Fuel starvation and massive detonation - RESOLVED
 

Hey guys,

I think one or maybe both the fuel pumps in the XJR are failing...

Yesterday afternoon I got the chance to give it a short blast entering the freeway, (full-throttle runs are hard to come by around here) and it was hesitating & felt sluggish.

Rev out in 2nd and at about 5000rpm there was a big hesitation and massive detonation, so loud it sounded like rocks peppering the windscreen.

It's been 15 years since I had a car that pinged, but it was unmistakeable.

Ambient temps were warm, low 30's / 90's, late afternoon. I took it easy on the rest of the trip & got to where I was going, but the car felt sluggish and not as reponsive as it usually is.

After the meeting around 10pm it had cooled down to the low 20's / 60's and I came back onto the freeway & tried again. Again hesitation and down on power, and pinging, but not as bad as earlier in the afternoon when it was hotter...

I don't have my Android phone anymore, so I don't have Torque, couldn't check the fuel pressure or other values at the time. When I got home I plugged in my pocket scanner, no codes.

This morning I hooked up the SDD, no relevant codes (but this car has always had a heap of random "background codes") and I checked out the fuel trims & rail pressure, 55psi at idle, long term trims around zero.

This massive episode of detonation has left no codes or clues, but it reminds me of a couple of old carb-fed vehicles I had, where it was just plain old fuel starvation.

Other info;

Car has now 174'000kms / 108'000mi
Fuel filter was replaced 19'000kms / 12'000mi ago
AFAIK it's on the original pumps
Fuel gauge seems to be working ok, I'd just put in a half-tank of fuel
Intercooler pump is working

I have heard of one or both pumps failing in STR's and X350 XJR's (same twin-pump setup) as well as older models. Pumps don't last forever so at 10 years / 108'000mi, maybe their time is up.

Problem for me is a pair of new pumps will be $1600 in parts alone... :(

I have searched through the X350 & the S-Type sections. A few mentions of the S/C models with failed pumps, but I could not find anyone who had these sort of symptoms...

I'm reluctant to go out & road test it with the SDD connected to monitor the fuel pressure, that detonation sounded like an engine killer :( but there has to be some trouble-shooting done.

I am 99% sure its fuel starvation at high demand, just a matter of figuring out what's causing it.

Candidates;

Blocked fuel filter (but it's quite new)
Blocked internal filters in the pumps (possible, age/mileage)
Bad tank of fuel (possible, how do I check this?)
Failing pump or pumps (again, how do I check this?)

What else?

I'll wait for the morning traffic to clear and go for a drive with the SDD hooked up. I guess the thing to look for will be fuel pressure dropping when it's hesitating...

I thought I'd seen a TSB about hesitation and detonation so I searched and found this:

http://www.jagrepair.com/images/TSB/...Detonation.pdf

Hope that helps

Thanks Mac. I'd seen that TSB but discounted it because the ECU in this car has been reflashed a few times in the last couple of years and has the most up to date revision.

Alright before going for a drive I figured i'd have a look with the Engineering Test Mode in the instrument cluster to see if the fuel pressure was visible there (easier that trying to look at a laptop) I found some other "interesting" things...

First, a heap of DTC's shown in the message center, which the SDD also picked up

A139 = B2139 = Instrument Cluster ID does not match ECM
5986 = C1986 = Power steering speed threshold
D262 = U1262 = SCP Network Fault (Audio)
E152 = U2152 = Invalid Steering Column Lock message
E150 = U2150 = Invalid Steering Column Lock message
E195 = U2195 = Drivers Door Module invalid SCLM data
E510 = U2510 = Incorrect PATS identification - IC/ECM
D900 = U1900 = PATS CAN communication fault
9601 = B1601 = PATS received incorrect ignition key

Some of these make sense, some not.

It's happened a few times that the car would not start, pull out the key & re-insert, it starts, that explains the B1601.

Frankly i'm not worried about them right now, but I don't remember seeing these last time I went into the ETM so who knows. The corrective action on a couple is to reflash the ECU & Inst. Cluster, probably the DDM as well. I don't think they would be related to fuel starvation.

What would be is the fuel pressure measurement but you can't see it in the ETM, only the individual tank levels;

RAW FUEL 1 = 135
RAW FUEL 2 = 156

"Displays present received fuel input 1/2 in decimal, fuel gauge will present filtered level. 000-009 = short circuit (below empty) 10-254 = normal range, 255 = open circuit (over range)"

FILTER FUEL 1 = 126
FILTER FUEL 2 = 154

"Displays present filtered fuel level percent status from sender 1/2 in decimal. 000-254 = normal range, 255 = open/short detected"

These are the fuel levels on each side of the tank. Why would I have a different level on each side? Or is that normal?

Tolerance in the senders? Or because one pump is not pumping as much as the other?

