XK8 / XKR ( X100 ) 1996 - 2006

Engine stalled and now won’t start

Thread Tools
 
Search this Thread
 
  #41  
Old 10-27-2019, 06:24 PM
giandanielxk8's Avatar
Veteran Member
Thread Starter
Join Date: Oct 2012
Location: Puerto Rico
Posts: 3,530
Received 1,442 Likes on 772 Posts
Default

Originally Posted by MikeAlford
Am i missing something here, but I think that the preoccupation with the fuel supply seems to defy logic?

To my way of thinking, and forgive me if I am wrong but if the car ran above idle, and you kept it alive, then the pump is delivering more fuel than is required for idle. Which would to my mind establish that it must be working satisfactorily. I mean why would it clog up on a run, that I don't get. Left for a year without running, i get, but hot and running?

So the question to ask is what is the difference between and idle and running engine? The answer has to be the mixture of air to gas of 14:1. When you open up the throttle the ratio becomes less significant, that is the bleed level for idle becomes less critical, so my logic would point to an air leak, or the the computer for fuel metering, or MAF, thinks there's less air going in than there actually is. So either the computer doesn't calculate the correct fuel metering, for whatever reason, and therefore the engine cannot idle?

An air leak world trip a code and there is none other than the P0191 which appeared once and disappeared.
 
  #42  
Old 10-27-2019, 07:29 PM
mhminnich's Avatar
Senior Member
Join Date: Oct 2014
Location: Arlington, Texas
Posts: 885
Received 696 Likes on 346 Posts
Default

He says that depressing the Schrader valve results in a dribble of fuel. He should be getting sprayed. After that he got it to momentarily rough idle before it quit and has not started since.

Failed /failing pump seems the likely conclusion.
 
The following 2 users liked this post by mhminnich:
giandanielxk8 (10-27-2019), RJ237 (10-27-2019)
  #43  
Old 10-27-2019, 08:05 PM
giandanielxk8's Avatar
Veteran Member
Thread Starter
Join Date: Oct 2012
Location: Puerto Rico
Posts: 3,530
Received 1,442 Likes on 772 Posts
Default

Originally Posted by mhminnich
He says that depressing the Schrader valve results in a dribble of fuel. He should be getting sprayed. After that he got it to momentarily rough idle before it quit and has not started since.

Failed /failing pump seems the likely conclusion.

Yes, that's exactly what it seems and what I feared. I sent an email to both Coventry West and Jagbits inquiring about a used pump after reading that maxdwg bought a used one and it worked out well for him. If the price difference is not enough, I may end up opting for a new one.

I also read that, on the coupe, it is possible to remove the pump through the subwoofer hole, but I only read about people doing that with the older style dual pump setup. Does anyone know if the larger single pump module of the 4.2 L cars will fit through the clearance between the subwoofer hole and the rear glass?

I do not look forward to removing the fuel lines and working on my back while the car is on jack stands in an effort to remove a gas tank that is half full.
 
  #44  
Old 10-27-2019, 08:12 PM
MikeAlford's Avatar
Senior Member
Join Date: May 2015
Location: Cornwall, UK
Posts: 112
Received 32 Likes on 24 Posts
Default

I tend to agree with mhminnich. But without fuel pressure readings we are all guessing here, is a quirt better than a dribble? So either replace the pump with a known good one, or measure the pressure in the line, otherwise we will all go round in circles with speculation. We don't know if the pump is working during cranking, or whether it is robbed of power when the starter motor cuts in. In my opinion we need to eliminate the pump and if it were me, I'd measure the pressure, and establish some facts.
If the pressure is down, then replace the pump, if the pressure is good look elsewhere.

Video for a 4.2 fuel pump removal which is the same as 2004.

 

Last edited by MikeAlford; 10-27-2019 at 08:18 PM. Reason: addional infor
  #45  
Old 10-27-2019, 08:24 PM
giandanielxk8's Avatar
Veteran Member
Thread Starter
Join Date: Oct 2012
Location: Puerto Rico
Posts: 3,530
Received 1,442 Likes on 772 Posts
Default

I'll go to the store and buy a fuel pressure gauge tomorrow. Does anyone know, which of the relays in trunk is the one for the fuel pump? First thing I wanted to do was swap relays but I don't have the fuse box diagram and can't seem to find one.
 
  #46  
Old 10-27-2019, 09:42 PM
Johnken's Avatar
Veteran Member
Join Date: Apr 2014
Location: New Jersey USA
Posts: 5,457
Received 1,810 Likes on 1,115 Posts
Default


Here is location.
 
