Engine Rough until Normal Operating Temp

Have been having a problem for a few months now and Im to the point where Im tired of it and want to fix it. It does it intermittenly so I am having trouble pinpointing it.

When I start the engine it will idle around 750rpm then drop down really low to where the engine about dies, but it never does. Then it will go back to normal and maybe jump to about 1000rpm just for a second, then back to normal. While driving I can hold the gas pedal steady but the enine cuts out so bad I can't go any faster than 30mph. At about 120*-140* it will cut out real bad and makes a loud popping sound possibly a mild backfire. The only helpful hint I can give is that it will do this until 195* then it runs just fine.

FYI it is a '97 Sport 4x4 4.0L and I have replaced TPS, CPS, IAC, ignition coil, and downstream O2 sensor

Well first I would try checker, pepboys auto zone what ev. and get them to plug in to the computer to see if there are any error codes that my be present.

next maby a computer reset is needed...

good luck, if I get a better brain storm Ill be back.

One clue here is that it happens when engine is cold, not hot. Try testing the coolant temperature sensor; not the one that feeds the coolant temp gauge, rather the one that provides input on coolant temperature to the computer. I'm not positive but I think I remember that if you unplug it, it will use default values. May be worth a try to see if unplugging it makes it run better....Good Luck!

birchlakeXJ said:

One clue here is that it happens when engine is cold, not hot. Try testing the coolant temperature sensor; not the one that feeds the coolant temp gauge, rather the one that provides input on coolant temperature to the computer. I'm not positive but I think I remember that if you unplug it, it will use default values. May be worth a try to see if unplugging it makes it run better....Good Luck!

Click to expand...

Thanks for the input, where might I find the coolant temp. sensor? I know there is a plug that goes into the thermostat housing but im guessing this is the one that feeds the coolant temp guage.

If your XJ is a '97 like your handle implies then that sensor is the only one you have (the computer uses it to monitor the engine and feed the gauge both, while earlier years had a separate sensor for the gauge)

There is only 1 on your 97, and that's it.

ehall and hubs; thanks for the clarification. It's been a while since I worked on a CTS and as I had an 88 and now a 99, sometimes the differences become fuzzy!

A short to ground on the coolant temperature sensor would create a situation where "no resistance" is detected, causing the computer to think the coolant air temperature was extremely hot and lean out the fuel mix. Not saying this is what is happening to yours but it could be. The coolant temp sensor can be tested but it involves a beaker of heated, cooled water and a calibrated thermometer; kind of a pain. If you suspect it, most guys just buy a new one.

I'm thinking there is some quick and dirty things you could also try to see if it is related. I'd try unplugging it or at least unplug the connector and get a resistance reading while the engine is cold, then plug the connector back in and warm the engine up, then turn engine off and unplug the connector again and get another resistance. If the resistance is similar, you've found your problem.

Good luck and post back what you find. Some of us guys get way too excited about working on this stuff and like to know the resolution. I really need to get out of the house more!!

How about the IAC :dunno: That controls the amount of air allowed into the engine when it is cold. In short, it should work like a choke (butterfly) on carbeurettor engines and allow less air when cold causing it to become rich.

IAC (Idle Air Controller)

techno1154 said:

How about the IAC :dunno: That controls the amount of air allowed into the engine when it is cold. In short, it should work like a choke (butterfly) on carbeurettor engines and allow less air when cold causing it to become rich.

IAC (Idle Air Controller)

Click to expand...

If you recall from the OP I have already replaced the IAC. Thanks for the idea tho.

Just last night the CEL lit up(after months of dealing with the problem) went to get the code scanned..turns out to be the upstream O2 sensor short curcuit. I replaced the downstream O2 sensor when my old catylitic conv. went bad. Somebody chime in if im wrong but wouldn't a bad upstream sensor cause the symptoms im having...I remember reading something about switching between open and closed loop based off the O2 reading

When no signal is received from the O2 sensor, as is the case when a cold engine is first started (or the 02 sensor fails), the computer orders a fixed (unchanging) rich fuel mixture. This is referred to as "open loop" operation because no input is used from the O2 sensor to regulate the fuel mixture.

It's possible you may have two separate issues. A lazy upstream O2 sensor and the issue creating the "cold driveability" problem you are experiencing. Checked your CTS yet?

birchlakeXJ said:

When no signal is received from the O2 sensor, as is the case when a cold engine is first started (or the 02 sensor fails), the computer orders a fixed (unchanging) rich fuel mixture. This is referred to as "open loop" operation because no input is used from the O2 sensor to regulate the fuel mixture.

It's possible you may have two separate issues. A lazy upstream O2 sensor and the issue creating the "cold driveability" problem you are experiencing. Checked your CTS yet?

Click to expand...

I think you are right about two separate issues. I went ahead and replaced the upstream O2 sensor and im still having the issue. What is my CTS and how do I check it?

97xjcrawler said:

I think you are right about two separate issues. I went ahead and replaced the upstream O2 sensor and im still having the issue. What is my CTS and how do I check it?

Click to expand...

Im an idiot...CTS=Coolant Temerature Sensor like you mentioned before.

Alright I did the CTS check. Sensor reads 1.84ohms @ 130* and 1.10ohms at 195*. What do you think birchlake? While doing this reading the engine was idling erratically and definately backfired while I was under the hood:explosion

Here the chart. You should really check it when it cold too to see if it shows the large resistance value.

The best way to do it is as follows:

1. Remove the coolant sensor
2. Put the tip of the sensor into a container of water
3. Connect a digital ohmeter to the two terminals
4. Using a calibrated thermometer, compare resistance of the sensor using the chart on this thread to the temperature of the water.

Remember when I said testint it was kind of a pain? That being said, if you test your sensor while still in the engine when the vehicle is cold (say overnight), it should correspond very roughly to the outside temperature resistance numbers on the chart. So if it is 50 degrees in Kansas overnight and you test the sensor cold, you should be somewhere in the ballpark of over 17,000 ohms. If however you are still getting those very low readings around an ohm and I'm guessing you will be, definitely replace the sensor. It doesn't have to match the chart perfectly but you should be seeing some very big resistance swings in testing with a hot engine and a cold engine. I'm guessing the sensor has gone south on you. Good luck and keep us posted.

Well im not sure if I have my multimeter set correctly but when testing the resestance(ohms) I put it on 20k ohms. When the engine is cool(equal to Kansas outside temp of 35*F) I get a reading of 9.60ohms. And when at operating temp. of 195*F I get a reading of 1.10ohms. Is this actually single digit ohms or is it in the thousands(i.e. 9.60=9,600ohms & 1.10=1,100ohms).

So either way it still seems my readings are off from saudades chart.

I guess the engine is still cooling down. Did another reading and it was 11.67 so maybe 11,670ohms?

I'd replace the sensor if I were you; I don't think they are too expensive.

Well there is no "k" in the display but the dial is set to "20k" on the multimeter ohms scale. In the display it says 11.67 and below the period is the number 20.

The settings available on the ohms scale are "200" "2k" "20k" "200k" "2M" "20M" Craftsman Digital Multimeter 82170

Generally speaking, those determine the max reading. So I would read your number above as 11,670 ohms (or 11.67K) If you had it on the same setting when hot, it may not give you a reliable number. Once fully hot, set the range to 2K and see what you get. Do the "cold" test in the morning once your sure it fully cold.