There is one code that has been showing up in this car for a couple of years;

P1235 = Fuel Pump PWM control circuit out-of-range low. ECM to REM fuel pump control circuit: open, short or high resistance. or ECM failure.

The P2135 is only stored in the REM & cannot be read with a regular code reader. I always figured it was just "noise" now i'm thinking perhaps not...

perhaps this could cause a momentary fuel cut = detontation

is this an indication of 1 fuel filter being clogged?

and you just got a 1/2 tank of fuel ... bad fuel?

Big clue here.

Depending on the wiring in your particular model, this can be an indication of a lack of current and voltage at the fuel pump.
This would be especially noticeable at high demand.

When you had the IDS/SDD read the vehicle did you have DTC P1582 saying there was Flight Recorder Data available?

You might be able to use the recorded data to see what conditions were at the point of the hesitation/detonation.

This document:

https://www.jaguarforums.com/forum/at...rder-notes.pdf

mentions that Data Recorder ECM parameters are recorded such as Fuel pulse width, fuel pressure, fuel intervention, ignition, ignition angle, throttle failure, MAF volts, etc.

Unfortunately, it doesn't mention how to decode that data like it does other data in the notes.

[QUOTE=Cambo351;1164214]
When I read your first post my first thought was a momentary interruption of the fuel injector power supply, but there's no separate relay for the injectors on the X350. The power comes from a singe fuse in the front fuse box via the EMS relay.

Your fuel pump P2135 code seems like the best clue. Looking at the schematic, there are LOTS of related connections between the fuel pumps, REM, RPDFB and Fuel Pump 2 Module and the associated grounds. As a starting point, it might be worth inspecting and cleaning all those connections, and replacing the Fuel Pump Relay just to rule it out.

Only at WOT and high RPM, that'd be a hell of a co-incidence..

"Filtered" means "corrected" not actual filtration. It's still the level in each side of the tank, just with some corrections.

Could be. But the car had been slowly losing power for some time I think. The max speed over the standing 1000m circuit at the SM1000 was down compared to the previous year, for no apparent reason.

If it is bad fuel, I'll have to drain the tank, not fun. I will go to that petrol station later and ask if they have had any complaints since yesterday...

Just going through the wiring diagrams now. They do not show separate pumps, just one which is odd.

Yes there is a P1582, there has always been with this car.

The link to the PDF doesn't work.

The code has been there for a while, and come back after clearing codes months ago.

Looks like i'll have to start checking connections...

Not sure which Electrical Guide you have. See page 67 of the 2005 guide at the link below, which shows the S/C fuel pumps:

http://www.jagrepair.com/images/Elec...al%20Guide.pdf

Sorry about the link, will try again:

https://www.jaguarforums.com/forum/a...rder-notes.pdf

Attachment 106472

Here's the DTC's that were read, all 28x of them :(

EDIT

Tried calling the petrol station, no-one picked up the phone. So I googled "BP Baulkham Hills bad fuel" and this is what I get http://forums.whirlpool.net.au/archive/1709071 it's from a few years ago, but doesn't fill me with confidence...

Mildly contaminated fuel could have been causing your sense of decreasing performance, but perhaps in your latest tank you got a higher dose of H20 than usual? When I was a teen working at a gas station, we had to check the fuel level in the underground tanks every night and apply a water-detection paste to the wooden gauge rod to check the depth of the water in the tank (there were always at least a few inches of water).

Regarding your multitude of DTCs, our '04 has been throwing off lots of random codes every couple of months, right after I start the car. I'm assuming it's due to borderline battery voltage detected at the ECM, either due to poor battery charging or corroded or loose connections in the battery power circuits and/or grounds I haven't cleaned yet. I just clear the codes and only the few permanent codes related to our ACC remain.

I found this, which might help you determine if either pump could be the cause of inadequate pressure at high demand. It's how to test for high voltage drop on the negative and positive sides. High voltage drop is the most the common fault when you have a working pump, but that doesn't keep with demand.

Good find, Mac,

I wonder if this test method is valid on a PWM-driven pump circuit?

BTW, Airtex was the last company my late father worked for before he retired. There, he primarily designed mechanical and electric water pumps for Ford, GM, Chrysler and Toyota, but also contributed to fuel pumps and other components.

Cheers,

Don

I've decided to bet on bad fuel, and am getting the car towed to the local Jag dealer service department.

The service advisor said "oh yeah, we see it now & then, still got the receipt for the fuel? the petrol supplier will pay"

At the very least they will be able to get the panel off the underside that hides the fuel filter, two of the captive bolts are spinning & i can't get the nuts off... :(

Oh, good point Don. No, it wouldn't be valid on a PWM system. IIRC to figure out if there was insufficient pressure at high demand you would need to a scope on the signals.

Sounds like a good plan. We'll all be anxiously awaiting the outcome.

Attachment 106479

Off it goes...