The following 2 users liked this post by Johnken:
giandanielxk8 (10-27-2019), sklimii (10-28-2019)
  #47  
Old 10-27-2019, 09:48 PM
giandanielxk8's Avatar
Veteran Member
Thread Starter
Join Date: Oct 2012
Location: Puerto Rico
Posts: 3,530
Received 1,442 Likes on 772 Posts
Default

Thank you John!
 
The following users liked this post:
Johnken (10-28-2019)
  #48  
Old 10-28-2019, 08:49 AM
fmertz's Avatar
Veteran Member
Join Date: Nov 2014
Location: Eastern USA
Posts: 2,621
Received 1,512 Likes on 1,060 Posts
Default

Also, remember this is a return-less system. It means there is a fuel pump controller (located in the trunk, by the filler neck, I believe). It reads fuel pressure at the rail (fuel pressure sensor), and the value is available over OBDII. Should be your first priority. A cheap ELM327 can read this number. The controller then drives the pump to reach the desired constant 55psi. If you lift the throttle (and bring vacuum in the intake), the excess pressure is supposed to be bled through a relief valve.

My understanding is that the "fuel pump" is really a combination of an actual pump plus some sort of pressure relief valve, all in the tank. Maybe your pump is fine (you are getting some pressure), but the relief valve is faulty and drains most of the pressure back to the tank, never pushing much fuel to the rail. From memory, there was a series of bad pumps a while back. Folks were ordering new pump assemblies, but kept getting these P0191 or P0193 (pressure too low or too high). Rev Sam did a lengthy search, and posted a bunch of details, I think. From memory, he even identified the relief valve, and posted the difference between the XK8 and XKR (did he order an XKR pump for his XK8?). I believe the main symptom was that the car was hard/slow to start because the fuel pressure was not held.

A cheap shot in the dark would be to use a nice quality injector cleaner (Techron?) and somehow run the pump for a while in the off chance this relief valve can be convinced to work again. Injector cleaner is not friendly to rubber, so do not go over the indicated amount.

The other theory is that the pump is somehow not getting enough power:
  • Swap the fuel pump relay in the off chance it is not offering enough Amps to the pump.
  • Check the controller for any sign of damage. If you can find the wiring diagram, try and bypass the controller to see if powering the pump more directly gives you enough pressure.
  • Check the power to the controller. Start with voltage...
  • "some say" try and over-supply (24V?) the pump for a very short bit in case it is slightly gummed up and needs a high torque to free itself up.

With the pumps that expensive and hard to reach, it is worth spending some time on diagnostics. The fact that there is some pressure makes me think the pump itself is possibly fine.

Best of luck, keep us posted.

PS: Once I was able to help an XK8 owner: the fuel pressure sensor was leaking fuel through the vacuum line and put fuel directly in the intake. Easy to check, pull the vacuum line, use a rag and run the pump.
 
The following 2 users liked this post by fmertz:
Diddion (11-02-2019), Johnken (10-28-2019)
  #49  
Old 10-28-2019, 09:29 AM
JagV8's Avatar
Veteran Member
Join Date: May 2009
Location: Yorkshire, England
Posts: 26,798
Received 4,548 Likes on 3,957 Posts
Default

Does this car have P1111?

If not (i.e. has P1000), it may be that it would like to flag codes but cannot due to the Catch-22 situation where it can't get to P1111 and can't figure which code(s) to flag.
 
  #50  
Old 10-28-2019, 06:56 PM
giandanielxk8's Avatar
Veteran Member
Thread Starter
Join Date: Oct 2012
Location: Puerto Rico
Posts: 3,530
Received 1,442 Likes on 772 Posts
Default

Originally Posted by fmertz
Also, remember this is a return-less system. It means there is a fuel pump controller (located in the trunk, by the filler neck, I believe). It reads fuel pressure at the rail (fuel pressure sensor), and the value is available over OBDII. Should be your first priority. A cheap ELM327 can read this number. The controller then drives the pump to reach the desired constant 55psi. If you lift the throttle (and bring vacuum in the intake), the excess pressure is supposed to be bled through a relief valve.

My understanding is that the "fuel pump" is really a combination of an actual pump plus some sort of pressure relief valve, all in the tank. Maybe your pump is fine (you are getting some pressure), but the relief valve is faulty and drains most of the pressure back to the tank, never pushing much fuel to the rail. From memory, there was a series of bad pumps a while back. Folks were ordering new pump assemblies, but kept getting these P0191 or P0193 (pressure too low or too high). Rev Sam did a lengthy search, and posted a bunch of details, I think. From memory, he even identified the relief valve, and posted the difference between the XK8 and XKR (did he order an XKR pump for his XK8?). I believe the main symptom was that the car was hard/slow to start because the fuel pressure was not held.

