Wiring/O2 sensors (very long) persistent problem

Important stuff is in bold.

1996 (note: OBDII) 2WD 4.0L Auto
intake manifold heat shield
cool air intake
62mm throttlebody

Problem: O2 sensors throwing CEL & codes ("voltage too high" for both the upstream & downstream, and for the sensor heating circuit). Have been driving 8 months w/this problem, running a little rich, but now emissions inspections are due for my Jeep & I need to get this fixed once & for all.

This began when I installed a new upstream Denso brand sensor 8 months ago. I did not have a CEL on before installing the new sensor, but I wanted better gas mileage than 16mpg (city), & the original sensor had 100 000 miles on it. I should have left it alone. But I installed the Denso & I reset the computer by disconnecting the battery & turning the ignition to "on" (not "Start," but one step before). Everything was fine for a while; the CEL came on after about 130 miles of driving. I thought to myself, "I must've bought the wrong brand sensor." Since the CEL had not come on with the original sensor, I put the original sensor back in in place of the Denso. I cleared the computer memory again.

30-35 miles down the road, the CEL comes back on again. Jeep developed a bit of hesistation (on acceleration) after both cold & hot startups. It seemed like the computer was restricting fuel flow to the injectors, presumably b/c of the O2 sensor readings. The hesitation on accel only lasts a few minutes, then goes away, I presume because the computer relapses into default mode at that point. Unplugging the sensors made the hesitation problem go away.

I tested the sensors, the sensors seemed fine, & read within normal operating ranges. I bought a downstream Denso sensor, & reinstalled both Denso sensors, upstream & downstream. Cleared the codes/computer memory, & drove it. 30 -35 miles later, CEL came back on!

Took it to a mechanic. Mechanic said maybe there was just dirt in my intake/exhaust or I got a tank of bad gas. Mechanic cleared the codes & reset the computer. Used some fuel injector cleaner. Cleaned the TB for the second time in 4 months. 30-35 miles later, the CEL came back on anyway. And it was giving me that hesitation again, on acceleration.

Unplugged both sensors from the wiring harness. Hesitation on acceleration goes away, Jeep runs perfectly, though rich (judging by the deposits left on my spark plugs. Gas mileage drops to 15mpg. Drove this way for 6 months.

I tested the output voltage from the wiring harness & compared it to a 1994 XJ Cherokee Factory Service Manual (FSM) which I had access to. The voltage coming from the harness seemed to read 1.5-2 volts higher than it should, but then the '94 Cherokees were OBDI & the '96 Cherokees are OBDII, so maybe that accounts for the difference in voltage?

Took it to another (different) mechanic. They tested the sensors & ran a diagnostic, decided I needed a new catalytic converter. I disagreed, but, I let them do it. 40 miles down the road, the CEL comes back on. I take it back to the mechanic, & they run a different test (that they didn't do before, the little jerks) & tell me that by grounding the wires at such & such, & doing such & such, they decided that the Jeep just doesn't like Denso sensors, & I should buy OEM sensors from the dealership.

Fine. I buy $220.00 worth of OEM sensors (both upstream & downstream) from the dealership, & install them myself, & clear the codes by the method of disconnecting the battery & turning the ignition key to "on." CEL stays off for nearly 90 miles, then comes back on again! Hesitation on acceleration reappears again, too.

That's where I'm at. I can try one more time to go to a mechanic, except now I'm virtually out of money.

Possible causes not yet investigated: two people (one of them rebuilds CJs for a hobby) have suggested to me (1) that the voltage regulator might be the culprit--I haven't played with it yet--but it seems to me that I would have other symptoms (besides the O2 sensors) if the voltage was out of spec--am I right or wrong here?--& (2) there might be a short somewhere in the wiring--I favor this possibility. I followed the wires as far back (from the sensors) to where they are taped together with the fuel injector wires. I chose not to unwrap (untape) the injector wires b/c I figured it wouldn't be worth all the work (that was several months ago, but now I'm thinking I might be willing to take all the wiring apart even if it takes me a week to get it back together).

