference the use of WD-40. Any other suggestions?

[BRRV]

This is what maid (sic) me reference the use of WD-40. Any other suggestions?

See my posts on Page 1.

Also, an exhaust manifold gasket leak (if I read that correctly) won't cause this problem.

[Rich Euro M5]

This has been mentioned numerous times and as far as I'm concerned it was the first thing I checked. I'm still troubleshooting this problem and plan to replace the O2 sensor today. I'll keep you guys updated

This problem sounds a lot like a sluggish O2 sensor to me. You can't simply take the Bentley test as gospel, when measuring the O2 sensor output, with the multimeter in DC mode it should be oscillating "dither" between about .4 vdc - .7 vdc. At idle it should dither at a rate of about 1 cycle per second. As you increase the engine RPM while monitoring the O2 sensor output voltage the cycling of the output voltage should increase in frequency. If you note the O2 sensor simply maintaining a constant voltage while measuring it's output iin DC mode, the O2 sensor is probably shot.

Another test is to simply disconnect the O2 sensor, run the engine open loop and see if the surging goes away. If it does, it indicates your A/F ratio is too lean in closed loop mode. This points back to vacuum leaks, O2 sensor issues, etc.

Rich

[Rich Euro M5]

This is what maid (sic) me reference the use of WD-40. Any other suggestions?

See my posts on Page 1.

Also, an exhaust manifold gasket leak (if I read that correctly) won't cause this problem.

I agree, typically a exhaust leak upstream of the O2 sensor introduces extra air for the O2 sensor to read. This resluts in O2 sensor sending a false lean mixture reading to the ECU, which then in turn richens the mixture. However it takes a decent sized exhaust system leak to induce a false reading by the O2 sensor.

Rich

[steve 87 535is]

"Sorry, you didn't mention it in your initial description and I, admitingly, was too lazy to read through the rest of the posts. My apologies. However, I still think you may have a vacumm leak. A vacuum leak on the exhaust side, before your O2 sensor, can cause similar problems and would contribute to your low MPG. It's also consistant with the problem occuring after your backfire. Check the exhaust manifod gasket for leaks.

And good luck. I know how annoying these things are. "

I just had my 86 535i auto back for a few days to fix the parking brake for the current owner. It has exhaust leaks at the manifold flanges to downpipes. I can feel them and of course hear them. This car does not currently or did it ever have the coasting surge that my e23 exhibits. They have about the same miles on them too (comparing throttle wear-I would assume it's pretty close). My e23 has the leaks after the 02 sensor, could that still be a factor, or am I looking in the wrong direction. I still think my throttle linkage needs just the correct adjustment (come on Spring, hurry the hell up).