I am chasing what appears to be a fuel delivery problem on my 95 XJS 4.0l. Very rough idle in a cyclic repeating pattern. Symptom goes away when I either lightly spray starter fluid/vapor into the air intake or disconnect the coolant temperature sensor. The disconnect causes the ECM to use a longer pulse to the injectors. So this seems to indicate that not enough fuel is getting to the engine at idle. So here is my question. I measure 40lbs of pressure at the rail ahead of the regulator with no vacuum to the regulator. I read 35lbs pressure when 15lbs of vacuum is applied to the regulator. Is the 35 lbs of pressure in the rail the correct pressure needed to properly operate the injectors at idle? Also, I am getting 40 - 45lbs pressure from the fuel pump. Your help is always appreciated!

 

Service manual says that the fuel system should remain pressurized between 35-45psi.

A bad temp sensor could throw off your idle.

Can you see what your oxygen sensors are doing?

 

I have tried swapping out the temp sensor. Both read the same resistance at 80 degrees and at engine normal operating temp. I haven't compared those readings to specs yet. Anyone have temp sensor specs? I don't know how to test the oxygen sensors so could use some guidance on that. The Ecm has thrown codes for weak cats for the 2 years I have owned it, but it always ran fine and passed emission tests with flying colors. I plan to go ahead and replace the injectors to see if that is contributing. I cranked the engine while I had the fuel rail pulled out of the manifold and could not see a spray pattern from all 6.

 

You say it runs normal when you disconnect the temp sensor or spray something in the air intake, so I suspect the injectors are fine.

 

Vee, you may be right. I suspect the injectors because I pulled them and cleaned them. The XJS ran better after the cleaning, lessened the symptom, but it did not return to normal idling. I am theorizing that one or more of the injectors are not opening fast enough to provide proper fuel when the ECM is using the normal operating temperature pulse width to the injector. When I disconnect the temp sensor, making the ECM think that the engine is colder than the ambient temperature, the ECM lengthens the injector pulse and possibly results in the engine getting sufficient fuel to level out the idle. Pure speculation on my part. The injectors that I have are likely the original Lucas injectors, so as this point, having replaced the ignition coils, which did help a bit, the moderate expense for a set of cleaned and flow matched injectors is worth a try. The new set may not fix the problem, but like the ignition coil replacement, might lessen the symptom enough to help zero in on the real cause. My symptom could be stemming from multiple weak and failing components and the oxygen sensors could well be one of the factors. I will research how to test the oxygen sensors and report back here on how the replacement injectors affected the problem. If anyone can tell me how to test the oxygen sensors, it will be greatly appreciated. Thanks Vee.

 

If you have the original oxygen sensors, then you may be experiencing the same problem I had.

I had a rough idle, but the driving experience was fine. I did stall out at stop lights.

It turned out to be dying oxygen sensors. They were just responding slower than normal, so a bench test may not prove them to be bad. You would have to see what the short term fuel trims were doing. Mine were reading in the +20s and went down to about positive single digits.

They were rarely going into the negatives...

If you're going to throw parts in, I would recommend picking up a pair of NTK oxygen sensors, which will run you about $100 a piece. That might resolve the random cat codes you were experiencing too. (There are a total of four sensors, but two of them, I believe, are simple check sensors just to make sure the two are working. I think you want the downstream ones, but I always mix them up)

If that doesn't work, you can go back to the injectors, which will undoubtedly cost you more money.

 

Does your car idle well when disconnecting the MAF?

 

If I remember correctly, the engine dies quickly when I removed the MAF connector. I can try it again on Wednesday, but I do believe that when I tried that last week, it died. Maybe a coincidence that it died. My 04 XJ8 will run without the MAF connected.

 

As stated, old / dying 02 lambda sensors will not necessarily throw any codes but can be on their way out.

If they're original they could be the problem, or perhaps part of it at least

Symptoms could include, poor idle, hesitant or lurching, stumbling when on the gas, poor fuel consumption and more.

Perhaps at least consider them as part of your problem.....

 

I don't know the history of the oxygen sensors but they could well be those installed by the factory. Should the downstream be replaced first or just bite the bullet and replace all four?

 

On a '95 I would assume its higer mileage, perhaps 130k or more?
Can you please advise, as it is a guess.

If you're doing it on a tight budget then you'd want some way to try and check them, as some may be ok!

If you're looking to keep the car for a while and money isn't a major issue, then I'd suggest ( depending on price ofcourse ) to replace all 4

Check our forum sponsors SNG Barratt / Jagbits and others for pricing

SNG have a parts catalogue I think, please check, and also offer forum members discount, but shop around for sure.

 

OK. Just read a bunch of stuff about the upstream and downstream O2 sensors. It seems that the ECM uses the lean/rich signals from the upstream sensors, along with input from other sensors, to control the fuel map/fuel air ratio. The O2 sensors,it seems, provide the input that the ECM is the most concerned about. It strives for a14:7 air to fuel ratio. The ECM uses the downstream sensors to determine cat efficiency by comparing the upstream readings of oxygen in the exhaust to the readings of the down streamers. So my upstream sensors could definitely be the problem or a significant contributor to the problem. Thoughts?

