[welly]

All, first post as far as I remember, but have visited many times and am looking for thoughts from the knowledge pool.

Cutting to the chase - E92 335d, 2006 pre LCI, 190k on the clock, custom tune with various supporting mods.

Had a map tweak a few weeks back, first chance to give some beans today. MIL with limp is the result. Code logged is 004560 which I know is rail pressure below requested. No other changes (map tweak was for DPF removal) and was fine on the dyno.

Not sure what the request is from the tune file, however I have asked and will find out. Tried to log requested/actual but Torque extended PIDs aren't working for me so gave up in the end. Will try tomorrow with ISTA.

Running a stock R70 CP3 pump currently, as far as I know with engine mileage.

I know for a fact that the rail pressure request was turned up in the new file, and logging actual pressure shows max achieved being circa 24500 PSI (so around 17000 mbar), and this is clearly less than is being requested. This seems very close to the reliable limits for the R70 based on my knowledge of the unit.

Fully aware that this fault is a bit of a git and can have numerous causes. Fuel filter is old and will be replaced, but was working fine previous. I'm also accutely aware that the LPFP has seen a few Christmases and may not be tip top, but again, no issues before. Sensors seem to be working fine from logging responses.

I will be buying a used R90 and fitting next weekend along with other work that needs doing (bottom pulley, belts, water pump), but interested to know what thoughts are on whether that will solve the issue....

[Yozh]

I think you answered your own question. And you made a really good analysis. However, without looking at the logs, you would not know for sure and May be throwing parts at it. Could be so many things. I suggest yiu download TestO and log your lpfp side then HPFP side of rail control.

[welly]

Originally Posted by Yozh

I think you answered your own question. And you made a really good analysis. However, without looking at the logs, you would not know for sure and May be throwing parts at it. Could be so many things. I suggest yiu download TestO and log your lpfp side then HPFP side of rail control.

I'll look into this, not an application I've used before but will give it a spin.

By way of an update, progress has been made, not good progress though sadly.

R90 installed successfully (along with bottom pulley and water pump because why not). Using ISTA to prime the system wasn't a huge success, long crank before eventual start.

Issue is now significantly worsened. Any use of WOT now produces the same codes as above. Absolutely fine on part throttle so the car remains driveable.

I've manually tested supply fuel pressure at idle after the filter, seeing between 38 and 40 psi so roughly around 2.7 bar at idle. This seems lower than expected as I believe this should be around 5 bar.

Based on these developments, I'm suspicious of either the rail relief sensor/valve and the lift pump pressure. Interestingly, I've had a fault code since owning the car for 'Fuel Filter' -the code escapes me. I wonder if this could be a manifestation of reduced supply pressure with the DDE listing a blocked filter as the cause... I've seen this fault over the forums over the years with some people unable to resolve.

I happen to have spares of both suspect items, so I'll fit these tomorrow and see what the result is.

In the case of the issue persisting, ill crack on with some logs.

[welly]

Rail change has made no difference whatsoever. It is a good second hand part with both sensors so I would be very surprised if it has failed in exactly the same way.

Was unable to do the LPFP as I would not get the bloody collar off and was starting the damage the tabs. I will be making a tool for removal as I've seen long ago on another post on here.

I DID however, (as suggested) download Testo. The standalone version simply wouldn't work for me so I had to break out the old EDIABAS laptop and fired it up on that.

Have done some logging today, well as much as I can before CEL and limp mode kicks in. Some very odd behaviour to be honest.

WOT pressure request appears to be off the measurable scale as it simply stops at around 1680 bar and is the same number from then on.

Actual is all over the shop.

I don't have the option to log pre supply pressure in anything other than litres per hour, and as I am going to assume that is based on voltage supply to the pump, it really doesn't mean anything. I will be driving with the pressure gauge in the line and in through the window to review.

Still suspect of the LPFP with the results below, but thoughts welcome...

[welly]

Can anybody please advise on what the pre supply pressure should be, or is on their 335d, or others as I'm assuming that they're all the same.

Pump swapped out but pre supply pressure is only fractionally more at 42 psi.

