Hvithaien

Hi!

My 84 928S (US L-jet) last year developed a bit of rough idle and a small stutter when accelerating. All this went away when I disconnected the O2 sensor. I later changed the O2 sensor for a brand new one, but the problem returned when this was connected. Since then I have been driving with the O2 sensor unplugged with no problems.
However, I can't stop thinking about the fact that something must be wrong in the L-jet regulating loop that the sensor normally is a part of.
Any thoughts what that might be?

Hvithaien

James Bailey

Most likely an air leak (unmetered air )on one or more of the rubber intake boots causing a lean misfire and confusing the O2 sensor and brain.

Five-8

My 82 (US L-jet) has much the same issue, so i am curious also to hear what the possibilities here are. Running with the O2 sensor disconnected here also...car seems to run fine...but i am pretty sure its far too rich and probably not running at its optimum.
Its on my list of issues to address, but i am still knocking through some other stuff first,so havent really dug into this one yet.

Thanks,

Hvithaien

Forgot to mention that all rubber in the intake system is new, and all vacuum lines are leak checked.

-Hvithaien

IcemanG17

I'm also leaning towards a vac leak....here is why, by disconnecting the O2 you make the car run VERY rich....which would compensate for the leak..... Try the "brake cleaner" test....spray brake cleaner around the intake-spiders...while the engine is running, if the rpm climbs....there is your leak. When I put intake spiders from one engine on another, there was massive leaks....turns out the spiders had slightly warped to fit the heads from the old engine.....a quick trip to the belt sander fixed it fast!

Keep in mind the L jet system is essentially analog.....so it is VERY sensitive to sensors being even slightly out of normal range. If the Temp II is off or the Temp I (inside AFM) it can be the difference from a car that won't run and runs perfect....my temp I was only a couple hundred ohms off....car barely ran...installed a rebuilt AFM..car ran PERFECT and still does

James Bailey

If there are no false air leaks then the airflow meter becomes the next likely suspect....the internal contact slides along a printed circuit foil and that varies the signal to the brain. The arm sometimes get bent (by mechanics adjusting it ) the contact foil gets worn and has gaps. Sop sends the brain the wrong signal. What ever is wrong is causing the car to run lean and the unplugged O2 sensor makes it default to over rich.

GregBBRD

These "very lean" 928s frequently ran much better with the O2 sensor disconnected, so if you don't find a vacuum leak, don't be overly concerned. [They also (generally) run better with more ignition timing and by "loosening" the spring on the barn door.]

There are two micro switches on the throttle assembly. One is triggered at idle and the other is triggered at about 3/4 throttle. The "idle" switch needs to work properly with the O2 sensor hooked up....it "puts" the engine into the "idle" map of the computer. If it gets stuck closed (common) the O2 sensor may think you are at idle....while you are trying to accelerate. Or, if it doesn't close, the car will be very confused about idle.

If the WOT switch doesn't work, the computer will continue to make the car run in closed loop, instead of allowing the engine to go into open loop, past 3/4 throttle.

James Bailey

Good points from someone who has fixed more 928s Than anyone else on earth....

IcemanG17

x2

Five-8

This is very helpful, thanks Fellas!

Cheers,

James Bailey

The 1980-84 USA injection was designed with one thing in mind...low emissions smog compliance. Performance was irrelevant as the national speed limit was 55 MPH and the speedo maxed out at 85 MPH. Kibort and others have proven that mixing in some Euro S parts and exhaust improvements can make a HUGE increase in power. Note the 1980 Corvette LG4 V-8 Ca spec made a total of 180 HP !!!! Those were dark days for automobile enthusiasts....

mark kibort

what the o2 sensor does,is make the system hunt for 14.7 mixture. (optimum for emisssios) so the idle will wonder a little, but it works and the engine runs cleaner. pop off the o2 sensor, and the idle kicks up a little (becuase now you are in rich land) and the idle is very smooth. the S4 does this too. I think anderson doesnt even run the o2 sensor on his car. so the car is fat during idle, but its a race car so who cares!
anyway, too lean in part throttle or idle shoud be just fine, especially if things are working well. the only time it will matter, is as Greg says, near full throttle and beyond. the microswitch makes it go open loop at near 75% throttle and WOT.and there should be no hunting of the fuel mixture. if it doesnt work, the system willl continue to hunt and you will go way lean at WOT, and the power will be very low.
the system works amazingly welll when working and all at the 5% of HP cost of the small opening of the AFM. its like a carburetor with no mess.
how amazing was it for me to get a 170rwhp 928 '84, and bolt on a bunch of stuff and get it up to 290rwhp with NO mods to anything on the injection or electrical systems!! (just a rising rate fuel regulator, which only really just tuned up stuff, from 13.5:1 to 12.5:1 vs not even touching anything)

GregBBRD

Exactly. The '80-'84 cars respond very well to increased fuel...regardless if you unplug the O2 sensor, loosen up the barn door, or add more fuel pressure.

GregBBRD

Not every engine "likes" 14.7.....it's not a "magical" number that every engine design works well with. It's taken the automotive companies many years to come up with engines that like being super lean and make lots of power, at the same time. Some engines are happy at 15.1. Some will barely idle at 12.5. It's all about design and efficiency.

In my experience, the '1980-'1984 928 engine design performed very poorly when the mixture was this lean. They surged driving down the freeways, with the O2 sensors working, they had idle issues, etc.

Like I said, give them some additional fuel and then some additional ignition timing and they were much happier.

mark kibort

I was using 14.7 as the ball park target number where the engine sits after an average of rich and lean hunting. if it averages more than that, yes, it can be more stable. generally, no internal combustion engine makes best power anywhere near Stoich . they can run smoothly up toward 17:1, and just before they start missing, that makes the best fuel economy without hurting anything. max CHT, and EGTs happen at 14.7 with gas engines, and they dont make as much power there, right? certainly to go to the lean side, you have to pull enough fuel out to get out of dangerland, and that usually starts in the 16--17:1 range. cruise, is an entirely different discussion.