Continuing the saga, after fixing special tool 9187 I resolved to finally set the mixture correctly. After another spirited drive I calibrated the Gusman's gas meter, disconnected the O2 sensor and hooked everything up. The meter read 1.4% - a bit rich. .4 to .8% is the target.

So I give a couple of spins on the mixture adjuster and no change. Hmm. Give a couple of more spins. Disconnect the gasmeter and check calibration - checks out fine. Reconnect. More spins. Check the voltage at the O2 sensor - 0.07V?? Gasmeter still reads 1.4%. So I disconnect the AC hose to simulate a large air leak. Car starts to die and the gasmeter goes up to 3.5%. Not what I expected, but at least it responds to changes in the A/F, so why no change with the mix adjustment. Reconnect the hose and take about 10 turns on the adjust screw. No change. Then things get strange.

I pull the MAF to check out the screw (again). When I get it on the bench there is no screw. It fell out in the valley. So I fish it out and take a closer look at the MAF. That's when I notice this:



Is this normal? Is this what it looks like - that someone capped the mixture adjustment off?? Anyone seen this before?

My questions are:

1) Did someone cap off the Mix adjust? Extra credit: Why?

2) Why when I introduced a massive air leak did the CO go up (rich) instead of down (lean)?

3) How come my O2 sensor is reading 0.07V (lean)? Is it toast?

4) The car runs fine so why am I doing this anyway?

This all started with the high idle which I suspect was done to pass smag as they couldnt adjust the mixture!

 

What mixture adjuster are you spinning if you have not pulled that cap off? I must be missing something here.

 

The adjustment screw (perpendicular to the cap - accessed from the bottom right in this picture) turns in and out, but it looks to me like this cap stops it from actually doing anything. What I am trying to figure out is why the cap is there. What normally goes in there if anything? Can I just pry the cap off and have another go at it? Why would it have been capped in the first place?

 

The AFM looks like L-jet, correct? IIRC, L-jet AFMs sometimes have a nut-locked adjustment screw on the outside of the casing(eg 92 Miata) - I think this is a stop for the barn door, but dont quote me. It is also normal 'anti-tamper' practice to blank off access to the mixture adjustment screw - I have seen instructions saying to drill a 2mm hole in the plug, screw in a pk screw (self tapper), and pull out the plug to gain access to screw. Thats what that cap looks like, but again dont quote me. The screw I have changed on a Discovery was a many turn one (10-20?). Also, on the discovery we used to measure volts across two of the pins(ign on but not running) on the connector. Eg to lift power (also co) by richening, we changed normal Volts from 1.5 to 1.9. You might find it useful to locate these pins by probing through the hood, and note the base setting before moving it. This applied to both L and LH I think.
good luck
jp 83 Euro S AT 48k

 

Do not pull that cap! If you look into the discharge directly behind that cap, you will see a small rectangular opening: it is the end of a cast-in air bypass loop, the other end of which is seen (open) at the lower corner of the closed air vane/flap. All the adjuster screw does is meter a certain air by-pass around the air vane ( which is articulated to a rheostat, etc.) Put the 3mm hex adjuster back in, and blow air through this passage with a straw - you get the idea.
Now, for why it doesn't respond .... need 5 min to think ...

 

The cap is still intact, but you caught me just in time :-)

Ok, I see how the air gets routed now. It seems like this little bypass channel could move a pile of air. It is possible that the screw was already completely leaned out which would explain why further turns would have no effect.

