Figuring a new AMM could be the answer to my crappy running I installed one from fcp and gave it a run.

The FAQ procedure for adjusting the AMM calls for building a small circuit (which I did), hooking it up to the CO test point and 12+ and while running the engine, look for a blinking LED while turning the AMM screw.

Well, soldering the circuit as shown in the figure left the LED OFF regardless of the number of turns in either direction.... reversing the wiring on the test LED circuit left the LED ON ALL the time.

Can anyone shed some light on this procedure? Got any other ideas? I'd really appreciate it.

Thanks.

Mike

I have used a LED but you can also just hang a meter on the test point. Analog meter works best.
--
David Hunter

I have a several comments.

First, I have built two of these simple test diodes circuits, and both work fine.

Second, it takes a lot of turning to make much adjustment on the AMM.

Third, I have had an AMM that would not adjust no matter how much the screw was turned.
--
john

John,

thanks for the reply.

FIrst off, just how much turning are we talking about? 10? 50? I am at the 50 point in each direction... no change.

second... your circuits... was the resistance EXACTLY 750ohms? I am assuming this is just to protect the LED from popping.

third, is there another port I can use instead of the test port?

Mike

The "Volvo Problem Solver" book says there are no fixed stops on the AMM screw, but that 15 turns takes you the full range from max lean to max rich, over a resistance of 1,000 ohms. (Measure between terminals 6 and 2 with harness unplugged. Terminal 6 is closest to the pot screw).

It also says you can back-probe pin 6 with harness connected and Key ON, and should see 0.0v (lean) to 2.6v (rich) over the adjustment range.

As I recall (from somewhere, Art?) setting the AMM pot to a nominal 360 ohms or so is a good ballpark figure. That's where I left it on my daughter's 240, which went back to Mass and passed emissions with excellent numbers.
--
Bruce Young
'93 940-NA (current) — 240s (one V8) — 140s — 122s — since '63.

It seems like the Radio Shack resistors were 680 ohm or so. That resistance may not be perfect, but it works fine.

Fifty turns and no change does not sound good. Try another AMM.

Instead of using the test port, you can put a multimeter on the O2 sensor output. You should get an oscilliting voltage from about 0.2 to 0.7 volts. If you don't, you have either a bad O2 sensor or a bad AMM.
--
john

Where in the FAQ was the original incorrect note?

http://www.brickboard.com/FAQ/700-900/EngineFuelinjection.htm#SettingBaseIdle

The test point on my '87 744T doesn't work either; I use a 'scope on the oxygen sensor lead and set the crossing for 450 mv.

why 450mv?

but more importantly, how do you get the voltage reading on the O2 sensor to NOT change? THe O2 voltage changes as the engine runs...

I don't get it to not change.
As John says, it should change from ~200mv to ~700mv; the average should be 450mv, which is the stoichiometric point on the oxygen sensor.
On 2 out of 3 of my LH 2.2 cars, the diode setup worked fine; it just doesn't on my '87 740T.

The LED adjustment procedure presumes everything affecting mixture is in the ball park to begin with. Stuff like vacuum leaks, fuel pressure, injectors will throw the mixture off more than that pot will correct. If the O2 sensor isn't working, e.g. clogged with soot from rich running, the LED test point will not switch. It is actually monitoring the output of the first circuit inside the ECU that handles the O2 sensor signal, so the flashing indicates the switching between lean and rich that occurs while the ECU is maintaining the correct mixture. Because AMMs are slightly different in response, an adjustment is given on the AMM to adjust the ECU to match it up until LH2.4 when the ECU became automatically adaptable to the slight variations in sensor output.

A better procedure to use in replacing AMMs would have you measuring the voltage on the yellow lead (LH2.2) before you adjust or disconnect the old one. Then you could at least put the pot back where it was if needed or set the new one to the same value before installing it. You could measure the resistance, but it is much easier to backprobe the voltage as you set the pot-- it varies from zero to about 2.7V-- then you don't need to guess where you ran out of adjustment. The ends are very hard to feel.

Check the voltage at the oxygen sensor; you'll probably find it is not oscillating between .1 and .9V. It must do this for the LED to flash. If not, it doesn't definitely mean the sensor is not working; it could just be reporting what it actually sees, either too rich or too lean.
--
Art Benstein near Baltimore

1. there are no air leaks

2. the fuel injectors, FPR, fuel pumps, O2 sensor, ECT sensor, knock sensor, TPS sensor, throttle body cleaned and just about anything else that can be replaced has been replaced ALL BOSCH PARTS. oh yeah, and the fuel pressure is fine.

3. A full diagnosis has been performed on all components through the ECU... all check out fine. Engine grounds are good.

4. the O2 sensor is fluctuating just fine between 0.1 and 0.9V.

Next?

Next?

Obviously, the ECU. What else is left?? I apologize for not following or recalling the prequels in the saga.

If your O2 sensor is generating a fairly even duty cycle between .1 and .9V, the system is in closed loop and the mixture is correct. The test point, if you are on it (pin 22 on the -544 ECU) is an open collector output protected by a 511 ohm series resistor-- pretty hard to damage externally. You don't really need the 750 ohm resistor, and its value certainly isn't critical to the function; just protects the LED if you drop the ECU end of the probe on a ground. The LED just shows you what the ECU sees as crossings of stoich point (like mcduck says about 450mV) with a set of comparators a bit on either side of that midpoint. I hope you find this helpful, 740ATL. Much easier to follow the light and guess duty cycle than observe a bunch of flashing numbers on a DMM.
--
Art Benstein near Baltimore