Sorry for the long post. Any suggestions/ideas/comments/questions/complaints?

and I DID check all the fuses & relays (though that was several months ago, & I haven't re-checked them since)

For your 96 XJ, the O2 sensors are wired like this:
Heater + from the automatic shut down relay
Heater - from ground

Sensor + from ECU
Sensor - from Senor groud (this ground came from ECU but is wired separated and is the reference ground voltage for diferent sensor, like the TPS) try checking for voltage difference between sensor ground and battery ground.

What are the codes?

Panama XJ said:

try checking for voltage difference between sensor ground and battery ground.

Click to expand...

I will do so, & I will report back w/my findings. It might be a few hours, or it might tomorrow before I get a chance to check voltage.

langer1 said:

What are the codes?

Click to expand...

"PO132 O2 Sensor Circuit High Voltage (Bank 1, Sensor 1)" (upstream, I presume)
"PO135 O2 Sensor Heater Circuit Malfunction (Bank 1, Sensor 1)"
"PO138 O2 Sensor Circuit High Voltage (Bank 1, Sensor 2)"

Sorry, I would've posted those first time around, but I was at work & the printout was at home. The only variation has been that, when I changed the sensors, for a brief time only one sensor (the upstream one) plus the heater circuit would cause the CEL, but after driving a few more miles the code for the downstream sensor would add itself to the list, too. Whatever's wrong seems to affect the whole circuit.

HoratioTheJeep said:

"PO132 O2 Sensor Circuit High Voltage (Bank 1, Sensor 1)" (upstream, I presume)
"PO135 O2 Sensor Heater Circuit Malfunction (Bank 1, Sensor 1)"
"PO138 O2 Sensor Circuit High Voltage (Bank 1, Sensor 2)"

Sorry, I would've posted those first time around, but I was at work & the printout was at home. The only variation has been that, when I changed the sensors, for a brief time only one sensor (the upstream one) plus the heater circuit would cause the CEL, but after driving a few more miles the code for the downstream sensor would add itself to the list, too. Whatever's wrong seems to affect the whole circuit.

Click to expand...

You need to fix P0135 first.

I greatly suspect that, once you fix the heater fault, the other two issues will self-correct. I don't have a 96FSM, but I've got a 97, so here goes...

Again - this is on the 1997 and not the 1996, but they should be similar.

The HEGO sensor is a four-wire unit with "divorced" ground to the ECM. The wiring for the sensors is as follows:
UPSTREAM
BRN/YEL - Sense Ground (ECM Pin A4 - both sensors)
BLK/DKG - Sense Signal (varies with HEGO output) (ECM Pin A24)
DKG/WHT - Heater Source (+12VDC from PDC Fuze F23, 15A BLUE, via ASD (Auto ShutDown) relay.)
BLK - Heater Ground (Engine Block - near starter.)

DOWNSTREAM
BRN/YEL - Sense Ground (ECM Pin A4 - both sensors)
TAN/BLK - Sense Signal (varies with HEGO output) (ECM Pin A25)
DKG/WHT - Heater Source (+12VDC from PDC Fuze F23, 15A BLUE, via ASD (Auto ShutDown) relay.)
BLK - Heater Ground (Engine Block - near starter.)

Since you're getting a code for "heater fault," that tells me that you aren't getting voltage either TO or THROUGH the heater. Check for battery voltage at one terminal of the HEGO harness (vehicle side,) there should be ONE AND ONLY ONE pin that reads a full 12VDC with the ASD relay energised or bypassed (check for voltage at Fuze F23 - or equivalent for 1996. I'm not finding a name for the fuze, so it may take a little digging.)

If you have +12VDC at the heater supply pin, the next step is to check the heater grounds. The heater grounds are shown as being somewhat above the starter (there is a batch of grounds probably 8" or so on the side of the block above the starter - them's those.) Find continuity (Key OFF) from a single pin to the engine ground. The sense ground should be isolated through the ECM - but if you pick up two grounds there you're probably OK. If you'd like to manually isolate the sense ground, pull ECM connector C3 (Grey) at the PCM.

