Hi, I have seen a few threads on here with a similar problem to mine, but I haven't found a solution yet, so I'm hoping you guys can help, as I'm really struggling now.

A few months ago my car developed a P0191 code, but initially it didn't affect the running. I run it on LPG, so I wasn't too worried about a fault on the petrol side as long as it would run long enough to switch to gas.
More recently it began stalling whilst driving, but would always restart without a problem and would do it once or twice before it was warm enough to switch to gas, then it wouldn't be a problem.

As it was getting progressively worse, I decided to do something with it. First I changed the fuel filter, then I swapped the fuel pressure sensor and finally yesterday I gave in and chopped a hole under the rear seat and changed the fuel pump. None of this has helped. It will still start first time every time and the fuel pressure according to the live data is around 53 psi initially, after a few seconds it will start falling drastically until the engine dies around 20-30 seconds later.

Is there something that can be telling the fuel pump to reduce its output by so much? All the things that I can think would affect it (MAF, vacuum leaks, lambda etc) would surely also affect it running on gas?

 

In the "Sticky" posts at the top of the forum you can find links to the repair manual with step by step checks for P0191.

I recently changed my fuel pump and I had hard starting after, as in I had to crank the engine for a long time to get it to run. I changed the fuel filter and that got much better. The fuel filter had been replaced not many miles ago the process of changing the fuel pump probably stirred up some muck. Doesn't sound like the same issue you have but fuel filter restriction is listed here as a potential cause of P0191.

 

Being in England, is your car petrol or diesel? Putting the info in your signature will be helpful in the future for others to identify challenges you may have. I say that tongue-in-cheek with your LPG conversion.

 

jaghelp.com: P0191 on your X-type

 

Hi, thanks for your replies. I started working through the steps in the manual and found high resistance between pin 2 of the pressure sensor plug and ground. Hopefully this is my problem. I'm going to splice in a new earth cable into it tomorrow and see if that solves my problem 🤞🤞🤞

Also, the car is a 2007 2.5 v6 petrol (with LPG) AWD estate

 

Assuming the IP sensor is good and that it has a good 5v supply and solid grounding tot he IP sensor. Then either the ECM is not getting good feedback from the IP sensor (I am somewhat doubting that because of your description), or you might have a problem with your fuel pump control module (located under the rear seat LHS of vehicle).
The fuel pumps on the 2.5 and 3.0L engine models are demand driven pumps, as there is no mechanical pressure regulator and return line in the fuel circuit.
Instead, the pump is driven in a variable way via the fuel pump module which is being commanded by the ECM that is receiving fuel pressure input readings from the IP sensor.

The fuel pump module receives a variable width pulse drive from the ECM at a fixed frequency rate of 150Hz on Pin1 of CA105 to the fuel pump module.
The pulse width modulation (PWM) ratio (the variation of on versus off - high/low) that commands the fuel pump module ranges from 4% on to 51% on depending on the what the ECM is calling for to maintain fuel rail pressure.

The fuel module in turn sends a status feedback pulse back to the ECM via pin7 of CA105 at the rate of 1Hz.
A 50% on / 50% off duty cycle is the fuel module indicating normal operation status back to the ECM.
A 25% on / 75% off is the module saying to the ECU it is receiving no drive signal, whereas 75% on / 25% off is feeding back that the fuel pump is currently off.

Unless you have a oscilloscope or digital meter which can display wave forms, you might be out of luck trying to get a reading on these. But it might be possible to get an indication of operation with a dwell meter as it will measure RPM and Dwell (duty cycle (on/off ratio) of your classic points based ignition systems.

If you are set on the 4 cylinder setting of your dwell meter, the 1Hz pulse rate may read as 15 RPM (60 pulses per min/4), whereas the 150Hz signal may read as 2250RPM (9000 pulses per min/4)
On the dwell reading should also vary depending on the duty cycle of the PWM signal.
Dwell will give you an angle reading which is the charge/discharge cycle of your old ignition coil system delivered by the points.
This could be a crude method of checking your PWM using what you might have in your garage tool kit.

My guess is that you might have a faulty fuel control module that is not reliably obeying ECM commands. You seem to have changed out everything else apart from wires.

Hope this helps....Good luck.

 

Your problem is in LPG installation. Fuel system without return line is very problematic to setup. You probably have some kind of emulator installed.
I also run car on LPG and have the same problem and so far I have following:
If I emulate fuel pressure while running on LPG between 379 and 381kpa, I will sometimes get P0191 but it won't affect petrol side.
If emulator goes bellow 379 or above 381kpa, ECU won't hold pressure on petrol, and car will stall. Reset for this is removal of 5A fuse F32.
If I don't emulate fuel pressure while driving on LPG fuel pressure goes to max around 480kpa and after few days I get same symptoms with fuel pressure dropping on petrol and P0191 code.
First thing I would check your LPG map, basis for every LPG system is to have perfect LPG maps so that petrol ECU doesn't try to add or remove fuel. You can simply check this by looking your fuel trims on LPG and then switch to petrol. They should stay the same.
So far I have not managed to remove P0191 completely, it comes back every few months.
My next plan is to make return line on fuel rail that will be active only while running on LPG. It should have 0.8mm nozzle, because that exactly simulates flow trough petrol injectors. Found few examples on LPG forums with this setup that work.
Hope this helps.

