Hello all. Climbing out from UNDER my bed, lol

So, I have the OE VDO (the exact type that was in the car) pump I got off the eBay listing Bass hooked me up with a week back... I disconnected the plug to the in-tank jimmied pump, and hooked the OE up outside of the tank to the OE wiring... It jumped in my hand but would not spin for the 2 or so seconds of "priming" at key turn. I then started the car with the residual pressure in the rail,,, car started,,, starved for fuel,,, I felt the pump jump, maybe sputter but never spin in my hand. I regrounded (jimmied) during this,, and the OE pump spun well and happily in my hand... Reconnected the jimmied ground wire and the non OE pump in the tank kicked in, pressured up and car went back to running as normal.

I have an ECU and PWM signal problem and I don't know how to solve it. I tried hard grounding the black wire that is supposed to be earth coming out of the fuel pressure module in the boot as Michael suggested and no change... I'm spun here and DONT want to be going down the road (in the future) with this jerry rigged set up.

Could it be the pressure sensor on the rail? If that fails does it shut down the PWM system entirely?

Could it be a half eaten wire between the ECU and the boot somewhere?

I don't have a scope,,, don't have the IDS diagnostic equipment... Can't afford it,,, what now? What a PITA.

If you were me (no laughing) what would you do next?

****Most everything else is coming along well.

NEED:
-Door lock actuators - no biggie
-HVAC recirc flap solenoid/actuators x2 - no biggie
-Clevis Pins for the top/roof - on the way thanks to Jag heaven-Convertible Pump - can find one when ready
-Upholsterer (found) to help restore Recaro seats. Got all seat functioning as normal except headrest and recliners.

And that's about it... Of course more will show itself once driving. Most likely ABS sensors, but who knows?

It's this damn PWM issue thats proving unsolvable with the tools I have at my disposal...

Still - I'm a very very happy Man! Turns out,,,, it's a Portfolio Edition!!! Holy smokes!

 

Sorry JayJay, was hoping for you on the pump or at least giving you a direction to the problem.
Good luck, it must be your turn.

 

OK, not running, but it's trying. Are you absolutely certain that the new pump is the correct item? Not doubting you - just something to tick off the list.


OK, so it seems the ECM sees the drop in fuel pressure, and tries to run the pump. That tells us the pressure sensor is working in some fashion, and that an ECM signal of some sort is being received by the FPM.


So we know the pump runs OK, and that the 12V feed to the FPM isn't dropping under load.


So bad ground to the FPM ruled out


The ECM does some intelligent monitoring of the output from the pressure sensor. If the value goes outside high/low limits, or the pressure doesn't change in accord with ECM demands, then it fails back to defaults. I need to re-find the documentation I read on this


Possible - but we know something's getting through.

Definitely still no codes thrown?

 

My TURN, lol...? Imma git ya, CJ!

 

Hey Michael!

No codes specific to fuel pump,,, but yes to a bank 1 lean code and yes to (what I think is a 02) sensor code. They have been with me all along... I think the down stream O2 sensor spent a long while under water... But thats all the codes I get. No other. I grateful for that.

Something is getting thru OE style. But I cannot read what, and even if I did,,, I must say, I still wouldn't know.

Yes, the pump is identical.

I'm not sure if the FPM was responding to the drop in FP due to ECU signal,,, and trying to run the in hand OE pump harder to compensate. Something was happening tho. There's no doubt in my mind about that. Just not the right thing. Like, I thought that the 2sec prime would just be a straight 12v. Not a PWM signal. But I don't know. Still, it didn't run for the 2seconds.

I don't think voltage drops under load,,, it's the 12v that holds the non OE pump running strong (70psi) and a ground that finishes the circuit. That's is thru the FPM.

It's a tough thing. Feels so close but really,,,, it's a million miles away from understanding. At least mine... Well, it seems that way.

I would be interested to know more about that drop back to default of the rail pressure sensor,,, if that means that it's failure would just alllow the pump to run fully,,, as opposed to not at all. Crazy Jaguar ****...

Michael! It's a Portfolio Edition... I did not know.

 

I saw this while browsing locally and thought of your plight. :-)
https://m.facebook.com/marketplace/i...ref=browse_tab

 

This is a problem, but let's consider what you know. First, spragging the negative PWM signal to ground turns it on. Not knowing anything more, this seems to indicate that the pump runs with a negative duty cycle (i.e. ground is the signal), and you have likely turned it on 100% by grounding the lower signal. This may or may not be a good idea for the electronics in general, but no matter. In this case, the pump output doesn't care what the ECM thinks and fuel flows.

