• Welcome to the new NAXJA Forum! If your password does not work, please use "Forgot your password?" link on the log-in page. Please feel free to reach out to [email protected] if we can provide any assistance.

Downstream o2 - Fuel trim or no?

B

blistovmhz

NAXJA Forum User
Location
Vancouver, BC
98 XJ - 4.0L.
Was getting 20-23MPG. Had to pull the exhaust for some work. Put it back together and now I'm running super rich and fuel economy dropped to 12-16MPG and slight power loss.

Upstream O2 sensor is working fine (should be better now that I welded up a small crack right before the upstream, but just south of the collector.
The downpipe wasn't welded to the cat - down, and was held together with an exhaust clamp. When I stuffed the connection back together, it of course didn't seal all that well.

So now I know my upstream O2 is working normally, and my downstream waveform looks ... normalish, and I'm not throwing a code, but running super rich, poor mileage, and decreased power.

So while this question has been posed a thousand times, everyone always has a different answer. I'm in the camp that believes that OBDII puters do in fact use the downstream O2, in combination with upstream O2, to set fuel trim. This would make sense in my case, as I know I've got an exhaust leak just before the cat, which is probably letting in O2, causing the downstream O2 to read very lean.

So, before I go tear everything apart again and start welding my pipes together (I'm not a good welder) does anyone know definitively if the downstream O2 has any influence on fuel trim in an OBDII equipped 98 4.0L?
 
blistovmhz said:
98 XJ - 4.0L.
Was getting 20-23MPG. Had to pull the exhaust for some work. Put it back together and now I'm running super rich and fuel economy dropped to 12-16MPG and slight power loss.

Upstream O2 sensor is working fine (should be better now that I welded up a small crack right before the upstream, but just south of the collector.
The downpipe wasn't welded to the cat - down, and was held together with an exhaust clamp. When I stuffed the connection back together, it of course didn't seal all that well.

So now I know my upstream O2 is working normally, and my downstream waveform looks ... normalish, and I'm not throwing a code, but running super rich, poor mileage, and decreased power.

So while this question has been posed a thousand times, everyone always has a different answer. I'm in the camp that believes that OBDII puters do in fact use the downstream O2, in combination with upstream O2, to set fuel trim. This would make sense in my case, as I know I've got an exhaust leak just before the cat, which is probably letting in O2, causing the downstream O2 to read very lean.

So, before I go tear everything apart again and start welding my pipes together (I'm not a good welder) does anyone know definitively if the downstream O2 has any influence on fuel trim in an OBDII equipped 98 4.0L?
Click to expand...

It does. It works with the upstream O2S to set a goal voltage. However, it may not cause a massive loss in fuel economy.

How do you know you are running super rich? STFT?
 
The downstream o2 sensors monitor CAT efficiency, and the PCM will throw a code when out-of-spec results are received.

To my knowledge--and if anyone can post up a reference to any make, model, etc. that is the exception, please do--NO downstream o2 sensors are used for fuel trim.
 
"To my knowledge--and if anyone can post up a reference to any make, model, etc. that is the exception, please do--NO downstream o2 sensors are used for fuel trim."

From Dodgeram.org:

[SIZE=+1]Ram Upstream Oxygen Sensor Goal Voltage
[/SIZE]From Star Center News #5, February 2000 - tests for those who have access to OBD II or OBD III code readers.
Since the introduction of OBDII with both upstream and downstream O2 sensors, the downstream O2 sensor has been used to evaluate catalytic converter operation and to control converter efficiency.
When looking at the upstream goal voltage, it should fluctuate above and below the normal center (.4 to .6 volts) based upon what the downstream O2 sensor is reading. The vehicle should be driven through all speed ranges when checking goal voltage. If the goal stays high or low all the time, a thorough check of the downstream oxygen sensor should be made. Be sure you check the sensor, wiring harness, connectors, catalytic converter, and inspect for exhaust leaks.
Although a MIL lamp may not be illuminated, there are instances when the downstream O2 sensor goal voltage is out of range, causing driveability problems or low fuel economy.
If the downstream O2 sensor tends to stay below center all the time, the PCM will adjust the goal voltage higher in an efforet to correct the low O2 readings of the downstream sensor. This can result in low fuel mileage.
If the downstream O2 sensor has a tendancy to stay above center, the PCM will set a low goal voltage which results in a lean fuel mix which can cause the engine to run hotter and aggravate a spark knock condition.
Sometimes a service technician will reset the PCM and take it for a test drive. The PCM does a recalibration after the reset, and the probelm may go away until the PCM has completed a relearn of the system and then begins to attempt to correct for the downstream sensor error. If the symptoms reappear after several days of driving after a service is performed, check the downstream O2 sensor, the exhaust system, the wiring, and the catalytic converter. Note: a worn out catalytic converter can produce symptoms similar to a bad downstream O2 sensor.
 