Check that out, and let us know what you find.

Once again, PLEASE NOTE - this is data from the 1997 FSM, and you may note variations between the 1997 and your 1996. However, this should get you pointed in the right direction, and I included as much detail as possible in case they did match up (and if they did, would you please let me know?)

5-90

5-90 said:

The wiring for the sensors is as follows:
UPSTREAM
BRN/YEL - Sense Ground (ECM Pin A4 - both sensors)
BLK/DKG - Sense Signal (varies with HEGO output) (ECM Pin A24)
DKG/WHT - Heater Source (+12VDC from PDC Fuze F23, 15A BLUE, via ASD (Auto ShutDown) relay.)
BLK - Heater Ground (Engine Block - near starter.)

Click to expand...

As you guessed, there is some difference in wire colors:
UPSTREAM
Heater Source is LtGreen/Orange
Sensor ground is Black/Blue
Black/Green - don't know which it is
Black/white - don't know

My downstream wires are too grimy to tell what colors they are. I may clean them & get back to you on that.

5-90 said:

Check for battery voltage at one terminal of the HEGO harness (vehicle side,) there should be ONE AND ONLY ONE pin that reads a full 12VDC

Click to expand...

OK. The upstream LtGreen/Orange wire reads a very steady 12V DC. HOWEVER, the downstream sensor reads anywhere from 13-16V DC, but never gets as low as 12V. What does that mean?

5-90 said:

(check for voltage at Fuze F23 - or equivalent for 1996. I'm not finding a name for the fuze, so it may take a little digging.)

Click to expand...

I couldn't find the exact fuse, but there were two 15A, blue fuses. I checked voltage at both, & both were a steady 11V DC (why 11 and not 12V?)

5-90 said:

If you have +12VDC at the heater supply pin, the next step is to check the heater grounds. The heater grounds are shown as being somewhat above the starter (there is a batch of grounds probably 8" or so on the side of the block above the starter - them's those.) Find continuity (Key OFF) from a single pin to the engine ground.

Click to expand...

I found the grounds, & the multimeter gave me a reading of 1-2 Ohms when contacting the sensor ground on the harness side (the Black/Blue wire). The 1994 FSM says it should read 5 Ohms or lower, so I guess it's in spec.

Summary: I'm getting 12V at the heater source pin for the upstream sensor, but 13-16V at the heater source pin for the downstream sensor. I don't know what to make of that. Is the high voltage from the harness to the downstream sensor what's causing my problem?

BTW, thanks, 5-90, the info is great--the more detail the better IMO (as you can tell from my huge description of the problem, for which I again apologize).

Panama XJ said:

try checking for voltage difference between sensor ground and battery ground.

Click to expand...

As near as I can tell, there's no voltage difference regardless of where I ground the black multimeter lead, whether sensor ground or battery ground (well, this goes for the upstream sensor, at least; I'm not too sure about the downstream sensor, partly b/c I can't tell which wire is which while creeping around under there). Thanks, though.

HoratioTheJeep said:

As near as I can tell, there's no voltage difference regardless of where I ground the black multimeter lead, whether sensor ground or battery ground (well, this goes for the upstream sensor, at least; I'm not too sure about the downstream sensor, partly b/c I can't tell which wire is which while creeping around under there). Thanks, though.

Click to expand...

Well it looks like a bad heater or plug on the up-stream sensor. Don't worry about the down stream sensor for now, it don't control anything.
The heater can be tested with your ohm meter, you should read 7-9 Ohms between pins a and b.

HoratioTheJeep said:

OK. The upstream LtGreen/Orange wire reads a very steady 12V DC. HOWEVER, the downstream sensor reads anywhere from 13-16V DC, but never gets as low as 12V. What does that mean?

Click to expand...

That's something I'd take the time to look into. "Nominal" source voltage is +12VDC - but actual reading can and should run anywhere from +12.6VDC to +15.0VDC - less means keep an eye on your battery, more means keep an eye on your alternator (of course, this is with the engine running. +12-13VDC is normal for Key OFF battery voltage.)