 

Phew, that was close :
Marc, too many things seem bass ackwards: When you introduce extra air by opening the oil fill cover, dipstick, whatever - the mixture goes false lean, and the CO should drop from a stoichometric ~14.7:1 A/F (lambda = 1) to show a decrease in CO of ~0.6% to 0.3% - and the O2 sensor (in open loop) should drop from ~0.5v to ~0.2v.
Conversely, if the mixture is forced rich (by pulling and plugging the vac line off the FPR), the CO should approach 2% and the O2 sensor approach ~0.9v.
If that is a CO meter, things are going in reverse; however, if the car is truely lean (with the AFM adj. screw backed out), then the O2 sensor is telling the truth, more or less.
Perhaps something to try is to reseat the adjuster screw and set it open ,say, 3 turns: install the AFM, and fire up. Put your new meter in the glove box (for a while), and lock it up! Now, with the O2 sensor disconnected (open loop), lean the mixture (counterclockwise) for the smoothest idle - then add 1/4 turn 'rich': Test drive slowly, looking for signs of surging - if so, enrich another 1/4 turn: when satisfied, reconnect the O2 sensor, break out a cold one, unlock the new meter. and have at it ....

 

Thanks Garth,

The gastester is safely locked away where it can't do any more damage. I reassembled everything, and set the mix adjustment about 5 turns out. Went out for another spirited drive to warm up the o2 sensor. When I came back, I checked the open loop voltage on the o2 sensor which was 0.41v - pretty close to optimal. The A/F meter shows proper closed loop operation.

So what have I learned?

1) The mix adjustment appears to have almost no effect. Maybe it is too subtle for me to pick up.

2) No matter what I do to the fuel system the L-Jetronic is good enough to compensate and make everything run pretty well anyway :-)

3) The voltmeter on the O2 sensor is far superior that trying to get the dopey gasmeter to work. Too bad about that - I had high hopes. It appears to be pretty temperature sensitive and need frequent re-calibration as it warms up from the exhaust stream. I may not have a properly functioning unit, as I never did understand the readouts (rich with a massive air leak??).

4) Most of my "Richness suspicions" came from the cold start exhaust smell. This is clearly the cold start injector firing and clears up pretty fast. The rest came from my gas mileage - about 15 mpg - which is more likely tied to the fact that I drive enthusiastically.

5) Ask Garth before removing any little metal caps I find :-)

Time for that cold one

 

Suspect the adjuster you moved is the idle air bypass, to set base idle or similar?
jp 83 Euro S AT 48k

 

The base idle adjuster is a screw closer to the front of the engine. That works great, one crank on that puppy and the idle came right into spec. The gizmo I am turning is back at the base of the MAF and shown on page 24-6 of the WSM.

I'm going to give Garth's technique a go and see how that works. I'll report back.

 

Mark,

I am having the same problem with mixture that you had, O2 sensor readings in the .07 range. Reading the posts I'm a bit confused about what parts you changed or what you adjusted besided the mixture screw. Any info would be really appreciated.

Phil Silano

 

Phil,

If you looked through some of the older threads, where I started with this whole thing was a high idel that would occasionally become a surging idle. I systematically went through the car replacing all the vacuum hoses, the AFM O ring and anything else I could find that might cause a vacuum leak. After checking the timing and everything else they tell you to do, I finally gave the idle screw a couple of cranks and it brought the idle nicely into spec.

I also picked up a little gizmo that gives a real time readout of the A/F mix. With the L jet you can see the O2 sensor warm up and the control unit go into a normal oscillation. Looking at that under WOT conditions, it looked like it was running a bit rich. Of course you want it rich at WOT, but it looked like it was a bit over the top. In addition at start up I got a hit of exhaust as I backed out the driveway that smelled rich. So I decided to crank on the mixture screw, with the results in the post.

The end result was that the o2 sensor was behaving correctly as was the AFM and the computer so I put the whole thing back together and sold the CO Meter on ebay. I recently passed smog with flying colors. But to answer the question, I didn't change any parts (at the end), I was only imagining I had a rich system based on some bad assumptions and some average instrumentation!

See Garth's post above to check how to verify the O2 sensor is working. There have been a bunch of other discussions on sources of running rich - leaky cold start injectors for instance.

Note, I have only played around with the L-Jet. What year is your car?

 

Marc,

Mine is an 83 S US version. Since my post I have pulled the AFM and removed and cleaned the mixture screw and air passage. Looked clean so I don't expect much help there. I have checked the O2 sensor to some extent but will do more tomorrow.

Phil