Jaguars fuel pressure fault code again!

[57jam89]

Hi all,
As many of you will know i fitted a Landi renzo system to a Jaguar 3.0 s type a few mths ago and within a few weeks had the fault code P0191 Fuel Rail Pressure Sensor Circuit Range/Performance.
So i fitted a AEB adjustable fuel emulator.
So i watched the fuel pressure then running on petrol and set the emulator to the same and then adjusted it every day for a week to get it spot on and its been fine, no engine light or code until now.
Then it starts cold or hot start (then its on petrol) the pressure goes to about 480Kpa and the normal pressure for petrol is about about 375kpa and the emulator is set to 372kpa.

Am i right in thinking maybe a fuel return would be best suited to this car??

Thanks Simon

[rossko]

Never had any luck with emulators. When you think about it, you won't be able to set your fixed emulation signal to exactly what the petrol ECU has as a target, so it'll always be a bit high or low. Over a long run, the petrol system will keep trying to tweak pressure with no response. Sooner or later, it will probably complain.
The problem is compounded if the "target" pressure is variable - it is on turbo engines but I don't know about this one.
It's all dependent on the petrol ECU inner algorithms which we can't see and might vary from version to version.

It is tricky 'tuning' the size of a solenoid controlled leak-back too. It's not perfect either, but at least pressure really will vary in response to pump drive commands, so it ought to keep the petrol ECU happier.

[Tubbs]

I disagree, i have had good results from pressure emulators on the 2.5 and 3 litre engines. You only sample the wire afterall. You dont cut into it. So any correction the pump makes is registered. But i do have one that seems to be giving me a constant set reading for some reason, which it shouldnt, so im suspecting its the unit that is dodgy. Im off to swap it for a different one next week. Ive never seen one do that before. They usually toggle the pressure up and down slightly dependent on throttle response giving the pump a nudge here and there. So stick with it. They are a PITA to get right. But it is possible. Just fiddly. Funnily enough i had a chap contact me about one the other day for a conversion. I told him he would have to expect to bring it back a few times because the engine management light would come on. They are a bit tricky to set up etc etc. I stopped myself at that point and said 'I'm not really filling you with cofidence am I ?' he replied 'well at least your being honest about it' and promptly booked it in.

[rossko]

They usually toggle the pressure up and down slightly dependent on throttle response giving the pump a nudge here and there.

Really? I've no experience with one like that, which would explain why I gave 'em up.

[Tubbs]

Thats what the unit should do. But as i say i have one out there thats not. And putting the k light on !

[rossko]

This is the AEB394 "J type"? Cant see how it can know anything about throttle response, there's no connection
http://www.aeb.it/products/emulators/394
perhaps its just a periodic waggling.

[Tubbs]

Sorry, im just presuming the change in pump pressure is connected to throttle response. It just seems to be directly contributable. Mind could just as easily be a response to increased injection times, air flow, map etc....just to avoid any misunderstanding it isnt connected to the TPS or anything. Just the one wire on the fuel rail.

[rossko]

Don't be sorry, its interesting to find out what we can expect from these things.

This one can only be applying a fixed clamp voltage or load to try to override the pressure sensor signal. The clamping won't be perfect, so large changes from the real sensor will appear as small variations still. Depends on the host petrol system if that happens, most could be changing pump drive to suit engine loads.

Still not going to work well on a car (probably only turbos?) that changes the target pressure under load. But it all depends on the petrol host, finding out just what the systems do and expect to see is damn difficult, and there's nothing to stop them changing the game from time to time.

[57jam89]

Thanks for more inf on the emulator, but my problem is the cars ecu is happy then its on lpg, but its then it starts on petrol.

As i understand this the emulator fools the ecu to think the petrol rail is ok then on lpg but as soon it switchs over or starts on petrol the ecu then sees the high pressure and then i have engine light/fault code.
I have checked this by cleaning codes then running on gas (pressure 369Kpa) and fault code does not return until i switch over or restart engine (pressure 483Kpa) on petrol. If i drive on petrol for about a hour the pressure slowly returns to about 375kpa and then it fine for about a week, then its back to fault code/high petrol pressure again.

Have i got this right?

Thanks Simon.

[rossko]

I have checked this by cleaning codes then running on gas (pressure 369Kpa) and fault code does not return until i switch over or restart engine (pressure 483Kpa) on petrol. If i drive on petrol for about a hour the pressure slowly returns to about 375kpa and then it fine for about a week, then its back to fault code/high petrol pressure again.

That sounds about right. The pump system is a learning system; there's a target pressure, it looks to see if it should turn the variable pump drive up or down to achieve the target. And remembers how hard to drive the pump.
Say the target were 374 and your emulator shows 369. The system will add a bit more pump drive, but nothing changes, so it adds a bit more, and so on. When you later switch to petrol, its doing far too much pump drive for normal use.
If you set your emulator to 376 and the target is 374 - the opposite happens and pump drive reduces and reduces. Folk have ended up with non-starting cars when it goes that way, so best to approach from the other side like you are.

[57jam89]

Ok, so say the target was 375 and the emulator is set on the same then the pump drive will not change, but whats not the true pressure is it?
If the engines running on petrol and pressure is 375Kpa then it switchs to gas the emulator turns on and the ecu still sees 375 but the true pressure in the rail has just gone up due to the pump drive is the same but no petrol is being used, so as soon as the the engine switchs over/starts on petrol the ecu will always see high pressure??