Sidewinder CC said:
"To my knowledge--and if anyone can post up a reference to any make, model, etc. that is the exception, please do--NO downstream o2 sensors are used for fuel trim."

From Dodgeram.org:

[SIZE=+1]Ram Upstream Oxygen Sensor Goal Voltage
[/SIZE]From Star Center News #5, February 2000 - tests for those who have access to OBD II or OBD III code readers.
Since the introduction of OBDII with both upstream and downstream O2 sensors, the downstream O2 sensor has been used to evaluate catalytic converter operation and to control converter efficiency.
When looking at the upstream goal voltage, it should fluctuate above and below the normal center (.4 to .6 volts) based upon what the downstream O2 sensor is reading. The vehicle should be driven through all speed ranges when checking goal voltage. If the goal stays high or low all the time, a thorough check of the downstream oxygen sensor should be made. Be sure you check the sensor, wiring harness, connectors, catalytic converter, and inspect for exhaust leaks.
Although a MIL lamp may not be illuminated, there are instances when the downstream O2 sensor goal voltage is out of range, causing driveability problems or low fuel economy.
If the downstream O2 sensor tends to stay below center all the time, the PCM will adjust the goal voltage higher in an efforet to correct the low O2 readings of the downstream sensor. This can result in low fuel mileage.
If the downstream O2 sensor has a tendancy to stay above center, the PCM will set a low goal voltage which results in a lean fuel mix which can cause the engine to run hotter and aggravate a spark knock condition.
Sometimes a service technician will reset the PCM and take it for a test drive. The PCM does a recalibration after the reset, and the probelm may go away until the PCM has completed a relearn of the system and then begins to attempt to correct for the downstream sensor error. If the symptoms reappear after several days of driving after a service is performed, check the downstream O2 sensor, the exhaust system, the wiring, and the catalytic converter. Note: a worn out catalytic converter can produce symptoms similar to a bad downstream O2 sensor.
Click to expand...

That fits with the way it was explained to me, in laymans terms. The CAT has evolved so it isn't universal and year/series constant, but the downstream O2 sensors main job (OBD 2 and later) is to make sure there is enough oxygen left over for an efficient catalytic process (CATs have evolved and many are now three stage CATs). And the way it was explained to me is the downstream has priority over efficient combustion or gas mileage (the upstream O2 sensor). Just guessing here, but only a few ways to cool down combustion enough to have the extra left over oxygen, timing or fuel. The leaner the mix the more complete combustion and heat, the less left over oxygen. So in my guestimation the downstream O2 sensor has more influence than you'd think. THe defaults may or may not work for or against you, depends on your driving, highway or city. Like I said this is more guesswork than first hand knowledge, but the logic makes sense. Emissions have priority over mileage, the EPA at work.

Off topic, but have a look at your battery temperature sensor, if it is out of range it can fool the PCM into running constantly in cold (or warm up) mode. My milage sure improved after I swapped mine out. I noticed I had an issue with my charging rate and after a battery temp sensor test and change, my idle lowered and my mileage increased. Like the downstream O2 sensor, that Batt Temp sensor has more influence (priority) than you'd think or surmise IMO.
 
Depends on the year.

1997-1999 OBD-II is a little vague on whether or not the downstream O2 is used for setting the fuel trim.

That said, 2000+ the FSM is extremely clear that all O2 sensors (2 in Federal/49-state emissions, 4 in California/50-state emissions) are used when calculating the injector pulse width. If your downstream O2 or cat is bad in a 2000+, the PCM will alter it's behaviour. This was to comply with increasing emissions regulations, much in the same way that an EPA Tier 4-complient diesel will go into limp mode if you run out of DEF.
 
This is the thing. We toss around the word "normal" a tonne in the Jeep world, but I often wonder how often the term is really accurate. I average 22mpg which to me is "normal", as that was the number the Jeep was supposed to make stock, from the factory, and I've owned 6 Jeeps that have gotten about that number. But then I hear most guys say anywhere from 12 to 17mpg is "normal" and this makes me wonder how forgiving they are in the assessment of their engines :). I do know a lot of guys who wheel, and very few even bother to check their fuel economy once in the life of the vehicle :p.

