Stubborn Purge Valve codes P0441 / INPA 250

[e39_touring]

So, I'm finally down to 1 lousy error code on the wife's old X (3.0) that I've been prepping to sell for like a year now (it takes that long when you only have time to work on it one afternoon every other month or so). It's the INPA 250 Functional Check TEV or OBD P0441 Incorrect Purge Flow.

Why can't the codes on this car be addressed by the normal/common fixes!?!

So, here's what I've done:

1) Remedied all vacuum leaks, which have alleviated all other codes - had to replace intake boot and dipstick to CCV hose. Smoke tested intake, normal fuel trims, and no related codes leaves me thinking I'm all good on vacuum leaks.

2) Replaced the TEV/purge valve with a new one (BMW original).

3) Swapped gas caps with my other X which has no codes.

4) Ran DTML cycle test in INPA multiple times. Result always a pass with everything intact. Result always a fail with 1mm+ leak if I leave the gas cap loose or disconnect purge hose from TEV/purge valve. This tells me the system is sealed and nothing's blocked between DTML pump, purge valve or tank.

5) Tested the new TEV multiple times - with engine running and valve closed, I feel no vacuum at port that goes to canister. With valve open at 20% or 99% (the two activation values INPA triggers), I feel good vacuum.

Everything seems to check out, and I am perplexed as to why this code keeps popping up after ~30 miles of driving.

My understanding of this TEV code is that when the valve opens, the DME looks for a corresponding change in the O2 sensor readings (lean?). If it doesn't see that, it opens the valve the next time the car's at idle and looks for a change in RPM. If it doesn't see that, then it throws the code and sets the CEL.

My O2 sensors, while probably original (140k mile), seem very responsive and don't cause any other problems/codes. The only other thing I can think of seems a remote possibility - a partial blockage in an EVAP hose or clogged air entry point into the EVAP system.

Does anyone know definitively where the air should enter the EVAP system when that purge valve is open? It's got to be either the filter on the DTML pump, or does the gas cap have a one-way valve that lets air in but not out? I think the only thing I haven't ruled out is some sort of flow restriction.

Has anyone run into a tough-to-eliminate P0441? All my searches lead to stuff I've pretty much ruled out (gas cap, new TEV valve, DTML pump, etc.).

[80stech]

Does anyone know definitively where the air should enter the EVAP system when that purge valve is open? It's got to be either the filter on the DTML pump, or does the gas cap have a one-way valve that lets air in but not out?

That's a good question, I have wondered that myself. The purge valve purges the charcoal canister so your problem might with a restriction or saturation there. I would imagine that the air inlet would be on the charcoal canister or a hose to it. There are some pretty good hose diagrams(it is fairly complicated) in the Bentley manual.

[e39_touring]

Originally Posted by 80stech

That's a good question, I have wondered that myself. The purge valve purges the charcoal canister so your problem might with a restriction or saturation there. I would imagine that the air inlet would be on the charcoal canister or a hose to it. There are some pretty good hose diagrams(it is fairly complicated) in the Bentley manual.

I think the air solely has to come in from the filter on the DMTL input and not the gas cap. There really aren't that many hoses as part of the system. I found this diagram in some BMW training material, and it lays it out pretty well.

It also says 'The pump also contains an ECM controlled change over valve that is energized closed during a Leak Diagnosis test. The change over valve is open during all other periods of operation allowing the fuel system to “breath” through the inlet filter (similar to the full down stroke of the current vacuum operated LDP).' in the functional description.

I blew some low pressure air/smoke (<5psi) back through the hose that goes from the canister to the purge valve, and smoke did come out of the rear fenderwell where the DMTL pump is located, so I assume that path is clear. Maybe I'll hook up a vacuum pump and gauge and see what kind of vacuum I pull going the other way.

[andrewwynn]

My bet; a bad connection on one of the myriad of hoses in the system. I think the test applies pressure so you can run the test to pressurize and then spray soapy water to locate. Also any leak in the fuel system so check the main tank seals under the back seat and those fittings. Look with your nose! You shouldn't be able to smell gas vapor anywhere under the car. Find gas smell you found the leak. Run the purge test

[e39_touring]

Originally Posted by andrewwynn

My bet; a bad connection on one of the myriad of hoses in the system. I think the test applies pressure so you can run the test to pressurize and then spray soapy water to locate. Also any leak in the fuel system so check the main tank seals under the back seat and those fittings. Look with your nose! You shouldn't be able to smell gas vapor anywhere under the car. Find gas smell you found the leak. Run the purge test

I've run the pressure test many times in INPA, and no leaks are detected. I also don't smell gas anywhere. To make sure the pressure test is working, I re-ran it with the gas cap loose and then again with the hose from the canister to purge valve loose. Both of those tests detected a >1mm leak.

At this point I feel the system is leak free, but as 80stech suggests, there's got to be a restriction in either a line, the canister, or the DTML filter. I'm going to see what kind of vacuum I can pull at the purge valve and maybe do the same on my other X5 which has no codes and compare the two. It's nice to have a 'Chinese blueprint' for these sorts of things

[80stech]

That looks like a generic diagram. There is a lot more going on if you check out the diagram and pics in the Bentley manual.

[e39_touring]

Originally Posted by 80stech

That looks like a generic diagram. There is a lot more going on if you check out the diagram and pics in the Bentley manual.

No, it's definitely specific to the M54 in E53 - it came from a training module covering what's new in the M54 in E53.

It matches what I have in my X5. This is only the EVAP system, which is pretty simple. If you add in CCV, secondary air pump, etc., then there are a lot more hoses.

[andrewwynn]

From the diagram I see a possible issue: there is a calibration orifice. If that got plugged the system will think any actual leak is much bigger.

[e39_touring]

Originally Posted by andrewwynn

From the diagram I see a possible issue: there is a calibration orifice. If that got plugged the system will think any actual leak is much bigger.

True that! However, in my case the system is not detecting any leak, but it is saying it doesn't see the change in mixture that it's expecting when the canister purge is called for. The M54 has separate codes for EVAP leak, purge valve inop (grounded), and incorrect purge flow. I'm only experiencing a code for the latter.

[80stech]

I had a look in the Bentley manual and there is a huge amount of hoses valves and connections to do with the expansion tank that sits between the canister and the fuel tank, It does look like though the only air inlet is the filter on the leak detection valve so the diagram you have probably shows everything relevant to the code if that is the case. So you are back to a restriction in the line(maybe from the purge valve to the canister, that's a lot of line!) the canister itself(probably not),the filter(maybe) or maybe the leak detection valve is not spool is not fully going into the run position(against spring).