AFM Vane Details

[Captain Obvious]

And that's why I was looking for a ballpark number. My mixture screw was about fifteen turns out, and it seemed way too far.

As for the AFM adjustments while running... I was doing my check at idle with the TPS idle switch closed so there shouldn't be any conflicting signals. My engine seemed to like more fuel at idle even with the idle enrichment switch closed. Might be related to a mixture screw that is fifteen turns out.

[FastWoman]

Obvious, it's interesting the repair made no difference. It made a huge difference in my '75, but again, the flap was bent in a different sort of way that I'm at a loss to describe here.

I might have recommended 5 turns open on the idle mix screw, but if so, it was just something I picked up elsewhere. I don't really find this adjustment makes much difference on my engine, so I have it set at 5 turns.

You've correctly determined that your engine is running lean. Now the question is "why?" I'd do the beer-can calibration on the AFM's clockspring to the factory tension, and make certain it is good electrically. I think I recall your doing the yogurt cup test to ensure that your intake is tight. I'd check all the plugs to make certain they're reading the same. Partially clogged injectors are another possible factor, and they're relatively cheap to replace (about $150). Of course check valve and ignition timing, as well as valve lash. Then after you've done all that, if you're still running lean, you'll find there's not much left to consider. Then it would be time to trick the ECU into running richer by adding resistance to the coolant temp sensor circuit.

[Zed Head]

On the subject of AFM tuning: I found this web page a little while ago while looking for injector information. It's from the days of Usenet, the very earliest days of world-wide web growth. Somebody collected a bunch of interesting posts and put them in a modern day web page format.

This page offers one modification method for someone with a lot of free time. Go to the Index link at the top of the page and browse and you'll see that people were asking the same questions about Z cars 20 years ago.

http://yarchive.net/car/z_airflow_sensors.html

[240260280z]

John De Armond is a genius!

[FastWoman]

Some nice ideas for tweaks in that article, Zed, but I strongly believe the best approach is the coolant temp circuit. That's because this seems to be the primary determinant of injector pulse width, and the temp correction is applied uniformly through all operating conditions. Corrections made any other way are not going to be applied uniformly. For instance, if the clock spring is loosened, the AFM will peg out prematurely, and there will be a lean condition beyond that. If the wiper is turned, it will overshoot, perhaps only resulting in a premature pegging as with the spring adjustment. If the linearity of the trace is tweaked, per De Armond's clever method, the correction is only locally applied.

Of course the thesis that the CTS is the best place to make the AFR adjustment is based on the assumption that the ECU has drifted in its fundamental operational characteristics. In my experience with 70's era electronics, this is a very reasonable assumption -- indeed an expectation.

[Zed Head]

I was just showing some old discussions for the sake of historical perspective. I think that Blue may be right, this guy might be a genius. He was doing some fantastic stuff back before the Google and the WORLD WIDE web were available. We have it easy.

The L-Jet system is primitive, by definition, and there are many different ways to get it to do what you want it to do. I have no idea if the ECU components have aged, causing a drift to lean mixtures, or if the lower energy content of 10% ethanol blends behaves like a lean mixture, or if the lean behavior is inherent to the tune the engineers gave it in the late 70s (the test reports from the car mags reported a lean surge behavior, brand new, using late 70s quality gas). It does seem that the water temp resistance is the popular modification, considering the number of resistors that pop up here and there when current (no pun intended, but still recognized) owners crack open their wiring.

It's all good! The more I learn the more I appreciate what the Megasquirt guys are doing. Tuning via computer using wide-band O2 measurement would be so much easier.