97 Ranger 2.3 o2 sensors go to 0v at idle

Hi all,

I've got a 97 Ranger 2.3l (manual). Truck starts rough, idles poorly and acts fuel starved. A few minutes after first starting it will enter closed loop and if under load/throttle opened to more than ~2300rpms, it will stay in closed loop and run quite well.

UNTIL I'm at a red light, or idling for whatever reason. For a couple minutes its ok, and then it drops out of closed loop and runs poorly, and it will not go back into closed loop until... it feels like it--I've gone 30 minutes cruising down the highway, underpowered, until it decides to enter closed loop again.

Codes: p0170 lean and p1151 HO2S not switching.

I've got a scan tool on it and have noticed that when in closed loop the o2 sensors are responsive and switching. But as soon as it's at idle, they both go to 0v flatline (upstream and downstream always drop out/come back online together). For the next couple minutes I can rev up the engine and they will respond. Also, if I slowly step on the throttle, at around ~2300 they start switching again. If I slowly let off, they go right to 0v as soon as I drop below 2300rpm.

Once it drops out of closed loop, the o2 sensors stay unresponsive and it doesn't seem to go back into closed loop until I restart the truck. I haven't had a scanner on it long enough to tell if the o2 sensors start sending voltages again, but I suspect that's the case.

I have tapped into the o2 circuit at the PCM and the voltages are the same as those seen through the scan tool. The upstream o2 sensor is new, I've bench tested old and new with a propane torch, they both work fine. Also, ohms across heater circuit is within specs.

I've tested the TPS: 5v ref, <.2 voltage drop to ground. I only have a digital voltmeter to test the signal circuit but it seems to be sweeping smoothly and voltage readings are between .8 off throttle to 5v WOT.

New fuel filter (old one was one of the dirtiest I've ever seen--runs much better under load now). Fuel pressure is perfect, between 30-45 psi and no drops while quick revs. Stays solid at 45 when vacuum line pulled off regulator.

STFT is below +10% most of the time (scan tool I was using up til yesterday didn't show long-term). It shoots to +30% at idle though I haven't found a vacuum leak yet. I intend to replace the PVC (even though it rattles) but they were out of stock. I couldn't see it causing my main symptom, though.

What am I missing?

Thanks in advance for taking the time to read this.

Fuel pressure <> fuel delivery.

You can increase pressure in a closed hydraulic system by turning a small screw into a bung tapped into the system. The pressure throughout the whole system should be equal, and a small change in volume will pop or drop pressure.
How much does the pump deliver in a 10-30 second time frame? Measure and you may find it is inadequate.
Given that the old and replacement O2 sensors both respond the same way(I expect you replace the original because it was doing what the new one continues to do), the problem is not in the sensors, but in the fuel delivered, OR a leak in the intake, OR possibly plugged injectors that cannot spritz enough ... I didn't get the trims to be able to think on that, but one would think the computer would see the low O2 and start trying to richen the mix.
Other reason the O2s quit is they get too cold. At idle, not much heat being generated. Do you have a leaking exhaust system that could be cooling the O2 down too much?
The seeming lack of power when it goes into open loop must relate to fuel supply. In open loop, the computer uses stored values... which should allow decent drivability. You report different, so the open is not even working...
tom

Before going any further... I would ensure you have power and ground on the sensors heater circuits. Sounds to me like they aren't working. If they aren't hot... they simply won't work.

Thanks for the input. Uncle Gump, I tested the heater circuit today, power and ground at the sensor are both perfect. The circuit gets system voltage as soon as the engine starts and stays on the whole time it's running. Less than .02 voltage drop on the ground side.

tomw, I get what you're saying about fuel volume, but wouldn't the demand for fuel volume under load be greater than at idle? Because I only have this issue at idle, and unless it's in open loop, it runs like a champ at cruise and accel. And the short term fuel trims are usually less than 10% driving around, while they are at 30% at idle. So it seems that under load, the computer is able to compensate for any necessary demand, unless my understanding is wrong.

Today, while it was idling and the o2 sensors were unresponsive, I pulled a (small) vacuum line off and squirted in some brakleen and still couldn't get them to go rich. Throttle snap, nothing. Everything I read says that the computer ignores the o2 sensors while it's in open loop, but I just don't see why they would be stuck at 0v at idle, until the throttle is pressed a certain amount.

Right now I've got a voltmeter measuring the voltage drop across the ground side of the upstream o2 sensor circuit and my plan is to drive around and monitor it. The only thing I can think is that it is an intermittent open. Right now the voltage drop is around .04v, which could not cause this problem.

Yesterday, after 30 minutes of highway driving I got off at an exit and for the next 7 minutes of stop and go driving, everything worked normally. o2 sensors giving adequate cross counts, even at idle. Then a few hours later, it started up beautifully and I drove home in closed loop. This morning, it was the same as before, flatline at idle and then dropping out of closed loop and then resistant to going back in unless I restart the truck.

I am so stumped but I appreciate your feedback.

what about the actual o2 signal ground? is it frayed or loose?

whats the pcm temp sender doing? is it freaking out? air bubble in cooling?

I don't know the location of the physical o2 sensor signal ground, but that is what I am currently monitoring electrically with the voltage drop. Right now, it's a solid connection but I'm going to watch it while I drive to see if it opens.