[Papogator24]

Originally Posted by welly

Can anybody please advise on what the pre supply pressure should be, or is on their 335d, or others as I'm assuming that they're all the same.

Pump swapped out but pre supply pressure is only fractionally more at 42 psi.

I'll check my photos. I manually Teed the fuel supply from lpfp and recall seeing between 50-60 psi.

[Papogator24]

This is with the car idling.

[welly]

Originally Posted by Papogator24

This is with the car idling.

Mint, you're a star. I'm seeing significantly less obviously, however to have the same pressure on two (different, I had to modify the pipe work as the lpfp was from a newer 335d) pumps suggests it's unlikely to be coincidence.

I wonder if the fuel control valve on the used R90 is allowing too much of the pressure through and it's returning back to the tank as bleed off from the rail. You would never see it really, and it would certainly cause the pre supply pressure to reduce without a clear reason.

Having done some post lpfp logs, rail pressure can happily reach 1650 bar actual against a commanded 1650 bar (or thereabouts) on part throttle. This is only a WOT problem, and doesn't happen exactly the same every time. Examples:



WOT



Really odd problem this one. I'll swap out the fuel flow control valve with a new one (both used so far are used) if no dice, I'll start looking at software tune as a possible cause. Feels hardware related though.

[welly]

Have now done some more logging, and significant testing of various components. Interestingly, it seems that there isn't a thread on here which ties the whole fuel system together as a whole, including specs of how various things are designed to work, so once I've got to the bottom of this, I will give a comprehensive explanation to help other in future.

Have done a quick and dirty leakdown test using my old home made leakdown tester from years ago - nothing really to report. I'm not seeing any real abnormalities in diag to suggest an injector, and the leakdown test is pretty even stevens at a range of idle to 2000 rpm spread - photos below:





I've also done some logging whilst waiting for my new VCV to arrive (which it has today). I've expanded the logging to cover off both of the two point control sensors on the system, those being the Volume Control Valve and the Rail Pressure Control Valve.

Some interesting results, and a definite pattern as can be seen below:



To understand the graphs, know that the more open the RPCV is, the more pressure is being bled from the rail. The more open the VCV is, the more fuel is being delivered to the pump for rail injection.

Runs are left to right, top graphs show rail pressure vs actual and the bottom shows VCV and RPCV actuation (but reversed to show percentage open, not percentage closed as the output duty cycle is backwards). As you can see, the rail pressure is all over the shop, but in a repeatable fashion.

Rail pressure starts to climb with requested, however fails to deliver and has a significant dip. Then it recovers back to requested, and then dips again, before recovering again at roughly redline.

Graphs show the RPCV relatively static at ~35% open (which on the output would be ~65% duty cycle - as I said, it's backwards) and holds that quite happily - doesn't seem abnormal, and can maintain the rail pressure at idle by itself when VCV is unplugged.

To manage and/or achieve rail pressure where is isn't able to, the DDE is requesting more duty cycle from the VCV to deliver more fuel into the HPFP to deliver to the rail - the resultant effect being the climbing of the rail pressure.

Once requested rail pressure is met, the DDE reduces the duty cycle for the VCV which results in the pressure dropping away from requested once again, so the above happens again. Seems very much that the DDE is having to intervene as the expected behaviour is not happening. Ultimately this is what a plausibility fault means, what was expected to happen based on parameter x and y being true, didn't.

Interesting that the RPCV doesn't increase duty cycle to help, however I am going to assume this is standard behaviour and that the VCV deals with this by default.

I am going to further assume that the duty cycle of the VCV is defined in the map initially based on an expected supply fuel pressure (within a tolerance) of say 50 - 60 psi.

As my LPFP is delivering at approx 42 psi at idle, all of the investigations done are starting to point to the supply pressure being low for what is a currently unknown reason, or a failing VCV.

Whilst this problem is annoying, I now know an awful lot more about the fuel system on the M57.

I'll keep updating as I have noticed a lot of threads seem to just go dead with no resolution listed - which is frankly bloody annoying!

[Nadir Point]

Those CP3s will generate pressure with little to nothing on the lift side at marginal throttle levels. You're still barking up the wrong tree.

