Steve,

There might be a clue in something you've described:

- When you leave it idling after the expected timeframe, it suddenly dies, but when you're driving it, it splutters and tries to keep running before cutting out.

I could be completely off-track, but just maybe that indicates that it isn't a sudden electrical signal failure. It seems to imply that when you've varying throttle movement / manifold pressure / fuel draw etc, it keeps going a bit longer before dying.

So maybe it is a fuel flow problem, rather than a fuel pump cutout or fuel pump relay failure?

I think there's a few things you can do.

- Immediately it dies, see if the fuel pump can run. Get someone to listen whilst you turn on the ignition (the fuel pump will run for 2 seconds, each time you cycle the ignition). Or crank the engine - the fuel pump should run then
- Immediately it dies, carefully disconnect the fuel pipe at the fuel rail and see if it's still under pressure, indicating that fuel was being pressurised immediately before it died. Once you've got the pipe disconnected, also crank the engine and get someone to see if fuel is spurting from the pipe.

Also immediately it dies, pull a spark plug and crank the engine and see if you've got a spark.

By checking the state of both fuel and spark plug immediately after it dies, you might quickly pin down if it's a fuel or spark problem.

Good luck

Paul

 

Thanks for that suggestion, Paul. I'll follow up and report back after I recover from this week's overdose of Jaguar troubleshooting.

 

Got some new information. Executive summary: It's a fuel problem, not an ignition problem. I determined this by pulling the #1 coil and holding it just off the spark plug so it was still firing but I could hear the spark. When the car went into failure mode (took much longer today that previously), I could hear the spark rate slowing as the engine died. In other words, it sparked all the way down.

After the engine died, I loosened the fuel pipe connection at the input to the fuel rail and a little fuel dripped out. I couldn't tell what pressure it was under, but the engine and fuel pump were off. Frankly, I'm wary of trying that experiment when the fuel rail is pressurized for fear of starting a gasoline fire, so I'm throwing it out to you guys for any ideas about what direction to take now that it's pretty certain to be a fuel problem. I will mention that I don't know how to listen for the fuel pump, which I guess is inside the fuel tank.

Thanks again for your time and expertise.

 

Clogged fuel filter would be possible. Cheap to buy and relatively straightforward to replace.

The best next step would be to pressure test the fuel rail, but there isn't any port or valve to connect to unfortunately.

 

IF the car runs fine for 20 or more minutes, I don't see how it is a typical fuel issue -- that just don't make sense. Fuel problems, such as filter or weak pump are pretty binary, they impede flow or they don't. The lack of pressure after shutdown isn't surprising if the ECU stops sending the ground signal to the fuel pump, there will be enough engine rotations to drain pressure from the rail.

Is there a ECU fault that is causing it to shut off the fuel pump? Or is the ECU fine and it's actually doing its job and shutting off fuel for safety?

How old are your coils? If they have been replaced recently, were they OEM quality? @ptjs1 mentioned coils above, and I think he is on to something that fits your issues. Before you drive yourself nuts, I would replace your coils, UNLESS you KNOW they are all perfect. My recollection of the AJ16 is that it is sensitive to weak or failing coils. Coils are the one thing I can think of that would behave themselves, until they get hot. They expand with heat until they reach a point that they arc or short out, and send a confusing signal to the ECU. ECU gets that signal, and cuts the ground to the Fuel Pump Relay.

 

Thanks, Guys. I think one of the central questions is whether the fuel pump continues to run when the engine dies. So I'm thinking I should put a voltmeter on the power to the pump and watch it when the engine dies. If the power to the pump is getting cut off, that points upstream to the ECU, the Fuel Pump Relay, or - and I just learned this is a thing - the intertia switch that cuts the pump in a collision. (Actually, I just read up and the intertia switch cuts power to the ECU and all ignition-powered circuits, so it doesn't sound like that could be the problem.) If the pump continues to run (be powered) when the engine dies, then that points to a fuel flow problem or something downstream from the pump. If that all makes sense, then is there anything I need to know about getting to the fuel pump power terminals? I haven't explored that section of the car yet.