 

Can anyone shed any light on this too? The manual says that both pins 2 and 3 are WG. On mine only pin 3 is White and Green, pin 2 is Brown and Green. Which one should go to ground, pin 2 or the WG one?

 

Really great info, thanks h2o2steam. However I don't have anything that can measure frequency. Does anyone know if there is a bench test for the fuel control module?

 

Has anyone come across a pressure sensor that has failed? From what I've read, they don't really go wrong but I have 2 that have failed the resistance readings detailed in the manual. They both read around 1400 ohms.

 

Technically you could rig up and bench test the module for basic functionality and responsiveness.
I don't have this fuel pump module in my version (2.1 engine) so I can't fully verify some of the correct drive parameters, but with all care and no responsibility I have drawn a diagram of how you might use a frequency generator PWM module from eBay (just a few pounds) to insert the required specific 150Hz frequency and variable duration pulses into the module to then test and see the resultant outcomes. The pump module will respond to PWM pulse settings ranging from 4% through to 51%. You would want ot preset the frequency to 150Hz and the PWM to 4% before you connect it to the pump module input.

What I can't verify from the schematic is whether the PWM signal input of the pump control module is requiring only a 0-5v peak input or 0-12v peak input level. That makes a difference in what supply voltage you would supply to the PWM test module, as the output pulse amplitudes from this will equal the supply voltage you apply to it. Too much signal voltage might overload and damage the input circuity of the pump control module. You could power the PWM module from a separate 5 volts supply (3 x 1.5V batteries in series would do). Just make sure you have all the earths (negatives) joined together so the two supply lines are relative. Too little signal will not trigger the fuel module input.

Using a small 12v lamp across the output pins of the fuel pump module (not LED) as a dummy load should give you a visual output so you can identify pump drive voltage changes - the brighter the glow the more the pump is being driven.
If you are also wanting to verify the 1Hz output operation of the module that feeds back to the ECM, I suggest a simple resistor and low current LED (20mA) will give you a responsive visual reading on the change of output state flashing at a rate of once per second.
It is just the the duration of flash that changes not the frequency.
Remember that a 50% on / 50% off duty cycle is the fuel module indicating normal operation status back to the ECM.
A 25% on / 75% off is the module saying to the ECU it is receiving no drive signal, whereas 75% on / 25% off is feeding back that the fuel pump is currently off.
Again I can't tell you if the output pin on the pump module if sending a 12v pulse back to the ECM or if it is grounding a high signal line, so I have drawn both scenarios for testing (either/ or). One will work if the other doesn't.

I really wish I had a module that I could rig up so I could be exact about PWM drive voltage and ECM feedback pin monitoring orientation, but for what it is worth these were my thoughts if you really want to try and bench test your unit.




Frequency generator with pulse width modulation

 

You are all neglecting the fact that he is running his car on LPG. his fuel injectors are closed all the time, so if his fuel pressure emulator is not working properly it's normal to get DTC code.
Also, if LPG map is not setup properly his fuel trims are off, and that causes symptoms he is having.

 

The reason that I don't suspect the LPG system is because I have this issue when running on petrol with the gas system disabled. Surely if it was related to LPG, it would only present a problem when selected?
On a separate note, I ordered a new fuel pressure sensor from SNG Barratt, as my other 2 failed the resistance test as I've mentioned above. Strangely, the brand new sensor doesn't show any resistance readings across any of the pins, does this mean it is DOA or will you only get those figures once it has been used?

 

Reset ECU, then run it only on petrol and see if you have the same problem.
You have the same problem as many of x-types running on LPG...

 

Do you mean reset the petrol ECU? My basic scan tool doesn't have that facility. Or can I do it by removing the fuse for a period of time? I'm still dubious that the LPG system is affecting this, as I have had the car nearly 3 years and it has only become a problem recently. 🤷‍♂️
Also bear in mind that I still need to find and fit a pressure sensor that returns the correct resistance readings as shown in the manual before I can rule that out.

 

Thanks again h2o2steam for your incredibly detailed response. I really appreciate you taking the time to go through all that, however I fear this is somewhat over my head. Even with the correct tools, I'm not sure I could replicate what you have shown in your diagram. I hope I don't end up down this road, or I may never find my way back! 😂

 

You can reset it by removing fuse F32, 5A or by disconecting battery.
As you say in first post, you had sensor that reads 51psi, return that one.

 

UPDATE:

The car is now running fine since I repaired the earth supply going to the pressure sensor. It didn't fix it immediately, but strangely it is OK now and I've been driving it for weeks without problem. It did give me the P0191 code again a few weeks ago, but it didn't affect the running and it hasn't come back since I reset it, so hopefully that's problem solved. Thanks for all your advice.

 

FURTHER UPDATE: It has started being a problem again 🙄 😥
It's started doing exactly the same again and I finally bit the bullet and bought a new pressure sensor from Jag and it has made no difference! 🤬
As I have no way of testing the fuel pump control module, should I try swapping that?

 

[QUOTE=Stevegledhill;2433416]ULTERIORE AGGIORNAMENTO: ha ricominciato a essere un problema 🙄 😥
Ha ricominciato a fare esattamente lo stesso e alla fine ho morso il proiettile e ho acquistato un nuovo sensore di pressione da Jag e non ha fatto alcuna differenza! 🤬
Dato che non ho modo di testare il modulo di controllo della pompa del carburante, dovrei provare a sostituirlo?[/QUOTE
]


hello, excuse the question, I have the exact same problem with LPG how did you solve?