Does anyone have a 4.0L XJ getting at least 20mpg highway, who can also check their sensor readings? O2 is a PITA as it's all waveform analog, but MAP and fuel trim might be useful.
Actually, my ex is in town with my old XJ (still gets 23mpg) so I'll try to compare against her ECM as well.

I've never had problem with the 4.0L that stumped me this bad before. All I did was pull the exhaust and put it back together. Seems like the obvious answer is that there must be a leak before the collector, but I can't hear it with a scope, and blowing pure O2 or sealing it doesn't change anything. The downstream o2 is the only one that still has a small leak. I suppose it's possible something else broke (maybe cat plugged internally when smashing exhaust back together?).
 
So took Jeepy for AirCare. Failed miserably.
Test - Reading - Max allowable
HC - 0.95 - 0.5
CO - 9.9 - 9.32
NOx - 4.69 - 1.24


So given I was getting around 23mpg before taking the exhaust apart, and after hammering it back together, I'm getting around 16-17mpg and failed AirCare miserably, how likely it is I just busted my Cat?

Also, does anyone know if the downstream O2 should be sitting pretty close to 0.45V, or should it be swinging wildly like the upstream? Seems to me that the downstream SHOULD remain pretty constant, as the Cat should stabilize output no?

At any rate, I've got 3 months to solve this before I won't be able to insure it anymore.
 
No it should swing wildly. Both of them work the same way by reporting lean/rich.

You able to do the vacuum test for a clogged cat? Cruiser's got a writeup around here somewhere. Vacuum leaks too if you haven't already, that's a lot of NOX
 
Yea, I'm stumped unless it's the cat. Haven't found any vacuum leaks yet. Compression test looked good enough at 159,155,142,160,162,152, and leakdown was good.

Pummelled the engine bay with propane trying to find a vac leak, but no change.

Engine definitely shakes a bit when idling. I've certainly had much smoother 4.0L engines, though none with compression this high.

I suppose symptoms are in order:

* Engine shakes a bit at idle, but smooths out above 1000RPM.
* Engine feels a bit boggy. Definitely not winning many races lately.
* Occassionally,when I've been wheeling it pretty hard for a few hours and I shut it down, when I start it back up it'll misfire on a cylinder for a minute before cleaning up. Doesn't throw a code.
* Cranks for a while before starting when it's been off more than 10 minutes. Fuel rail pressure definitely drops after about 5 minutes when not running, but isn't leaking out the injectors so must be return line.
* Massive drop in fuel economy (23 down to 17mpg highway, and 24 down to 18 city). Yes I know it's weird I get better economy in the city. Always have with every Jeep. Must be driving style.
* When starting from cold, makes a lot of white exhaust, smells rich. If I hold it at 2000RPM for 2 minutes, the exhaust cleans right up to invisible and mixture smells about right.

What is this cat test you speak of. Havne't found it yet.
 
Last edited:
Also just tested Cat. Let run at 2000RPM for 3 minutes before checking.
Inlet = 350F
Outlet = 410F.
Seems good?

Also, once cat is hot, the exhaust is clean.
 
Very strange. I'd go ahead and replace the O2 for fun anyway, NTKs are very cheap on RockAuto.

1998 doesn't have a return line, but it's not uncommon for the check valve in the pump to go out. I wouldn't think anything of a drop in pressure after 5 minutes of being shut off.

Misfires on restart are usually a sign of a fuel problem and/or heat soak. Is the temp gauge accurate in the dash?

cruiser54 said:
CRUISER’S VACUUM TEST FOR EXHAUST RESTRICTION
Your vacuum gauge should come with an instruction booklet outlining the procedure. Hook the vacuum gauge up to a source on the intake manifold. Start the engine and note the vacuum reading. Usually 17 to 21 inches of vacuum. Throttle the engine up to 2,000 to 2,500 RPM for 20 seconds or so and the vacuum reading should stabilize to the same reading you got at idle. Let the throttle snap shut. The vacuum reading should shoot up about 5 inches of vacuum higher for a second and then come quickly down to the original reading. If the vacuum reading stays high and comes down slowly with jerky needle movements, you have an exhaust restriction.
Click to expand...