 

Steve,

There's a slight flaw in your theory of testing the pump. The pump only runs under 3 situations:

- when the ignition is cycled to Pos 2, the pump will run for 2 seconds then turn off
- when the engine is being cranked, the pump will run
- when the engine is running, the pump will run

So, when the engine stalls out, the pump will not then be running anyway. So your proposed test isn't valid in that respect.

I'd suggest you take a different approach.

Work out how to check the "theoretical" operation of the pump by checking the voltage at the relay, which you will see only under the above 3 conditions.
Work out what the pump sounds like by getting close to the tank from inside the boot, until you can recognise the sound of the pump running for 2 seconds under ignition cycle to Pos 2.

Then when the engine stalls, get someone to cycle the ignition to Pos 2 whilst you listen for the pump running for 2 seconds each time. If you don't then hear the pump, run a voltage check at the relay under the same conditions. to see if it's a relay breakdown situation.

If you want to check petrol flow, rather than just pump operation, immediately after stall, then work out how to pull the fuel rail connection and point the pipe into a large receptacle and check the fuel flows under the ignition cycle 2 second routine.

Cheers

Paul

 

Paul, thanks for the refinement. To plead my case one minute longer: Yes, the fuel pump will cut when the ECU fails to detect RPM from the CKPS, but that will happen only after fuel fails to reach the injectors. So the critical question is "What happens when?" If the pump powers down before the engine stops, that's a different root cause from it powering down after the engine stops.

Okay, with that said, I like your suggestion as it doesn't require a multimeter and mucking with connections. I'll give it a shot tomorrow and report back.

["tomorrow's" update]: With a helper, I identified the sound of the fuel pump relay clicking and the fuel pump running and then stopping as the key is switched to position 2. But even with my stethoscope, I couldn't hear the sound of the pump over the engine noise when the car is running. So I don't know how to use listening as the detection mode. I'm going to try to get a voltage probe on the pump relay, since I really don't want to have to pull the bulkhead cover to get at the fuel pump connector.

[Another update] I put a voltmeter on the fuel pump relay and waited (about 45 minutes) for the engine to die. It died while I was watching the voltmeter, and it clearly died BEFORE the relay powered down the fuel pump. Power to the fuel pump cut only after the engine had come to a complete stop. So now I think I need to look for either a fuel problem that fits these symptoms (honestly, sounds like a longshot) or an electrical problem that is shutting down the engine without turning off the fuel pump. I'm coming back to Mac's suspicion about the coils. I ordered a set to arrive Saturday. Any other ideas?

 

So what of the filter sock at the intake of the fuel pump is gunked up? Would it make sense that the negative pressure caused by the sucking of the fuel pump could allow it to work for a little while, say 25 minutes, before it just can't keep up with the load? I don't know. I'm just spitballing theories now. (fuel filter blockage could also follow the same logic?)

You did make me consider where the security system could be to blame? Is it plausible that something happens where the ECU loses communication with the security module and shuts down?

I was also thinking that the engine temperature could come into play here. Something happens when the engine gets hot, or enough time has passed causing a shut down.

 

With so many things now eliminated, a relatively simple question comes to my mind: If and when the ECU gets some input signal that triggers an engine shutdown, just HOW does the ECU execute the shutdown? The two ways I know about are shutting off the ignition coils and shutting off the fuel flow. I think we've established that spark continues as the engine dies, and that the fuel pump continues to run as the engine dies, so exactly WHAT is killing the engine? Is there anything that could shift the ignition timing into a shutdown?

 

Steve,

A couple of quick thoughts:

- the fuel.pump runnin isn't necessarily the same as fuel getting to the engine and being burnt. Fuel starvation through a cavitated filter etc can cause that. You have to know that fuel is at the rail and that injectors are being pulsed.