Again, in general, you want the ECM controlling the fuel flow since 100% duty cycle is probably going to kill the pump sooner rather than later. So, either the PWM signal is not appropriately getting to the pump, or it is getting the appropriate signal. In the former, wire connections would be a good bet. You could rule this out by looking at the PWM signal, but not possible at this point from lack of equipment, so moving on. In the latter, everything from the ECM to the pump is working fine, but your inputs are either not what you expect or not what the ECM expects to provide fuel (some negative duty cycle value that turns the pump on, may be just above zero). So, this could be wires/connections/sensors/assumptions. The 'assumptions' part is less certain, it may just be that you/everyone/me is making an assumption that is incorrect which leads us to expect it to work, when it shouldn't.

So, first I'd try to beg borrow or steal a scope to determine if the PWM is working into RECM, and out of RECM, that gets rid of an entire branch above and may give you enough information to solve your problem. Second, i'd try to make sure all of the inputs are okay, like in the text below. So, for instance, if CKP is not received after 1 second of prime, it doesn't turn on.

In any case, admire your tenacity.

Note below: this is the 32 bit Denso system, could not find similar diagrams/information for the AJ27, but didn't spend a vast amount of time searching, so it may be available somewhere and is likely quite similar.

 

For completeness more than anything at this stage, here's the text I found in the The on-board-diagnostics-v6-and-v8-engine-obdii-monitoring-systems-jaguar-cars.pdf regarding monitoring of the fuel pressure sensor. (The DTCs below show the default actions):

"
6.18 Fuel Rail Pressure Sensor
6.18.1 High/Low Input Failure
These are continuous monitors. The voltage from the sensor is compared to a failure threshold defined in the software. If the voltage is below the low threshold, then a timer starts to increment. Once this timer exceeds another threshold, then a failure flag is set and a DTC is stored. If the voltage exceeds the high threshold defined in the software, then a timer starts to increment. Once this timer exceeds a threshold, then a failure flag is set and a DTC is stored.
6.18.2 Stuck Detection
Stuck at monitoring executes when closed loop fuel pump control is executing. It checks that the fuel rail pressure signal has varied by at least 5 kPa over a range of demanded fuel pump duties. The maximum and minimum fuel rail pressures are updated each time. The change in demand duty is integrated and when the integral reaches 4%, the variation between the maximum and minimum values is checked. If it is less than 5kPa, failure judgment is made; otherwise, a normal judgment is made.
6.18.3 Offset Detection
This part of the monitor executes when the vehicle is idling. When closed loop fuel pump control is executing, a settle timer is incremented. After the counter reaches 5 seconds monitoring can be started. This is to allow the system time to settle after a transition from open to closed loop fuel pump control. Once the counter is greater than 5 seconds the target pressure is checked against the actual fuel rail pressure. If the error is less than the failure threshold, a normal counter is started. If the normal counter reaches 1 second, normal judgment is made. If the target to actual error is greater than the failure threshold, a failure counter is started. If the failure counter reaches 5 seconds then failure judgment is made.
"

'Failure' I imagine will result in the the relevant code being thrown. P0191-P0193, P1234, P1236 and P1338 are all related to fuelling issues. What is strange is that the car doesn't seem to think anything is wrong:- no codes thrown.

To repeat the section in the 2003-xk-service-training-guide:
"
The ECM use a frequency of 150 Hz (PWM) signal during average conditions, varying its duty cycle between 4 and 50% to control fuel delivery rate or turning it to 75% to turn off the pump.
"

I agree with crbass. You have to somehow break this loop. I appreciate that it's easy to say, but either a scope or SDD/IDS to see what's going on around the FPM.

She's encouraging you and baiting you at the same time. Nice to know, though

edit: take a look at this youtube video - it's based on a Ford. Although not identical, it gives quite a good testing overview:

 

Lol... Thank you Michael,,, but I have to careful I don't end up forcing myself (small NYC apartment) into having a "yard sale" - I have NO yard, for one... Eventually,,, for a lot of good reasons,,, I will need to and get an osilloscope that runs on my PC... For money reasons more than anything,,, this hasn't happened yet BUT thanks for thinking of me! Really!

 

Bass, Michael,,,,

You guys have said a LOT here. A good bit of it is over my head. I am going to come back to this to better digest it, where possible...

One thing (IF I read you right). Bass, just to make sure. My jimmied ground if rigged into the wiring harness that plugs into the pump. Yes, the R and YR wires emanate/ go back and forth between the FPM and pump,,, but the groundings interruption is not as much at the FPM than it is in the wires between the the positive wire on the pump and the ground return... I did try straight grounding the black wire on the FPM which didn't seem to have any effect.

And Michael... I have never had any of those codes... The ones related to fuel pressure... It seems I SHOULD with the rigged grounding and high pressures (70.1psi) that I have whenever the pump is running.