[welly]

Originally Posted by Nadir Point

Those CP3s will generate pressure with little to nothing on the lift side at marginal throttle levels. You're still barking up the wrong tree.

We're not talking about marginal throttle levels though, we're talking about a problem only at WOT or high load. If I'm barking up the wrong tree, a suggestion as to the right tree would be useful would it not?

With respect, that post is about as much help as telling me that it's the giggle pin in the laughing shaft.

[robnitro]

I'm not sure how the 3er are, but if you can log the lpfp pressure, it would help.
Liters per hour is not helpful, as even if pressure drops, it could be flat on the flow, as the fuel that isn't going out the injectors gets dumped into the return, i think.
Do you have an Android device with otg cable to plug the USB dcan into?
Or do you have a Bluetooth device?
Try the app linked here ( https://play.google.com/store/apps/d....pw.btool.lite ) if so... It can show lpfp pressure . If you need the full one, direct message me.


Seeing the other graphs for the other regulators and how wot vs just part with commanding high pressures make me think it's neither at fault.
They aren't hunting or undershooting....

[welly]

No sensor on this engine for monitoring fuel pressure on the low side, which is annoying. Ive searched through everything, no dice. There are parameters available, but they only function for the X5 and I just get a baseline figure that never changes so no joy.

I'll be manually logging fuel pressure under load using a gauge, it just needs modifying to be long enough.

I agree, I am happy that the high pressure side is working based on graphs. I changed the VCV and no difference, but that wasnt really a huge surprise.

I cannot figure out for the life of me why I only have 40 psi at idle supply pressure when everyone else I've spoken with has 55 psi ish - I'm going to chase this problem for now as the graphs look and feel like fuel starvation from experience, and a bar adrift is a reasonable amount.

Anyone got any good calls on the low supply pressure? EKPS module for example. In tank pump has been replaced with a used pump already. As I said before, I have a 'Fuel Filter clogged' code in the DDE which has been a long standing code. How does the DDE come to that conclusion though. I will likely change the fuel heater just to rid the code, but any ideas appreciated.

[welly]

For reference as I couldn't see anything anywhere regarding what the supply fuel pressure should be, this is direct from TIS for 2006 335d.



For reference, my supply pressure is currently ~2757 mbar so not meeting the spec.

Does anyone know where the EKPS gets the required voltage information from? Does it come from the DDE or is it hard coded into a map on the module.

[robnitro]

Low Pressure Fuel System - E90
The low pressure fuel systems differ between the E70 and E90. The E90 is a “speed regulated” system which means that the fuel pump speed is regulated by the EKP module based on request from the DDE.
The fuel pump will be activated with the “ignition on” signal. If the engine is not started, the fuel pump will be switched off after a defined time period. When the engine is switched off, the fuel pump is switched off as well.

Low Pressure Fuel System - E70
The low pressure fuel system on the E70 is a “pressure regulated” system which uses the signal from the fuel pressure sensor located in the low pressure fuel line.
The fuel pump operates with "ignition ON". If the engine is not started, the fuel is switched off at a specific pressure. When the engine is running, the fuel pump is regulated on-demand by the EKP module in response to a load signal from the DDE in order to ensure a uniform fuel pressure at the inlet to the high-pressure pump.
The functions of the low pressure fuel system are integrated into the DDE control module. The DDE uses the pressure information from the combined fuel pressure-temperature sensor to determine the current actual pressure in the low pressure system.
In order to maintain the approximate delivery pressure of 4.8 to 5.0 bar, the DDE uses a number of input variables. The input variables relevant to determining the adjusting value are:
• Actual pressure in the pre-supply system
• Engine speed
• Injection volume
The adjusting value is sent from the DDE to the EKP module in the form of a CAN message.
Fuel Pressure-temperat

[welly]

Originally Posted by robnitro

The adjusting value is sent from the DDE to the EKP module in the form of a CAN message.

Very useful indeed, thanks.

This feels more and more like a software problem as we go....

[robnitro]

Originally Posted by welly

Very useful indeed, thanks.

This feels more and more like a software problem as we go....

I'm still confused as to how the dde determines how much fuel to flow in your setup.