- a spark on one plug isn't sufficient to run the engine. You'd have to know you've got more than 4 sparks and they are being timed.

I think youre going to struggle to find what is failing immediately before shutdown. i would focus on what isnt working or signalling when you try and start immediately after shutdown.

cheers

Paul

 

Thanks, Paul. Yes, you are quite right that the information I have is not enough to assure a running engine. I'm not quite sure what you mean by focusing on what isn't working or signaling when I try to restart. The engine has always restarted immediately after a shutdown, but it only runs for a few seconds or maybe a minute or two if it's been sitting for a while. If you're able to elaborate that, I'll be happy to follow up. In the meantime, I have new coils on the way and I think I'll have a close look at the fuel filter.

 

95XJSteve just humor me here for a second. Have you personally checked that your throttle valve has .002 inches clearance? I cured your exact symptoms by setting my throttle valve. It only takes a few minutes and if it doesn't help, nothing is lost. Make sure your throttle cable is disconnected and your idle control valve is fully retracted when you check. Best to do it with a hot engine.

 

RickE, I have not personally checked it because I don't currently have a suitable feeler gauge. But I haven't put a priority on it because the car dies while driving, as well as while idling in the driveway, and this has pointed me away from that root cause. Should I reconsider?

 

That describes how my car was running. Here's my theory and maybe somebody can correct me if wrong. When the engine is up to full operating temperature the stepper motor (Idle Control Valve) should be fully retracted. If the throttle butterfly valve is not properly adjusted (fully closes) and you take your foot off the gas while driving, the throttle valve will fully close and thus temporarily stall the engine. Your instinct will be to pump the gas to keep the engine running. You may or may not be able to do that. Your ECU is now totally confused and will try to correct the condition rusulting in a misfiring or poorly running engine. You may be able to drive it in this condition but it will stall again when you take your foot off the gas. Sometimes it will cause a check engine light and sometimes not and sometimes will cause a transmission fault light just because the engine stalled.

My car would start, idle and drive fine for a few miles and then just stall out when I took my foot off the gas. It would start right up again but I could tell it wasn't running just right. Sometimes it would be so bad it felt like it was running on three cylinders. The only real code I ever got was a misfire on number 6 coil. I replaced the coil and still had the same problem. I was convinced it was a sensor that acted up when warm and tried to troubleshoot for months. One day the accelerator was hard to push and that is when I removed the throttle body and discovered that the butterfly valve was completely closing. There is a set screw to adjust it which I did and reinstalled it on the car (you do not have to remove the throttle body to adjust the set screw). I have taken several drives since then and the problem has not resurfaced. Hopefully this simple adjustment can help solve your problem also.

 

RickE, the difference is that engine on my car will cut out under light, medium, or heavy throttle. If the transmission is engaged and the car is rolling, it usually catches again within a few seconds before the problem gets progressively worse and it will only run for a second or two before it dies. I'm skeptical about the butterfly adjustment, but if the problem persists after I change the coils and check the fuel filter, I will get a proper feeler gauge and check it.

 

Feeler gauges are super cheap and very useful. Like well under $10 on Amazon.

 

Yep, I have two sets - SAE and metric - but they are blade types and not suitable for setting the butterfly gap. Need to get a wire type.

 

Sorry I misunderstood. My car never cut out while I had my foot on the gas. Hope you find the problem soon.

 

Well, Guys, there may be a ray of light here. Today, I replaced the coil packs, and drove and idled for an entire hour without any cutting out. That's at least 10 minutes longer than it's ever run since the problem developed, and today was warm and the engine temperature gauge moved will into the area above "N." So I'm cautiously optimistic that one or more of the coils was faulty-when-hot and, for whatever reason, that was causing the cutout. I keep thinking that "absence of evidence is not evidence of absence," so it will take more than one day's experience to really prove the case.

In the meantime, does anyone have an explanation how a bad coil can cause the engine to cut out?