Check engine light doesn't work - can't pass emissions test | Ford Explorer Forums - Serious Explorations

  • Register Today It's free!

Check engine light doesn't work - can't pass emissions test

Austin Healey

Member
Joined
October 8, 2016
Messages
42
Reaction score
12
Location
Northern Virginia
Year, Model & Trim Level
1997 Ford Explorer XLT
Took in my '97 Explorer for emissions test the other day, and it flunked because the check engine light doesn't come on in self-test mode -- ie when the ignition is first turned on.

Here's what I've checked so far, based on other advice -- some from this forum, some scrounged elsewhere:

--gained access to instrument panel, checked bulb -- it's good (ie has continuity and resistance).
--swapped bulb and housing/connector w/ another one from the panel to see if it would work. No.
--plugged in code scanner to make sure OBD system was working, had power, etc. Yes.
--removed connector to main engine computer on firewall, checked to make sure pins and connectors appeared intact, clean, etc. No issues, no rust. I've owned the truck for approx 3 years, and this was the first time I unplugged the computer, so nobody's been messing with it. No change.
--also removed the big wiring connector on driver side of engine that connects to a bunch of engine sensors & misc. Checked pins, cleaned lightly w/ sandpaper, though they looked fine. No change.
--just for fun, cleaned the ground terminal for the computer connector, on firewall. Looked good, had no rust. No change.

The only other thing I can think of is to take apart the instrument panel and start tracing circuits to see if there's a poor soldered connection, etc. Any other suggestions?

I find it difficult to believe that the computer decided to fail ONLY on the circuit for the check engine light. All other warning lamps, indicators work fine, and the truck drives w/ no issues.

Thanks for any help.
 



Join the Elite Explorers for $20 each year.
Elite Explorer members see no advertisements, no banner ads, no double underlined links,.
Add an avatar, upload photo attachments, and more!
.





Aside from the issue with the light bulb in the instrument panel not working, the OBD2 scanner should tell you whether the check engine light status in the PCM is set to "check engine light on" or to "check engine light off" or "MIL on/off"
A good scanner should also tell you which of the specific emissions tests were successful and which ones failed or did not complete (and also which of the tests are not available in your vehicle - this can be tricky and misleading on some scanners, if they list all unavailable tests as "not completed").

Here's the Ford drive cycle procedure with all the specific requirements for each of the different tests to start and to complete:
Ford Motor Company Driving Cycle (obdii.com)

Hope that helps resolve the PCM side of things.
About the problem with the light bulb in the instrument panel not illuminating, I'd say you're right on track with that so far.

Good luck.
 






Took in my '97 Explorer for emissions test the other day, and it flunked because the check engine light doesn't come on in self-test mode -- ie when the ignition is first turned on.

Here's what I've checked so far, based on other advice -- some from this forum, some scrounged elsewhere:

--gained access to instrument panel, checked bulb -- it's good (ie has continuity and resistance).
--swapped bulb and housing/connector w/ another one from the panel to see if it would work. No.
--plugged in code scanner to make sure OBD system was working, had power, etc. Yes.
--removed connector to main engine computer on firewall, checked to make sure pins and connectors appeared intact, clean, etc. No issues, no rust. I've owned the truck for approx 3 years, and this was the first time I unplugged the computer, so nobody's been messing with it. No change.
--also removed the big wiring connector on driver side of engine that connects to a bunch of engine sensors & misc. Checked pins, cleaned lightly w/ sandpaper, though they looked fine. No change.
--just for fun, cleaned the ground terminal for the computer connector, on firewall. Looked good, had no rust. No change.

The only other thing I can think of is to take apart the instrument panel and start tracing circuits to see if there's a poor soldered connection, etc. Any other suggestions?

I find it difficult to believe that the computer decided to fail ONLY on the circuit for the check engine light. All other warning lamps, indicators work fine, and the truck drives w/ no issues.

Thanks for any help.
Lets start here.

Do you have continuity (1-2 ohms) between both ends of circuit 658? Computer to dash connector?

When you turn the key, and connect a multimeter between pin 9 (instrument cluster) and ground, do you see voltage? I don't know if it is low current circuit, the instrument cluster may switch on the light. But there should be some voltage with a digital multimeter.

If those tests fail, then the internal driver in the computer could be bad. If you just need it to turn on, and you know the monitors are set for their test, you could probably share another wire (like seatbelt) that proves out and lights it at start. The MIL should never normally come on anyway after prove out. (at your own risk of course).

1611415077049.png


1611415522460.png
 












Did some more diagnosis and re Circuit 658, as shown above in the diagram for the connector to the instrument cluster: it doesn't exist (at least on my truck). In that space, the connector lacks both the metallic female fitting and any wire to it. I poked around behind the cluster to see if there were any signs of a loose or cut wire ... and there weren't. Also, if you simply wanted to stop the MIL (check engine light) from coming on, you'd simply cut the wire, not remove the metallic female fitting.

I did find a slightly different pinout diagram for the PCM module for the 1997, specifically for the 4.0 liter OHC engine:

1611783617032.png


However, it's quite similar, and still shows the same #2 pin and pink/light green wire for that circuit.

I did find one reference online to there being late and early versions of the instrument cluster in 1997, with different circuit numbers for the MIL light.

So ... not much progress. Any other ideas?
 












Back
Top