EDIT: no idea about the cat inlet/outlet temp, sorry
 
O2's were both replaced about 10k ago. I may try replacing them again, but im goona test other stuff first. I hate throwing money at a problem without a clue :)

Hrm... Also took some pictures of my spark plugs while I was doing the compression test. They all look really good still. Now that said, I've only been running with this problem for a few days, but I'd expect the plugs to start showing a rich condition by now. I'm starting to wonder if I'm not actually rich at all, and that the whole problem is the cat?
What causes white exhaust, and isn't coolant ? :p
I'd guess water in fuel/crank case, except that the exhaust goes normal after heating up.
 
Last edited:
Pretty much fill the entire driveway with a white cloud every time I start it from cold, since doing the exhaust. Could be steam, but no one else is doing it.
 
White or blue-ish white smoke = oil
Black smoke = fuel/rich
White steam = condensation or head gasket
 
blistovmhz said:
98 XJ - 4.0L.
Was getting 20-23MPG. Had to pull the exhaust for some work. Put it back together and now I'm running super rich and fuel economy dropped to 12-16MPG and slight power loss.

Upstream O2 sensor is working fine (should be better now that I welded up a small crack right before the upstream, but just south of the collector.
The downpipe wasn't welded to the cat - down, and was held together with an exhaust clamp. When I stuffed the connection back together, it of course didn't seal all that well.

So now I know my upstream O2 is working normally, and my downstream waveform looks ... normalish, and I'm not throwing a code, but running super rich, poor mileage, and decreased power.

So while this question has been posed a thousand times, everyone always has a different answer. I'm in the camp that believes that OBDII puters do in fact use the downstream O2, in combination with upstream O2, to set fuel trim. This would make sense in my case, as I know I've got an exhaust leak just before the cat, which is probably letting in O2, causing the downstream O2 to read very lean.

So, before I go tear everything apart again and start welding my pipes together (I'm not a good welder) does anyone know definitively if the downstream O2 has any influence on fuel trim in an OBDII equipped 98 4.0L?
Click to expand...

I've been perusing the Bosch Automotive Handbook, 8th ed for a definitive answer, haven't found it yet.

However, as I have been given to understand it:

The "upstream" heated exhaust gas oxygen sensor (UHEGO) forms a feedback loop to provide a signal indicating whether the fuel mix is "rich" or "lean." Generally a "knuckle-type" sensor, the widest variation in voltage relative to oxygen content close to the ideal mix. Response falls off as the mix gets farther off.

When the UHEGO fails, the ECU will default to a "slightly rich" condition without the feedback trim. This is done to prevent engine damage due to "lean burning" (which can cause temperature-related damage,) but it is not rich enough to result in "cylinder washdown" (liquid fuel collects on the cylinder walls and will "rinse" off the oil that remains there, which accelerates ring & bore wear.)

The "downstream" HEGO (DHEGO) simply monitors catalyst efficiency. I believe the ECU compares its readings against the UHEGO output to determine if the exhaust gas contains excess oxygen and how much - once it gets thru the catalytic converter, oxygen content should be effectively zero (slight oxygen content is fairly normal, but a need for oxygen should not be indicated.)

If the DHEGO is to fail, the ECU will throw a CEL/MIL, and the catalyst efficiency will not be monitored. Nota bene: In OBD-II, the same DTC may be used to indicate either DHEGO failure or catalyst failure! The ECU detects that the DHEGO signal is out of range or not switching properly, but cannot tell you why. It's up to you to sort it out.

That last bit is why I don't look at OBD-II as an "end-all, be-all" troubleshooting method - as far as I'm concerned, the DTC merely tells me where to start looking. Still gotta do some investigation - just because OBD is telling you that a part has failed, doesn't mean that the part has failed. It may not be able to read what it's supposed to for some other reason, or it may have a wiring fault in the circuit, or ...
 
Damn. Have to get even more basic I think.
If I force the ecm into limp mode by disconnecting an 02 sensor entirely, I still get the same white smoke/vapour at startup. No change whatsoever.
So whatever the problem is, it's not at all related to any electronics.

Will try to track down a vacuum gauge to test cat today. It's seeming like the best option now.
[
 
You must log in or register to reply here.

Similar threads

C
How's My Fuel Trim Look
Replies
4
Views
2K
maxbraketorque
M
S
2000 STFTs stuck at 98% ??
Replies
8
Views
2K
RCP Phx
R
U
O2 Sensor differences?
Replies
3
Views
2K
Uncle_Enzo
U
G
1997 Sport running rough after warming up
Replies
8
Views
2K
Gilpine
G
J
2001 O2 Sensor pigtail
Replies
4
Views
1K
jmh9000
J
Back
Top