6.8 mpg, check engine light, NSS adjustment, 98 XJ

First off, I apologize cause this will likely be long.

I got a 98 XJ with 196K on the clock not too long ago. The check engine light wasn't on, but the backup lights didn't work, and the thing only started in Neutral. It would start in Park if it felt like it. So, I searched and came up with the NSS removal, cleaning, and adjustment procedure. I did this... put it back in, according to the correct adjustment procedure. Then, it tripped the check engine light.

Now, I drove it maybe 5 miles before I did this with no check engine light. Then I did the NSS fun, and the CEL came on. I called codes and got a lean mixture code, and 3 transmission control codes (sorry don't have the actual numbers handy)

Anyway, since then I smogged it, and it failed horribly in every catagory (again sorry, no numbers). It smells bad, I have smoke coming out of it, so I wasn't too shocked.

At this point, I've now driven it like 100 miles, I've gotten an average of 6.8 miles per gallon with half highway and half city driving. I've adjusted the NSS 3 times now, and still get the 3 check engine codes in addition to the lean engine condition code.

My questions are:
1) Would adjusting the NSS cause a horrible gas mileage/rich condition?
2) If not, would an O2 sensor possibly cause this bad of mileage?
3) would the coolant temp sensor possibly do that?

I've not checked the compression on it yet, hoping to do that tomorrow... but it doesn't seem to be burning oil, no oil apparent in the coolant, no coolant apparent in the oil.

Yeah, I've searched... but really this is the worse mileage I've seen or heard of and I'm doubtful of a simple sensor causing this crappy of mileage.

A gratuitous picture of the Jeep in question

What happens if you let it warm up to operating temp in the driveway w/o driving ?
I'd start from scratch and pull the Throttle body, clean it out well, clean the sensors, etc.
Then I'd do a tune up on it, plugs, wires, cap, rotor.
From there I would revisit the NSS thing, see if you pulled a wire or damaged something while you were doing the NSS. Usually when you do some work and another problem just pops up it's from getting an elbow or something where it did not belong...

and get some new O2 sensors....

My check engine light has been on all the time for the last three months. I did a compression test yesterday and it got it to go away. Maybe it will help for you too, I don't know. Good luck, Matt

No, a NSS won't cause crappy gas mileage, but can throw the CEL on and make for some odd shifting/loss of over drive if it's not adujsted properly.

A bad o2 sensor can definitely cause bad mileage, if it's the proper o2 sensor. The on a 98 there are two. The upstream sensor is the one that controls your fuel mixture. The downstream one (the one right after the converter) is there to make sure the converter is doing its job properly. It serves no purpose in fuel mixture. I would probably go ahead and replace the first sensor.

A coolant temp sensor could probably cause some bad mileage if it were telling the computer a lie and saying it is actually colder/hotter than it actually is, which would make the jeep run either rich or lean, depending on if it's saying its cold or hot.

You say you have some smoke coming out of it, I would assume it's black smoke?

In addition to the above, I agree with Rich 100%. Do a good tune up, use good quality parts. Also take the throttle body off and clean it well, in addition to the sensors. Search around on here for instructions on how to clean some of the sensors with out messing them up.....some of them are picky to being cleaned.

Yeah, its black smoke.

BTW, I checked for leaks in the exhaust system and didn't find any... no cracks in the manifold, manifold bolts all present and accounted for, all tight, no black deposits around manifold gasket- that's been the cause of 2 of my last three jeeps running rich when purchased.

The transmission seems to shift fine, backup lights come on, etc. Right now, it will start in park but not neutral. However, when I had it adjusted so it started great in park and neutral, I still got the codes.

After every reset (disconnect the battery) the check engine light actually stays off until I let it sit and idle. If I drive it, it is fine.

Gotcha on the upstream/downstream O2 sensors... but couldn't that error come from either one?

I intend to do the tune up, I usually do when I buy a new Jeep. Problem is, I'm trying to pinch pennies for a few paychecks. Right now I just don't have the cash to throw at it... I'm planning new plugs, wires, cap, rotor, and likely a high output coil just as soon as the pocketbook allows (might forgo the coil thing initially)

Glad to know the NSS won't cause the mileage issue. I do think it caused the CEL to come on though.

