328 oil temp switch question

The labeling of the oil temp switch on the schematic in your OM (page 73 in OM 396/85) and looking at how it functions to disable the air injection system if the oil temp is "cold" (text on page 72) indicates that the switch you received is correct -- i.e., if the switch is below 33 deg C, the contacts should be closed, and if the switch is above 57 deg C, the contacts should be open (and in-between its state is dependent on where it was last). When you "jump" this switch closed, it "tells" the ECU: "I'm cold, so default to a fixed FV duty cycle".

Are you still getting an unusual FV duty cycle with the (single wire from the) O2 sensor unplugged at warm idle (when the oil temp switch is open)? If so, the water thermoswitch has the same logic (cold=closed and warm=open) so you might just try unplugging it (simulating the correct warm, open state) to see if that has any effect on FV duty cycle at warm idle.

When you plug the O2 sensor back in at warm idle and make mixture screw adjustments, does the FV duty cycle change at all?

 

Of course, you are right -- I wrongly substituted the numerical values from the water temp switch specs when typing my post.

Although closing the oil temp switch gives you a more reasonable FV duty cycle, I don't think that this absolutely "proves" that the logic should be should be the other way around. If you look at that schematic, you will also see that when the WOT switch is closed (the SR wire side in the throttle microswitch assembly "M"), this also puts a ground signal on the same line that the oil temp switch does when it is closed -- i.e., in both cases, the ECU goes "dumb" (open-loop) and just puts out a fixed FV duty cycle, and the relay "C" actuates -- which turns the air injection system "off" (by leaving the electrovalve "I" unactuated regardless of the water thermoswitch state). If you did use an oil temp switch with the logic that Carl stated, I believe the FV duty cycle would just always be stuck at the fixed value when warm -- the car could actually run OKish in that state when warm as it is similar to a K-Jet without Lambda situation (but the air injection system would have inverted behavior when the water is cold -- i.e., running when the oil is really cold and the water is cold, not running when the oil is a little warm and the water is cold, and not running when the water is warm).

Do check your water thermoswitch, as if is has failed, and is always (wrongly) closed, one symptom is that the air injection system would also remain "on" during warm running (if the oil temp switch and WOT switch are open). I know that you reported finding a bad component that was causing this problem (and your "lean" prior condition at the tailpipes), but this is maybe another clue that the water themoswitch is kaputt.

 

Almost sorry to hear your water thermoswitch is OK as that would've been the easy fix

Unfortunately, I think F mangled the test procedure, and you are getting the right result (for the given test conditions). When the injection ECU connector is unplugged, I think there is still a path from pin 5 to pin 7 thru the coil in the electrovalve and the heater in the O2 sensor as I highlighted in green in this jpeg:
Image Unavailable, Please Login

Consequently, under the test conditions given in SB 80-29 (key "off" and ECU unplugged), I think, when measuring the resistance between pin 5 and pin 7 in the harness connector, you should measure the resistance of the electrovalve coil + the resistance of the O2 heater (not infinity) when the engine is warm and the water thermoswitch is open. I just measured the resistance of my O2 heaters (a couple hours after shutoff so they are still a little warm) and got ~4 ohms. At room temp, their resistance may be even lower, but the value is already pretty small compared to the electrovalve. You might try unplugging your 2-pin O2 heater (and plugging the electrovalve back in), and repeat the 5-to-7 resistance test and see if you do get infinite ohms when warm and the water thermoswitch is open (although if the oil switch is cold/closed, I think you still wouldn't get infinite ohms, but rather electrovalve coil + relay C coil).

On a side note, I thought the electrovalve coil resistances were more on the order of ~150 ohms (but that's a vague, vague recollection and could easily be wrong), but really don't know if 44 ohms would cause a mis-detection of the water thermoswitch being open at the ECU. At warm idle (with the O2 single wire unplugged), you could just unplug the electrovalve coil (since the electrovalve is unenergized during warm running anyway) and see if it has any effect on the FV duty cycle. Other than that, I'm running out of ideas