HoratioTheJeep said:

I couldn't find the exact fuse, but there were two 15A, blue fuses. I checked voltage at both, & both were a steady 11V DC (why 11 and not 12V?)

Click to expand...

I really wish I could have found a fuze name for you - yours may not even be 15A! It may take a little digging. Anyone here have a 1996 FSM?

HoratioTheJeep said:

I found the grounds, & the multimeter gave me a reading of 1-2 Ohms when contacting the sensor ground on the harness side (the Black/Blue wire). The 1994 FSM says it should read 5 Ohms or lower, so I guess it's in spec.

Click to expand...

That sounds livable - but I'd still prefer being thorough, and pulling the ECM connectors to isolate the sensor ground (at which time, there should be no continuity of any sort - unless they've done something creative.) The difference between the "sense ground" and "source ground" is simple - the "source" ground is a return loop for devices, and the "sense" ground is a +0VDC reference to which sensor values are compared. Therefore, the "sense" ground is usually somewhat isolated W/R/T the "source" ground to prevent false signals and fluctuations.

I'd be more concerned with the fact that you're getting the low heater supply voltage at the upstream sensor - especially since that's the sensor that your P0135 code is referencing (1/1 is the upstream sensor, 1/2 is downstream. Later versions with split exhausts will see 1/1, 1/2, 2/1, and 2/2. The first number is the "Bank" number (from V-configuration engines,) and the second number is its order in the exaust stream - #1 is closest to the engine, #2 is next, and if there were a sensor somewhere near the tailpipe, it would be #3.

As far as figuring out which is which without a manual, here's how to go about it.

1) Disconnect the PDC connectors (face it, you're boned anyhow, and you have to keep clearing the thing.)

2) Find the grounded pin - confirm by checking against the actual ground. This will be the heater ground.

3) Reconnect the PCM. Turn the key ON.

4) Find the pin getting +12VDC (nominal.) This is the heater source.

5) Start the engine.

6) Find the pin (you'll have to backprobe for this one) that shows somewhere between +1 and +5VDC - this is the "Sense Out" lead.

7) By elimination, the remaining pin is the "Sense Ground" lead.

5-90

5-90 said:

I'd still prefer being thorough, and pulling the ECM connectors to isolate the sensor ground

Click to expand...

I see what you mean, now, about disconnecting the ECM. I can see how it would affect things. I should've done that earlier, but didn't. Now it's dark, & I don't have light enough. I'll have to put that off 'til tomorrow morning, I'm afraid, but I'll report back tomorrow.

Regarding the following:

5-90 said:

That's something I'd take the time to look into. "Nominal" source voltage is +12VDC - but actual reading can and should run anywhere from +12.6VDC to +15.0VDC - less means keep an eye on your battery, more means keep an eye on your alternator (of course, this is with the engine running. +12-13VDC is normal for Key OFF battery voltage.)

. . . .

3) Reconnect the PCM. Turn the key ON.

4) Find the pin getting +12VDC (nominal.) This is the heater source.

Click to expand...

Here's the thing: With the key in the ON position I'm not reading +12V DC from any pin (on the O2 harness-side). If I remember correctly, I saw 1.8V DC with the key ON, on one pin (two of the others read 0.1V & the ground, of course, read 0.0). But I definitely have to start the engine to get 12V thru the O2 harness. Is that a bad sign?

So you're saying that, with the key ON and the engine off, I should see +12V DC? I'm definitely not. I just assumed those wires were unpowered with the engine off. And w/the key OFF I see only 0.01V DC, on one of the pins. I haven't checked battery voltage, but it sure starts fine every time.

Could the voltage regulator (which is located in [actually is a (small) part of] the alternator) cause low/nominal voltage at the upstream sensor (w/the engine running), when voltage should be higher? Or should I still look at this as a wiring problem?

I'm not up on the ODBII system, but if the sensor has the heater, isn't there normally a relay that actuates the power to the heater circuit?