 

Yes, this is an essential mystery. Based on the notional operation, the car should throw codes based on fuel rail response (pressure much too high for commanded PWM input), among other things, when the input is grounded. Or, it should throw codes for no pressure with a commanded PWM input when not grounded. Or, if crankshaft position or other sensors aren't reporting...

So, one could conclude that the ECM has not determined anything should be fueled, thus no codes.

Just stabbing in the dark now, could it be that the system doesn't like the fact that your pump is in the air, thus didn't properly prime, thus the ECM concludes the car is out of fuel, thus it doesn't spin the pump to pressurize the rail knowing there is no liquid in there? Spins fine when you artificially make it 100% duty cycle even though the ECM doesn't want that.

 

Lol.... Although less technical (smarter questions) welcome to my mind hahahaha hahahaha... Especially because I know less than you,,,, it's driving me nuts!!!!

 

That was also one of my trains of thought, and led to my going off at a tangent... There is a data dialogue between ECM BPM and the key transponder module that is labelled 'OK TO FUEL'.
I haven't been able to find the raison d'être for that yet. It seems to be tied to the security system, although quite why fuelling would be treated as a separate item to inhibiting crank (why stop crank and allow fuel, or vice versa?) eludes me at the moment. I'm sure Jaguar didn't just do it for fun, and P1260? is supposed to be thrown if there's an issue with key recognition.

We really need to see those PWM outputs.

@Jay, I wouldn't be distracted by that diversion just yet. Clutching at straws here, but you introduced a variable by changing the FPM. Perhaps try refitting the original if it hasn't suffered water damage, or another if you have one?

Kudos for your perseverance.

 

Back at it... And look what I found!
Mouse in the house????
UNDER the ECU in the engine bay.

For ***** and giggles I bought a 25$ rail pressure sensor and was putting it in, noticed the colors, traced them and came to this plug, under the ECU... Colors look the same to me. But why an external plug out side of the ECU?

I don't know if this will change anything and am going to have a hard time checking. Messed with the CAN system someway while repairing the convertible top over my last two visits... Had the battery discon,,, needed to manually open the top, close it, run the windows up and down and mess with the sensor-ed hydraulic ram... CAN had a seizure I think and now she won't even crank... Tried the hard reset x5,,, one of the PARTS that need to be in place with the roof operation I think is out of wack. When I connect the battery after hard reset,,, the 1/4 windows auto go down...

This car is making me nutts!!!

Rail pressure sensor Plug... Under the ECU

Rail pressure sensor Plug at the rail... Craziness!!!

 

Good Morning...

Is the IP sensor in this diagram the fuel pressure sensor on the rail? RH front of rail, top of the engine.


curious about the location of this junction...


New discovery under the ECU. This plug connects to a (something) that seems to terminate under the ECU. I can't find it on any diagram.

This is the plug that goes into the rail pressure sensor at the rail.

 

Good Morning (and Good what ever time O day where you are) 😊

Got "layed off" yesterday thanks to Corona.

Still looking for help here.

Soooo, since my last post I replaced the fuel pressure sensor at the rail with a temporary cheeeepO from carID... Installed it and because I am dealing with another NO CRANK NO START issue, I just got to seeing this... I NEVER got the p0193 code before and that's WITH the pump running full blast (due to being directly grounded) and 70+ psi at the rail.

Then suddenly, a DTC code, GOOD NEWS, lol...

Is the PWM coming back to life?
What would YOU do or look for next?

 

I think that is a good code for a change as it is correctly determining that your fuel pressure is not where it should be. Have you tried undoing your manual fuel pump ground yet? If the system can see the fuel pressure now , who knows?

Sorry to hear about the layoff. I hope things get back to normal for you quickly.

 

Thank you, Mh... As do I, kinda. Not sure where you are in the word but our Prez has freed up some money that make my ends meet for this off period. If that happens, and I can blow up my XJS rewiring the fuel system, lol,,, something will have happened during the lay off...

I'm at the car now trying to straighten out some things I think I discovered (always dangerous) with the no crank, p1638, and that damn Realgauge set up... If I can get her running I will start experiments with the fueling...

A new pump for the top is waiting at the UPS,,, all arms in the top have been replaced that need replacing and,,, I'm looking forward to a green shower once I hook all that up. But, first things first.

At this point, the fueling (where she will run with) and the no crank first... Fingers crossed. I'm almost afraid to know whether this is going work or not. Know that feeling?

 

GOT IT..... SOLVED! FINALLY!

What a thing! Today I cut the INERTIA switch out of the system and the fueling went to RIGHT NORMAL... 55 +/- lbs per...
There's no telling when water is involved, I guess..
Seems the minor voltage difference in the 3 systems directly in line with the inertia switch was enough to disable the PWM and normal fueling.

CRAZY MAKER!!!!!!!

2 codes to go....