A cheap shot in the dark would be to use a nice quality injector cleaner (Techron?) and somehow run the pump for a while in the off chance this relief valve can be convinced to work again. Injector cleaner is not friendly to rubber, so do not go over the indicated amount.

The other theory is that the pump is somehow not getting enough power:
  • Swap the fuel pump relay in the off chance it is not offering enough Amps to the pump.
  • Check the controller for any sign of damage. If you can find the wiring diagram, try and bypass the controller to see if powering the pump more directly gives you enough pressure.
  • Check the power to the controller. Start with voltage...
  • "some say" try and over-supply (24V?) the pump for a very short bit in case it is slightly gummed up and needs a high torque to free itself up.

With the pumps that expensive and hard to reach, it is worth spending some time on diagnostics. The fact that there is some pressure makes me think the pump itself is possibly fine.

Best of luck, keep us posted.

PS: Once I was able to help an XK8 owner: the fuel pressure sensor was leaking fuel through the vacuum line and put fuel directly in the intake. Easy to check, pull the vacuum line, use a rag and run the pump.
Interesting, I will check this out, hopefully tomorrow. I must admit I didn’t have time to get to it today due to it being my mother’s birthday and studying for the two tests I have this week. It is possible that I don’t get the chance to tackle the problem until the weekend.

Originally Posted by JagV8
Does this car have P1111?

If not (i.e. has P1000), it may be that it would like to flag codes but cannot due to the Catch-22 situation where it can't get to P1111 and can't figure which code(s) to flag.
It has P1000 and I suspect you’re right.
 
  #51  
Old 10-29-2019, 02:11 AM
JagV8's Avatar
Veteran Member
Join Date: May 2009
Location: Yorkshire, England
Posts: 26,798
Received 4,548 Likes on 3,957 Posts
Default

Next, use OBD to find unset monitor(s) and then attempt their drive cycle(s).

See if there are any other pending codes - a bit hopeful so expect none.

Also check sensor values for plausibility.
 
  #52  
Old 11-02-2019, 08:43 AM
giandanielxk8's Avatar
Veteran Member
Thread Starter
Join Date: Oct 2012
Location: Puerto Rico
Posts: 3,530
Received 1,442 Likes on 772 Posts
Default

I tried Charlie’s suggestion from the other thread to unplug the connector and try wiggling the power wires.

The car started on third crank. Trying to rev up the engine would result in some sputtering while it was cold. Once it warmed up revving the engine was fine but allowing it to drop to idle would result in sputtering and stalling.

When warm, I tried starting the car. It fired on the second crank but sputtered and died.

I’m guessing the pump is fine. Still no codes over OBD II.

I’m off to buy the fuel pressure gauge manometer at the time of this writing.

Does anyone know the location of the fuel pressure sensor and regulator on this car,
 

Last edited by giandanielxk8; 11-02-2019 at 09:00 AM.
  #53  
Old 11-02-2019, 09:24 AM
fmertz's Avatar
Veteran Member
Join Date: Nov 2014
Location: Eastern USA
Posts: 2,621
Received 1,512 Likes on 1,060 Posts
Default

Originally Posted by giandanielxk8
Does anyone know the location of the fuel pressure sensor and regulator on this car,
Fuel pressure sensor is at the radiator end of the rail, passenger side. There is no regulator. The sensor at the firewall end is fuel temp. The shrader valve is at driver side, with a green cap.
 
  #54  
Old 11-02-2019, 09:30 AM
giandanielxk8's Avatar
Veteran Member
Thread Starter
Join Date: Oct 2012
Location: Puerto Rico
Posts: 3,530
Received 1,442 Likes on 772 Posts
Default

Originally Posted by fmertz
Fuel pressure sensor is at the radiator end of the rail, passenger side. There is no regulator. The sensor at the firewall end is fuel temp. The shrader valve is at driver side, with a green cap.

No regulator? Didn’t expect that.
Would a failed or dirty sensor trip a code?

What’s the round metal thing on a fuel hose that connects to the drivers side fuel rail?
 

Last edited by giandanielxk8; 11-02-2019 at 09:32 AM.
  #55  
Old 11-02-2019, 11:34 AM
$jaglife's Avatar
Junior Member
Join Date: Nov 2019
Posts: 3
Received 1 Like on 1 Post
Default

Maybe loose rubber hose at fuel pump, blown off by clogged filter? The rubber hose just sits on the pump causing intermittent no start
 
  #56  
Old 11-02-2019, 11:51 AM
giandanielxk8's Avatar
Veteran Member
Thread Starter
Join Date: Oct 2012
Location: Puerto Rico
Posts: 3,530
Received 1,442 Likes on 772 Posts
Default

Originally Posted by $jaglife
Maybe loose rubber hose at fuel pump, blown off by clogged filter? The rubber hose just sits on the pump causing intermittent no start
I’m going to check this next.

