M90 Part throttle hesitation

[Das_Prachtstrasse]

Hi all,

Around 12 months ago I purchased a 1982 South African built e12/28 M535i which had been sitting in a shed for about a decade. Over this last 12 months I've been base lining the car and it's now registered, road legal and time for some proper shake down time.

Overall it's now quite a solid driver, however it does have an annoying part throttle stumble I cannot seem to iron out. In off idle conditions, slowly easing the throttle on any more than about 15% results in the engine bogging, hesitating and at very low rpm in first gear occasionally backfiring with very little willingness to increase in revs. Increasing the throttle input to wide open immediately alleviates the issue, and it pulls well. Idle conditions are also spot on, it blows no smoke and shows good compression numbers. It will happily rev out smoothly in neutral, it's only under load where the issue presents itself.

I'm very familiar with the Motronic M30 range of engines, although have limited experience with the nuance of the L-Jet M90. In attempting to resolve this and a range of other running issues I've done the following using only new and genuine factory replacement parts:

-Checked and reset the throttle butterfly to the factory specification including the necessary TPS adjustment, Aux air valve adjustment and all inlet and associated vacuum hose replacement.
-Confirmed the brake circuit vacuum lines/booster check valve etc is all leak free
-Confirmed the vacuum advance/retard unit is operating correctly
-Confirmed the cold start injector is operating correctly
-Confirmed static engine/ignition timing is correct
-Confirmed fuel pressure is correct, replaced filter and soft lines
-Replaced the CTS
-Replaced the AFM
-Valve adjustment and cover gasket replaced

The previous owner replaced the leads with genuine BMW and fit some generic replacement plugs to get it to fire up - I've since gapped them correctly and although new W8DC's are now on their way I don't expect plugs will resolve the issue. The coil looks to be original and I'm having a hard time finding any specification on it's correct resistance values to test if it's actually ok. I would prefer to test it before just throwing another one in there for the sake of it, but am suspicious. The ballast resistor is within spec. Distributor points all ok.

Another interesting quirk is a whistle from the engine bay under part throttle that almost coincides with the stumble, but not always. Naturally this leads me to believe there is a vacuum leak somewhere that is not present at idle - I've tried misting prepsol around all the usual suspects at idle with no obvious increase in rpm. A smoke test is on the to-do list.

Are there any other things worth looking into that I've overlooked in the interim? Are the ECU's generally pretty solid? It really feels ignition based to me and although the whistle is peculiar I would expect an off idle vacuum leak to present itself at WOT too.

I've spent a bit of time searching here and over on bigcoupe which has resolved a range of other M90 issues I didn't even know it had, but this one is persistent so any further advice would be appreciated.




Christian

[528i-1981]

First off, really nice car dude.

It does suggest a vacuum leak to me, also. Did you check for leaks around the intake runners? The lower plenum is easy to miss with the starter fluid test. I assume you've carefully checked the AFM boot that connects to the throttle body. Cracks in the bellows can cause intermittent problems. Also you said you checked the brake vacuum line check valve for leaks, I assume you've checked it's operation.

The only other thing that occurs to me is to check the idle switch on the throttle body linkage. I don't know about M90s, but the L-jet on the M30 has two contact switches, one for idle and one for WOT. Assuming the M90 is the same, I'm thinking your idle switch might be faulty so that the ECU might not always know it's off-idle (unless its WOT). On later L-jets there's a coast-down cutoff that could be causing your stumble if the ECU sees the idle switch closing.

Good luck,
Eric

[Das_Prachtstrasse]

First off, really nice car dude.

It does suggest a vacuum leak to me, also. Did you check for leaks around the intake runners? The lower plenum is easy to miss with the starter fluid test. I assume you've carefully checked the AFM boot that connects to the throttle body. Cracks in the bellows can cause intermittent problems. Also you said you checked the brake vacuum line check valve for leaks, I assume you've checked it's operation.