Have you changed the fuel filter?
I don't think the fuel filter heater will set off the clogged filter code, unless the fuel is so cold that it's restrictive in going through the filter:

Here's the pdf https://archive.org/download/BMWTech...ogy_WB_web.pdf

Fuel Filter Heating - E90
On the E90, the fuel filter heater is not controlled directly by the DDE. A pressure switch and a temperature sensor are located in the fuel filter housing.
The fuel heater only works with the ignition switched on and when both of the following conditions are fulfilled:
• Temperature drops below a defined value
• A defined fuel delivery pressure is exceeded due to cold,
viscous fuel.
If the filter is clogged, a corresponding signal is sent via a diagnosis line to the DDE. This is the case when, despite a sufficiently high temperature, the fuel pressure upstream of the filter does not drop. The conditions for fuel filter heater operation are as follows:
• The fuel heater is switched ON when - the fuel pressure is greater than 6 bar AND the fuel temperature is less than 2°C. • The fuel heater is switched off when - the fuel pressure is less than 5.5 bar for a duration of greater than 5 minutes OR
• the fuel temperature is greater than 12°C OR
• during the starting process if the electronics in the fuel filter detect a battery voltage of less than 7.5 V for longer than 0.2 seconds.
The fuel heater is not activated by the DDE control module. However, the fuel heater reports a detected filter blockage via the signal DIAG_DKH to the DDE control module. The DDE control module then stores the fault.

[welly]

Yeah, changed the filter early doors in this problem, so definitely not likely to be an issue as it's an OEM filter.

I guess the fuel filter heater may just be faulty, or giving false readings. I do have the related code logged in the DDE, and I wonder if that impacts the flow requests sent by the DDE to the EKP module - I cant find anything definitive one way or another to say whether that's accurate or not. Makes sense that if there is a restriction reported, the DDE would reduce flow to save stressing the LPFP.

Again, as you've already said, how the DDE determines fuel flow requirement to send to the EKP puzzles me as well. The 335d low pressure system hasn't got a huge amount of intel on it - even TIS isn't very useful beyond expected pressures.

[welly]

Also, that PDF is epic. Where did you find that?

[robnitro]

Originally Posted by welly

Also, that PDF is epic. Where did you find that?

I had the pdf for a while but couldn't attach it for some reason.
So i just searched the words of the name of the document and got a page on archive.org with this and others in there.

https://archive.org/details/BMWTechn...iningDocuments

Link for every pdf, in one file here
https://archive.org/compress/BMWTech...gDocuments.zip

[robnitro]

After the fix with the rail pressure sensor, I logged low pressure fuel.
It is 3.9 bar steady at any rpm and load. Not sure why the pdf said 4.9, unless they mean absolute, which is +1bar.
So if you figure almost 4 bar is 60 psi.

Can you get a fuel hose and Tee and use that to watch your low side while driving and come up with the condition? I've read that the cp3 pump actually can run without any pressure (in some applications there's no lift pump!), it's suction capable, like the vw tdi rotary ve pumps.

But of course you don't want to run vacuum under load, with a fuel pump, because many of them aren't designed to flow through, like the later vw tdis that had a lift pump

[tleader]

I've briefly looked into my EKP and why I was not getting the 60 PSI of pressure according to those docs. I was also not getting any activation of the LPFP with the "key on", again as listed in those docs. I was told the diesel does not activate the LPFP with the key on, but only while cranking and once started. I also found out that my LPFP would receive about 8 volts of power when at idle and that was the reason for the reduced fuel pressure. But, because it is a "speed" related system, that was normal for idle. I jumped my LPFP and gave it a solid 12 volts and it increased the pressure up to about 58 PSI.

I tried another EKP and that didn't work for me, but when I recoded the original EKP, it increased the pressure a bit more according to my inline fuel pressure gauge. Now my problem is not exactly like yours, but I never really got to the bottom of it because my fault was very sporadic and far between one another.

It would be nice to know if the voltage and pressure do in fact increase while accelerating, that is one thing I did not log. According to the docs, when accelerating, the voltage should increase at the LPFP while increasing the fuel pressure simultaneously. Maybe jump and run the LPFP at 12 volts the whole time and see if it remains steady.