I measured the fuel pressure and got about 2 psi. The gauge barely moved. The test kit came with a multimeter. I clipped it to the red and yellow with red wires with the key in the II position and also got no reading. Is that the wrong place to use the multimeter?

The filter I bought is slimmer than the original because the store only had the filter for the earlier cars. Fitting it required that I fabricate a rubber spacer so that it would not slip from the bracket, could this be causing it? What if I installed it backwards? I installed it with the flow going towards the engine.
 

Last edited by giandanielxk8; 11-02-2019 at 12:03 PM.
  #57  
Old 11-02-2019, 01:14 PM
giandanielxk8's Avatar
Veteran Member
Thread Starter
Join Date: Oct 2012
Location: Puerto Rico
Posts: 3,530
Received 1,442 Likes on 772 Posts
Default

I checked and the hose is nice and tight.
 
  #58  
Old 11-02-2019, 04:50 PM
ccfulton's Avatar
Veteran Member
Join Date: Jan 2012
Location: Phoenix, AZ USA
Posts: 2,953
Received 1,108 Likes on 764 Posts
Default

With you gauge connected, if you cycle the key a time or two, does the pressure come up to around 55ish? The system regulates to 55 over manifold pressure so what you see on an analog gauge will vary.

A bad pressure sensor should give a code if it reads out of range but FWIW with my problem, there was never any code at all. Sometimes it would fire right up. Sometimes it would stumble and die at idle. Sometimes it would never start. Sometimes it would quit while I was driving.

It’s interesting that you fiddled with the electrical connection and something different happened. If you have access to SDD you can see both the pressure sensor output in real time plus you can see what the ECU is asking the fuel pump to do.

Normal for key on position 2 would be no voltage. The system is using PWM to vary the pump voltage (speed) to control the pressure, when the car isn’t running, the pump is off. As an aside, it will also shut it off on overrun, like highway speed and completely off the throttle. This can be a llittle disconcerting if you happen to be watching it, because say you have a fuel pressure issue you are trying to fix.

When you first switch the car on the pump will run full speed for about 1sec to prime the system. If you have a friend turn the key while you are back there with the liner pulled down, it should be obvious to hear. A high pitched gear/motor kind of noise.

When idling you can still hear it, although quieter. You should read a voltage of 7 or 8 volts at idle if you are using a digital multimeter.
 
  #59  
Old 11-02-2019, 04:53 PM
ccfulton's Avatar
Veteran Member
Join Date: Jan 2012
Location: Phoenix, AZ USA
Posts: 2,953
Received 1,108 Likes on 764 Posts
Default

Originally Posted by giandanielxk8
What’s the round metal thing on a fuel hose that connects to the drivers side fuel rail?
That is the fuel line from the tank. The round thing is just a special kind of connector.
 
  #60  
Old 11-02-2019, 05:11 PM
giandanielxk8's Avatar
Veteran Member
Thread Starter
Join Date: Oct 2012
Location: Puerto Rico
Posts: 3,530
Received 1,442 Likes on 772 Posts
Default

Originally Posted by ccfulton
With you gauge connected, if you cycle the key a time or two, does the pressure come up to around 55ish? The system regulates to 55 over manifold pressure so what you see on an analog gauge will vary.

A bad pressure sensor should give a code if it reads out of range but FWIW with my problem, there was never any code at all. Sometimes it would fire right up. Sometimes it would stumble and die at idle. Sometimes it would never start. Sometimes it would quit while I was driving.

It’s interesting that you fiddled with the electrical connection and something different happened. If you have access to SDD you can see both the pressure sensor output in real time plus you can see what the ECU is asking the fuel pump to do.

Normal for key on position 2 would be no voltage. The system is using PWM to vary the pump voltage (speed) to control the pressure, when the car isn’t running, the pump is off. As an aside, it will also shut it off on overrun, like highway speed and completely off the throttle. This can be a llittle disconcerting if you happen to be watching it, because say you have a fuel pressure issue you are trying to fix.

When you first switch the car on the pump will run full speed for about 1sec to prime the system. If you have a friend turn the key while you are back there with the liner pulled down, it should be obvious to hear. A high pitched gear/motor kind of noise.

When idling you can still hear it, although quieter. You should read a voltage of 7 or 8 volts at idle if you are using a digital multimeter.

If I cycled the key the pressure gauge would not move up from 2. I will check it again tomorrow.
 


Quick Reply: Engine stalled and now won’t start



All times are GMT -5. The time now is 12:25 AM.