[rossko]

Ok, so say the target was 375 and the emulator is set on the same then the pump drive will not change, but whats not the true pressure is it?

That's correct. The true rail pressure will be up on the top limit, as the pump will be driving normally into a dead-end while no petrol is being used.

If the engines running on petrol and pressure is 375Kpa then it switchs to gas the emulator turns on and the ecu still sees 375 but the true pressure in the rail has just gone up due to the pump drive is the same but no petrol is being used, so as soon as the the engine switchs over/starts on petrol the ecu will always see high pressure??

Yep. But - it will drop quite quickly back to normal as the fuel is now being used. So long as the memorized pump drive is normal-ish! The petrol ECU doesn't ring alarms until things look wrong for some time. This sort of thing could happen on a petrol-only engine parked up with 'normal' pressure in the rail, heat soak will cause a pressure rise. So they have to allow for it.

The emulator problem is that the pump drive might not stay normal-ish, but creep away in one direction or another because the closed-loop feedback has been broken.

[stype3]

Hi All,
if you want to be able to reliably use petrol and lpg, on the S type, then the best way is to put a bypass in the petrol line. Put in a t piece, with the straight through part going to the injectors, from the third leg of the T put in a restrictor then solenoid valve and return line, feeding into the metal part of the petrol filler pipe is probably best.

The ECU will cope enabling a seamless transition from LPG to petrol and back. Having said that if you run the petrol tank almost dry, it will give strange results

[Tubbs]

The conclusion I have recently come to regarding the aeb fuel pressure emulators is .... They only seem to work on Bosch and not Denso. As jaguars have mostly Denso systems now I think they are pissing in the wind. So I spoke to a mate, who is a jag specialist and been in the gas trade for years. He confirmed the same in his experience. But he told a really neat solution, so I thought I would share. Go down the motor factors and buy a new fuel filter. Tap a spud onto the inlet side of the filter seal it in with loctite on the thread. As it comes out put a .8 mig tip in the fuel hose with a clip around it to hold it in place. Back to a shut off solenoid, tap a spud into the filler neck. All in all about 20 mins work. Job done. No more fuel pressure problem. And a new fuel filter into the bargain
But thinking on, I have a few Denso ones. Pretty late ones too with pressure emulators on that seem to be absolutely fine. Never checked the k light on at all.

[rossko]

So much depends on the software design of the pump management - what does it expect to see under different circumstances, what does it do about it, when does it ring alarms. Actual results almost certainly depend on driving pattern. A single-level emulator is doomed to failure in my view.

[Tubbs]

I was surprised the other day. I had an st220, 07 plate in for conversion. Expecting to see a fuel pressure sensor on the rail. No return, so I was expecting to have to fit a return. I plugged in my reader and it didn't have any read out for fuel pressure. Just air flow in the menu. So I thought I would suck it and see. Luckily It was with me a week so no stress. Whilst running the front to rear I clocked where the filter was. Had a look and it has a tee on the outlet of the filter returning back to the tank. So the job was already done for me. Also it had proper cables connecting the pedal to the throttle. But stlll the same 3.0 v6. The most hassle I've had from a jag was the 3.0 xj6. The ally one with the air suspension tank in the spare well. Got it sorted in the end though. It would be fine for 1500 ish miles. And eventually it would put the k light on and go into limp. They do all kinds of funky things in the ecu. Like pull the fuel trims about, expecting the fuel pressure to change I would presume. Eventually causing it to limp. Bloody thing gave me stress. And a few train rides to Manchester. I think as Ross says its a lot to do with driving style. And the chap was a big commuter. 5-600 miles a week. So I got to the bottom of it pretty quickly ! It was going wrong once a fortnight. I fitted a fuel return. It's not missed a beat since. The customer is over the moon with it. Nice car actually. Just goes to show you. Doesn't matter how long you have been in this game, your always learning. And we all get problems every now and again. I'm not proud, I'm quite happy to admit when my origional opinion was wrong

[57jam89]

Well i removed the emulator which i fitted. I would set it up and the car was happy then about a week later engine light back on, readjust and again a week later light back on. So i removed it and fitted a petrol return.

[Fabian]

Hi all,
I have just converted a Jaguar 2.5 X-type. So I had fuel pressure problems. As I have no fuel emulator at the moment I did a fuel return. After that car has P01236 code (control pump). I did a drill on petrol rail and I screw a 1.75mm diameter tail. If I drill more do you think I can solve this problem meanwhile I receive fuel emulator?

Thanks in advance

[stype3]

Hi,
for a fuel return line to work, you need a restrictor in it. Its probably best to put a T piece in your pressurised fuel line. put the relief valve on one let of the T make up a plug to fit in the inlet side of your relief valve, and drill a small hole in it. (you need the pump to be able to generate some pressure, not too much and you only need some fuel to go back to the tank.)

An alternative is to put a purpose made fuel pressure relief/control valve in to the fuel circuit. These are made for people who do high performance tuning of more ordinary cars.

best regards
and good luck

[rossko]

You'll need to use a live-data OBD reader to see the fuel pressure, and find out if your restrictor needs to be bigger or smaller. Run on petrol while driving at 30mph or so, switch to gas and observe if the pressure goes up or down.

As we've said, the emulators don't always work anyway. Better off making a nice job of your return system for a long term solution.