The only other thing that occurs to me is to check the idle switch on the throttle body linkage. I don't know about M90s, but the L-jet on the M30 has two contact switches, one for idle and one for WOT. Assuming the M90 is the same, I'm thinking your idle switch might be faulty so that the ECU might not always know it's off-idle (unless its WOT). On later L-jets there's a coast-down cutoff that could be causing your stumble if the ECU sees the idle switch closing.

Good luck,
Eric

Thanks. I have been extremely thorough with misting for vacuum leaks, particularly around the intake runners and as mentioned, all rubber intake boots are new. I removed the cover from the TPS, cleaned and confirmed its operation is as it should be. It’s calibrated to the throttle position perfectly. If only it were that simple!

[528i-1981]

Lock was having power loss symptoms with his L-jet under various load conditions. His solution was replacing it with Motronic! Hopefully someone else can come up with something to help you. The only other thing I can think of would be the AFM vane, but you've replaced the AFM (and assume from your thoroughness that you've verified it's in spec) and have the same problem.

Sorry for another easy suggestion, but have you checked the vacuum lines to the fuel pressure valve and its operation? You mentioned fuel pressure was good, but perhaps you're losing pressure at the rail with the change of manifold vacuum on decel.

-Eric

[PatinaBeforePolish]

Whistling sounds are almost always a vacuum leak. Next step should be a smoke test.

[maybeillbuyit]

I agree vacuum could be it. What happens when you pull the dipstick while idling? I would also suggest getting a spare ecu for that engine. It should be the black one ,114. Hard to find as it only came on the M90. I think the ecu's can always be suspect on these cars not just the M90. For $100-150 having a spare is a good idea in my opinion.
On another note, Amazing car!! My dream car there. Do you have any pics of when you acquired it? Can we see more pics of the interior etc?

[Das_Prachtstrasse]

I built a smoke tester and have found a leak from the throttle butterfly pivot shaft. It leaks out through the TPS on one side, and the linkage on the other and is sufficient enough to probably be the cause of the whistle. It is more pronounced with the butterfly part open, so I would guess the shaft is worn mainly in the part throttle region which would be consistent with typical use.

I have a spare TB off a later M30 I can dissect and will determine whether it's possible to turn up some bushes or retrofit a seal or similar before trying to source a new throttle body which of course is NLA. I suspect these things probably leaked from new given the design and am skeptical but hopeful that this is the cause of the stumble. The whistle seems likely, maybe a temporary dab of grease on either side will determine that.

Thanks for the suggestions and compliments, I'll share some more pictures of the car over the next few days.

[maybeillbuyit]

Its interesting, I wonder how many of us have this issue without the whistle and have no idea. I have seen that Porsche 944s actually have rubber bushings to seal the TB butterfly pivot which I thought was a good idea. ( don't own one but maybe someday).
Looking forward to pics. Are those 16 In Mahles?

[Das_Prachtstrasse]

I disassembled the throttle body and found shaft seals on either end that are both worn, and one is split. BMW must've done away with these on all later throttle bodies, I've never seen them before. These are NLA (13541257265) so I 3D printed some rubber replacements.




I then re-assembled everything, reset the throttle plate and TPS again and there is still no change to the running condition; the whistle and stumble are both still present. Re-smoke testing shows zero leaks anywhere so at least my seals are doing their job.

The only components I've not tested now are the injectors and the coil. If someone can provide the correct resistance values across the poles of the coil that'd be great. I'll save pulling and flow benching the injectors until I can confirm the coil operation as that's a nuisance of a task!

Christian

[maybeillbuyit]

I'm surprised to hear there were rubber seals in the the TB. I just took mine completely apart and never saw anything like that. That I can remember anyways I wish I could just print a part when I need it!
I'll try to measure the resistance of my coil tomorrow. I think I tested it previously and confirmed the numbers I was seeing from somewhere , Maybe it was in the FAQs here.

Can you post a video of the whistling? Maybe that wouldn't work but maybe worth a try.

[tlake]

The exhaust manifold gasket can be a source of leaks too. Had an engine die completely because of that.