Oh, and see, on cleaning the throttle body, that's the thing... this might be the first Jeep I've had where the throttle body and air box didn't have a ton of oil in it. I mean, it doesn't look as though it was cleaned before I got there... but it isn't wet.

Some reason I doubt that a simple copmression test will fix it... lol... but fingers crossed. I'll likely do plugs too while they are out.

BTW: I found the codes-
P0171- System Too Lean
P0700- Trans control System Malfunction
P0753- Shift Solenoid A Circuit Electrical
P0743- Torque Converter Circuit Electrical

Not sure I understand what you mean with the oxygen sensors.....it's not throwing a code for one of them is it? Do you have access to a scanner that can do some digital read outs, such as oxygen sensor readings, coolant temp. sensor readings etc etc? If you can get your hands on one, or know somebody who would do it for you and could get some readings, it would really help to determine if your sensors are going bad or not.

As for the throttle body, it looks like you (judging by your picture) have experience with older 4.0's, which are known for blow by and the alike. The newer ones dont have that problem, so carbon build up is the issue here.

I meant the system too lean error... couldn't that code come from either sensor? And, yeah I do have experiance with all kinds of 4.0s... Renix and HO. I have or have had an 88 MJ, 89 MJ, 93 YJ, 94 YJ, 97 XJ, 98 TJ, 01 TJ, and have wrenched on friends jeeps too... seems like every one but the 01 have or have had some kind of blow by.

Compression test results:
Cylinder 1- 170
Cylinder 2- 170
Cylinder 3- 157
Cylinder 4- 166
Cylinder 5- 153
Cylinder 6- 157

When I searched last night, I saw a lot of numbers from 120-150... is 170 abnormally high? Is it abnormally high for a motor with 197,000 miles on it?

As I pulled each plug, I checked the gaps on the existing plugs. It looks like they'd been in for a long time, and the gap was never less than .060, with a couple of plugs being over .080. Stock coil and ignition parts... or stock replacements, BTW. As I checked the existing plug and the compression, I put a new Champion truck plug back in gapped to 040. Well, after that I decided to see if it ran any differently, but it still ran rough. So, I decided to bite the bullet, and got a brandy new bosch upstream O2 sensor and put that in it. Reset computer and voila! New Jeep.

Took it for a drive, and using the on board overhead console computer, I got an average of 14-18 mpg on the highway. While that was more than double what I was getting before, I still felt it wasn't what it should be. Oh yeah, CEL is off now after that drive.

So, with the well maintained spark plugs in mind, I pulled the cap. I've never seen that middle electrode button worn all the way through the metal before. And, the rotor itself had seen better days too. So, I went ahead and replaced both of them too. I've yet to drive it yet, I'm on lunch now... and debating doing wires. I know it could use them, but I'm really over budget on this right now, and trying to hold off.

So, in the end, an O2 sensor alone would likely have helped, but plugs, rotor and cap made it run like a new Jeep again.

I'll see how long that lasts when I go to work tomorrow... but fingers crossed.

Oh, and no, I don't have access to a code reader that will do anything but call codes... no realtime monitoring stuff.

Thanks for the guidance guys.

over2land said:

I meant the system too lean error... couldn't that code come from either sensor?

Click to expand...

No...like he said, the upstream is the only one that controls the fuel mixture.

The tranny codes are most likely the NSS being misadjusted or not working right. When I adjust it, I set the parking brake with the key on and engine off. Put the shifter in Reverse, crawl under and adjust the NSS. Turn it one way until the backup lights just come on, then go until the turn off and set it in the middle of the range where you see the lights come on.

I would check the front O2 sensor as well. Granted it's over on the other side, but I've seen the wires get knocked loose and get up against the hot exhaust. If it's the original with 196k miles, it's probably time to change it anyway. You may also have an issue with a bad injector leaking extra fuel in. Does it take longer than normal to start, like the fuel rail is loosing pressure when you shut ut down?

JohnX said:

No...like he said, the upstream is the only one that controls the fuel mixture.

Click to expand...

Yes, I can read, thanks.

Just seems like a waste of a second O2 sensor. Why even bother if it is basically an on/off switch?

And, I mean no disrespect to anyone, I just find it hard to beleive... it just doesn't make sense to me that all that O2 sensor does is monitor the cat. Why?

over2land said:

Yes, I can read, thanks.