Maybe you already did this and just did not mention it.

I have a 96 and had a similar issue after replacing O2 sensors.

Turned out I had caused a pin in the wiring harness to back out and therefore not make a proper connection.

I, too had a couple of mechanics tell me the same things you have mentioned.
Just my .02

old_man said:

I'm not up on the ODBII system, but if the sensor has the heater, isn't there normally a relay that actuates the power to the heater circuit?

Click to expand...

Yes, usually the B+ Latch Relay (RENIX) or ASD Relay (OBD.) I think I mentioned that earlier, but I may have forgotten. I was up to my ears in wiring diagrams...

Went back and looked - yeah, I did mention it (but I didn't really call it out, so I can see why it could have been missed.) It might be useful to "bridge" the relay by taking it out and temporarily connecting the power lead to the output with a heavy jumper wire - don't leave that hooked up too long. I'm pretty sure the ASD relay is labelled as such - but it would be helpful if someone with a 1996FSM would chime in and refine my data...

5-90

5-90 said:

but it would be helpful if someone with a 1996FSM would chime in and refine my data...

5-90

Click to expand...

got mine on my lap, what do ya need?

Cool. Could you ref my earlier posts and translate my 1997 data into 1996 data? His wiring is a little different than what I had to refer to, and I'd like to point him inthe right direction.

Also, if you could identify the fuze and relay that powers the HEGO sensors, that would give Horatio a big help...

5-90

5-90 said:

I greatly suspect that, once you fix the heater fault, the other two issues will self-correct.

Again - this is on the 1996

The HEGO sensor is a four-wire unit with "divorced" ground to the ECM. The wiring for the sensors is as follows:
UPSTREAM
BK/LB - Sense Ground (ECM Pin A4 - both sensors)
BK/DG - Sense Signal (varies with HEGO output) (ECM Pin A24)
DG/OR - Heater Source (+12VDC from PDC Fuze F23, 15A BLUE, via ASD (Auto ShutDown) relay.)
BK/WT - Heater Ground (Engine Block - near starter.)

DOWNSTREAM
TN/WT - Sense Ground (ECM Pin A4 - both sensors)
BK/LB - Sense Signal (varies with HEGO output) (ECM Pin A25)
DG/OR - Heater Source (+12VDC from PDC Fuze F23, 15A BLUE, via ASD (Auto ShutDown) relay.)
BK/WT - Heater Ground (Engine Block - near starter.)

Since you're getting a code for "heater fault," that tells me that you aren't getting voltage either TO or THROUGH the heater. Check for battery voltage at one terminal of the HEGO harness (vehicle side,) there should be ONE AND ONLY ONE pin that reads a full 12VDC with the ASD relay energised or bypassed (check for voltage at Fuse F20, a 30 amp,(in the Power Distribution Center) - for 1996. I'm not finding a name for the fuse, so it may take a little digging.)

If you have +12VDC at the heater supply pin, the next step is to check the heater grounds. The heater grounds are shown as being somewhat above the starter (there is a batch of grounds probably 8" or so on the side of the block above the starter - them's those.) Find continuity (Key OFF) from a single pin to the engine ground. The sense ground should be isolated through the ECM - but if you pick up two grounds there you're probably OK. If you'd like to manually isolate the sense ground, pull ECM connector C3 (Grey) at the PCM.

Check that out, and let us know what you find.

5-90

Click to expand...

THIS HAS BEEN UPDATED FROM ORIGINAL POST.

If you need the pinout drawings, e-mail me and I will e-mail them to you. does not allow posting them here.

5-90 said:

It might be useful to "bridge" the relay by taking it out and temporarily connecting the power lead to the output with a heavy jumper wire - don't leave that hooked up too long. I'm pretty sure the ASD relay is labelled as such - but it would be helpful if someone with a 1996FSM would chime in and refine my data...

Click to expand...

I do know which one is the ASD relay. A few months ago I swapped that relay out, with no change in my problems.

I don't have any jumper wire. I could go get some, though. What guage would be considered "heavy" in this application?