The temp sensor for the PCM (the temp sender is for the gauge, if I remember correctly) looks fine. And if it wasn't, I'd expect a uniform lean or rich condition at any load/throttle position whenever a temp problem would occur. This one is only causing a problem at idle, or after it idles for a while. A temp sensor could be dead and the o2 sensors would still be doing their job.

if temp sensor is in low fluid or at idle because its contaminated or coated the drop in temp and flow could put it into cold start stratagy and ignore the o2. but should not turn it off...if anything the heater circuit should be cranked...and this model has heater circuits controlled by pcm...a failure in that system may be why it does not report i suppose...


does it use any coolant? the rough cold start could be valves or gasket/crack issues with o2 variation..causing bad reads...or rather good reads...


the cut in power makes me think bad connection in the pcm or harness though.


as to the 02...the circuit for the generation side is on the common..so where and how are you monitoring that? right at the connector of the sensor?



so you have jumpers built? thats what i do...then right to case on the ground side for loop monitor

if you dont...are you sure about that signal? seems i remember the signal ground on the o2 being separate then the heater circuit on all of them and direct in the early models...might be remembering it wrong i guess.


eventually and typically it is the 49 on obd1 and goes right to block ground big orange bastard with a eyelet.... if the break is internal then you get an intermittent short thats hard to find...seen it with stangs. these newer units are a bit different.

and pin 91 on the obd2 is the common and the rest all have separate pins for heater and signal......but if its lost before it ties in, the ground monitor your using isnt telling you anything if your doing it at the pcm connector.

2.3 should be 95 and 93 for heater grounds on my data with 60 and 35 being reported voltage if i have it right. dont have the actual pinouts with me. some you have to change on the jumpers....


bad valves/gaskets and physical exhaust leaks that are hardly noticeable can cause all sorts of readings that dont make sense. wiggling the harness at idle when the o2 power goes to 0 is something to try a few times if you have not..

Starts rough, idles poorly and acts fuel starved ... more or less.
Runs fine in closed loop, and may/may not drop out of closed loop on a whim.
...
Fix the first part and the second may take care of itself. In open loop, it should run without problems, idle smoothly, with ECT and ACT sensors reporting to the computer the coolant and air temp, and the O2 just fiddling around staring at the manifold until it gets hot.
You should have an idle above 1,000 rpms when cold, and it should step down in chunks as the coolant heats up. If not, start there and figure that one out.
Once you have that, clean IAC I suppose along with an intake that needs to be flushed on its surface with cleaner, and a gummy throttle plate, etc, the idle should pick up properly, and step as heat comes in.
I would do the above and also check fuel pressure when cold. The old fuel filter being the dirtiest you have seen may be a clue that the tiny screen inside the injectors needs some cleaning also. The screen fits into the inlet of the injector, and could be limiting fuel... Pressure and delivery from the pump are fine, but. Limited flow from the injector?
I had a misfire that seemed to hit two cylinders at a time at random conditions. I have a distributor, and went so far at to pull the fuel rail & injectors, the TFI, and finally found bad plug wires. It would go to dead miss, and then run fine. The screens in the injectors were clear, FWIW.
I do not know, but FoMoCo may have a strategy to ignore the O2 when certain conditions happen. If it finds there's misfire it may cut the fuel, ignore the O2 being LOW, and wander off until it is reset by a shutdown and restart. I am trying to think what also gets changed when the ignition is turned to OFF. What other thing can 'latch' and stay latched until reset? Either way, I'd find the cold idle problem as it may be causing the following problem in closed loop.(drop back to open)
If the MAF were being optimistic or pessimistic, the computer would / could be cutting back on fuel supply ... making for a poor idle, hesitation, etc. Have you looked to see if it has schmutz on the wires?
tom

SOLVED.

I had replaced the MAF originally: it was throwing a MAF code and behaving like a dead MAF so I put in a replacement from rockauto. Because the O2 sensors and MAF share a ground and the only thing that made sense was a weird problem with the ground circuit (not saying I'm right on that, it was just my line of reasoning) plus given the PCM runs on data from the MAF, ECT (and TPS?) in open loop... I decided to toss the old MAF back in to see what it would do.

And it was totally dead. Start and die, start and die.

BUT I noticed the new MAF wiring element had barely separated at one of the posts. Now, it is within the realm of possibilities that I damaged that while installing it... but I am pretty careful around hair-thin wires... anyway, I soldered it back on and re-installed it. Now, everything runs perfectly, everything is back within specs. Even the 30% fuel trim at idle went negative until I drove it around for a while and now it's jumping around 0%.

The O2 sensors are back on, exactly as they should be, cross counts at idle and everything. No more flatlining. No more rough idle, no more zero power until it warms up.

So... that was fun. Weeks of troubleshooting thinking it couldn't be the MAF because the system outputs changed so much when I installed it the first time, and it was responding smoothly and predictably on the scope. I guess the parameters were just off and I didn't recognize the units so didn't catch it. It started throwing the lean codes after I installed it (and went from start-and-die to running rough). I thought I just had two problems and the first one was taken care of.

So thanks for all your suggestions. It feels SOOOO nice to have a smooth running truck again.