Just seems like a waste of a second O2 sensor. Why even bother if it is basically an on/off switch?

And, I mean no disrespect to anyone, I just find it hard to beleive... it just doesn't make sense to me that all that O2 sensor does is monitor the cat. Why?

Click to expand...

I was behind a chrysler suv, forget the model, looks like a cross between a durango and a pacifica, it was on a car hauler, I counted two extra O2 sensors, one in front of the muffler and one on the tailpipe, whats up with that...

lawsoncl said:

The tranny codes are most likely the NSS being misadjusted or not working right. When I adjust it, I set the parking brake with the key on and engine off. Put the shifter in Reverse, crawl under and adjust the NSS. Turn it one way until the backup lights just come on, then go until the turn off and set it in the middle of the range where you see the lights come on.

I would check the front O2 sensor as well. Granted it's over on the other side, but I've seen the wires get knocked loose and get up against the hot exhaust. If it's the original with 196k miles, it's probably time to change it anyway. You may also have an issue with a bad injector leaking extra fuel in. Does it take longer than normal to start, like the fuel rail is loosing pressure when you shut ut down?

Click to expand...

Thanks for the hint on the NSS... putting it in neutral and lining up the marks isn't working for me.

As for the O2 sensor, like I said, I replaced the upstream one... also known as the front one, and I think that was the major issue.

No, it doesn't take more time to start, but I hadn't even thought of that, thanks. It starts like every other 4.0L Jeep I have... roughly 2 or 3 cranks and its on.

RichP said:

I was behind a chrysler suv, forget the model, looks like a cross between a durango and a pacifica, it was on a car hauler, I counted two extra O2 sensors, one in front of the muffler and one on the tailpipe, whats up with that...

Click to expand...

I'm clueless.

I can't even wrap my mind around using a second one to monitor the cat... LOL.

Oh yeah, drove it 45 miles since the cap, rotor, plugs, O2 sensor. Let it sit for 15 minutes (when the light would come on before), some around town driving and about 30 miles on the highway... no check engine light still. I wonder if I run codes if the NSS codes will still be there. Are they "silent" codes?

over2land said:

I'm clueless.

I can't even wrap my mind around using a second one to monitor the cat... LOL.

Click to expand...

Yep, you are.

I wasn't trying to be an ass, I just figured you missed something. The second O2 is used stricktly to monitor the condition of the cat. It does this by checking reading from the first one against the one after the cat. The first one should go rich, lean, rich, lean, etc. The wave form from it would look like jagged mountains up and down. If the Cat is performing efficiently, then the waveform from the second O2 will look like gently rolling hills. If the computer sees something else, then it knows there is a problem with the cat, malfunction, poor performance, or maybe missing all together?


BTW....now I was being an ass, because you were rude when I was trying to help.

Good luck with the problem.

JohnX said:

Yep, you are.

I wasn't trying to be an ass, I just figured you missed something. The second O2 is used stricktly to monitor the condition of the cat. It does this by checking reading from the first one against the one after the cat. The first one should go rich, lean, rich, lean, etc. The wave form from it would look like jagged mountains up and down. If the Cat is performing efficiently, then the waveform from the second O2 will look like gently rolling hills. If the computer sees something else, then it knows there is a problem with the cat, malfunction, poor performance, or maybe missing all together?


BTW....now I was being an ass, because you were rude when I was trying to help.

Good luck with the problem.

Click to expand...

Yep, and I was being rude cause I thought you were being an ass... lol. My reading comprehension is usually pretty good... but I can tend to take stuff I read on the internet with a grain of salt. There are a lot of people handing out advice as first hand knowledge when it is something they heard someone say or saw them post once. So, if something doesn't make sense to me, I question it. Actually, come to think of it, I did that all through school too. Now that you've explained it in a little more detail, I think I get it. Not what I was thinking it was at first.

So even the first one isn't a wide band? Just a true/false setup? I would have thought that by 98 they'd gotten away from that 80's era tech.

Makes more sense to me now though. Thanks.

In that application I would agree.

However in my 98, I only found two... so would think it'd read differently.

And, FWIW... by my jackass comment to John X earlier, I was hoping to elicit that response (cut and paste of FSM description of the function of the system), but with a '98. I would have thought that was how the 98 worked as well... not just "to check the function of the cat"

When did they go to 4 sensors?