 

Yep. The mileage on my 95 is 134000, so if these O2 sensors are the originals, they are toast.

 

You may find this link valuable as a guide....

http://www.google.co.uk/url?sa=t&rct...,d.d2s&cad=rja

There's no guarantees of course, but my guess would be that they are at least part of the problem.

Please by all means take further advice and do further checks, just add these to your list of possible causes / fixes.
The last thing I'd want to do is cause you to spend money you don't have to, but chances are that replacement will help with your issues.

Had similar issues on my sons Vauxhall Astra some time ago and a new 02 lamnbda sensor was the cure, this at around 60k miles or thereabouts.

 

Great site you suggested Jim. Based on your and Vee's input, I am leaning heavily toward replacing the O2s. The car is 19 years old with 134,000 miles, so some things are going to wear our and fail. So I cannot really complain if I have to replace all four O2s in the process of solving the problem. Thanks for your support and keep the ideas coming. I will keep updating as I continue working on this problem.

 

I sell the oxygen sensors for $159.95 made by the OEM manufacturer, simply plug them in no splicing, exactly correct for your car:

Jaguar LAMBDA OXYGEN SENSOR - LHE1682BA-X

 

Just so I can share what I was told by the Jag shop I took my car to, I had a combination of lazy oxygen sensors (I replaced all 4) and failing coils. Apparently he had tested them, and they were weak?

It seemed strange that no codes were being thrown to indicate either problem (although I was getting other codes) AND nothing had indeed failed. If the shop is to be believed (and I don't really have any good reason not to) then it does go to show that it can be a combination of issues causing your problem.

In any case, I think it would be money well spent to replace those sensors and then see what happens.

Going back to my car's problem and solution: The car's idle did get much better once I disconnected the MAF sensor. The fact that it's dying with yours may indicate something else is going on. Once unplugged, the computer should resort to a pre-programmed set of values to run on. The fact that it can't do that seems strange to me.

 

I plan to replace the injectors this coming Friday and if the problem persists I will also replace all four O2 sensors. I had already ordered the injectors and they are due to be delivered on Thursday. Although they are looking less likely to be the primary cause, it certainly could be a multiple input/cause problem because of the car's age and mileage and known good injectors shouldn't make it worse. Once they are installed and I can start the car, I will try disconnecting the MAF to see if it dies or continues running just for grins. I believe that you are correct in assuming that the ecm should use a default map when the MAF is disconnected. At least that is what my 04 XJ8 does. Maybe the earlier ECMAs don't behave that way. My 1993 Dodge Stealth would die when the MAF was disconnected. It may vary at the discretion of the engineering design team. Aside from the MAF experiment, if the rough/hunting idle persists, I will replace all 4 O2s. I greatly appreciate everyone's input. I will keep the thread updated.

 

Thanks and hope to hear how it goes for you, fingers crossed

 

Update: I installed a set of refurbished/manufactured and flow balanced injectors today. The XJS started up and actually ran better for 10-15 seconds, then died. After some coaxing, I was able to get the xjs to run and idle. The idle was almost the same hunting/stumbling idle that it had before the injectors were replaced. I did the Idle Air Control Valve set procedure that I found on this site and tried to set the idle speed. Side note: The idle air speed screw on my throttle body had little effect on the idle speed. I don't know if this is normal or if the problem's cause is overriding it.


So, and this is the confusing part, after playing with the idle air speed screw, the xjs began to idle with a lope (like it had done before). I could still give it a little squirt of starting fluid at the air intake and the idle would level out beautifully, but that does not solve the problem. Now that the xjs was running, with a lope, it went into a cycle of having the lope diminish and then without much warning just randomly die. The idle was lope, lope, lope..............lope, lope, smooth, smooth, smooth, stumble, stumble, lope, lope, lope......................... Then randomly die. I let the xjs idle for about 20 minutes. As it idled, the loping became less pronounced and after about 30 minutes it maintained a smooth idle longer than it loped. So the new pattern became: Lope, lope, lope, smooth.............................stumble, smooth....................................stumble, smooth........................lope.
The stumble would drop the idle speed about 150 rpm each time. So, feeling lucky I took the xjs for a drive. It ran better than it had, but randomly had a very brief hesitation when coming off of idle and would randomly die at stop lights. Also, once or twice, it would slightly stumble while driving at 45mph. I was in town, so that was my top test speed.


So, progress made, maybe. I ordered new O2 sensors which are due to be delivered on Saturday (Rock Auto $59.75 each, NTK OE). I also just received a used, known good Throttle Position Sensor (Thanks Scott if you are on here). While waiting on the O2 sensors to arrive, I will install the 'new-to-me' TPS. It only has 87K miles on it. My TPS has 134,000 miles on it, so it is worth the effort to install it to see what impact it may or may not have. At this point, I strongly suspect that Jim and Vee are correct in pointing to the O2 sensors. I plan to install the new ones this weekend. Fingers crossed. Thanks for all your help.