[tlewis]

2004 300M with a variety of PCM error codes related to the evaporative purge system and a variety of "fixes". Starting last year, would infrequently get PO440 (general evaporative system failure) and PO441 (evaporative purge system performance). Visually checked for missing/split/loose hoses. Changed gas cap to OEM gas cap. No real noticable improvement. Last fall, replaced the purge solenoid under the hood, and the codes went away for a long period of time. But in the last month, the codes started coming back every other day (I was killing them with my OBD2). So I got lazy and put it in the local Chrysler dealership, who said they did a smoke test, and replaced the evaporative canister which required dropping the gas tank, to the tune of $400. Then they said the PCM was bad and wanted $700. I drove it for a week, and now was only getting the 440 code, the PO441 code has not come back. So I replaced the PCM with another one myself, but the 440 code is still popping up every other day (after I erase it).

I am ready to do what I should have initially done, regarding checking wire continuity, replacing all rubber hoses etc. But I did want to put it out there to see if someone has any real world experience in ridding their 300M of the 440 code. Thanks.

 

[Dave Z]

In my case it was the hose and yes, they dropped the gas tank. Rather than just replacing the one bit of hose that was a problem, they asked if they could change out everything that needed a tank-drop and I said heck, yes. So my guess is either they screwed up their fix, or they left something there that's now failed. Either way, if it was within one year, they should be dealing with it, not you.

My parents had a similar issue with their import, as a side note.

 

[ImperialCrown]

I have found the NVLD (natural vacuum leak detector) as a common cause of this fault code before. The switch won't close, even under the forced test with the DRB III.

 

[Dave Z]

Oh, that could be it. Interesting. In my case it actually was rotted hose, but my car's four years older and was never garaged. The tech was still a bit surprised.

Gee, it's fun watching the big blue square reply box slide down when I post.

If I post again will my count increment?

 

[ImperialCrown]

Cracked (dry rotted) rubber hoses or swollen oil-soaked hoses near the engine are common evaporative leaks as well.
I had one case where the steel vapor line to the front of the car had rusted inside from moisture and rust had plugged the fuel vapor passageway. The steel line looked fine from the outside.

 

[AllanC]

I have found the NVLD (natural vacuum leak detector) as a common cause of this fault code before. The switch won't close, even under the forced test with the DRB III

Many years ago my 2003 Dodge Neon displayed code P0440. I removed the NVLD device and could not get the switch to close and pass current. I believe the switch closes but it only takes about 1 inch of vacuum or about 1/2 pound negative pressure to close the swtich. I believe corrrosion, carbon arcing, dust, etc can settle on the contact points and prevent the small current to flow through the switch. Replacing the NVLD device will probably fix the error code.

 

[FIREM]

I have found the "switch" in the NVLD causing a problem.
"YouTube" search NVLD, there are two great videos of how the system works and some easy testing to isolate the problem

 

[tlewis]

Well, the NVLD was replaced with the canister, and now it is not only throwing 440, but 456, which it did not do before. I took a few hours this weekend and examined all the hoses I could get to. The rubber elbows all looked good. I was surprised to see some clip type connectors to connect the plastic lines to the metal line. I took the time to put oil resistant RTV on all coonections I could reach, and perhaps that will resolve the issue. If not, I might just use some electronic trickery to keep the PCM happy.

 

[AllanC]

Well, the NVLD was replaced with the canister, and now it is not only throwing 440, but 456, which it did not do before. I took a few hours this weekend and examined all the hoses I could get to. The rubber elbows all looked good. I was surprised to see some clip type connectors to connect the plastic lines to the metal line. I took the time to put oil resistant RTV on all coonections I could reach, and perhaps that will resolve the issue. If not, I might just use some electronic trickery to keep the PCM happy.

Did you clear all existing diagnostic codes after replacing the NVLD and canister? Did you clean the 3 wire electrical connector terminals at the NVLD?

 

[tlewis]

Codes cleared twice after dealer worked on it. Hopefully the dealer cleaned the contacts.

 

[AllanC]

Codes cleared twice after dealer worked on it. Hopefully the dealer cleaned the contacts. /

From your description of services performed it appears the dealer service department was only guessing. I would check the electrical contacts at the NVLD device.

There is a vapor recirculation line that connects the top of the fuel tank to the fuel inlet pipe just slightly beyond the receptacle area that accepts a fuel nozzle. That needs to be checked and made sure it is securely mounted. The purge solenoid is somewhere close to the engine compartment. Since there is heat in that area check the evaporative system hoses at the purge solenoid and make sure they are tight, pliable and not petrified. Follow the line from the purge solenoid and look for any service ports. Some systems have vacuum ports for attachment of a vacuum pump. When not being tested a cap is in place and it must be secure and hold vacuum and not allow any leaks.

 

[FIREM]

FWIW the port cap is Green in color, 04 should be near the brake booster

 

[ImperialCrown]

It should also show the evap hose/line routing and component locations on the underhood vacuum diagram label. The NVLD (P0440) won't necessarily also set an evap leak (P0456) fault code. That may be a separate problem. Any leak (P0456) will prevent vacuum from building and therefore prevent the NVLD switch contacts from closing (P0440). It is the contacts closing that verifies a secure vapor recovery system.
You shouldn't have to use RTV to seal hoses. They should fit the nipple firmly however. Any breach in the system will set a leak fault and silicone-based RTV can decompose under HC exposure after a while. Certain greases and oils can also deteriorate rubber hoses. A soap/water solution is a good general rubber lube, if needed.

 

[tlewis]

In theory one should not have to use RTV, but a vehicle with 116,000 miles and ten years old could easily have leaks at the hose connections, and a high temperature, oil resistant RTV will outlast the vehicle while hopefully removing the problem. Sealing possible leaks was also more of a diagnostic tool than a "fix", since I have never seen the 456 code until I got it back from the dealer's shop. Unfortunately, I have not been able to drive the 300m enough to verify if the RTV fixed the leak, will report back next week.

 

[tlewis]

PROBLEM SOLVED!

No, it was not the RTV or system vacuum leaks. I have been working on my own cars for 40 years, from the simple slant six, a 318V8, two turbo 4s, two sets of 3.5 V6 and a 2.5 V6. In general I have always discounted wiring problems, simply because the wires are typically routed and protected to outlast the car. But my 440/441/456 problem was caused by a wiring problem that I consider to be a design flaw (note that this is not a fatal design flaw, or one worthy of a recall or lawyers, more of a design flaw like the crack that appears on the dash near the passenger air bag of every 300M I have ever seen).

Three wires run from the Power Control Module to the NVLD solenoid and vacuum switch. They enter the cabin near the parking brake, and are routed to the right (passenger) side of the car and then towards the back. The connector 308 is under the back seat cushion, and provides the connection to the NVLD solenoid and vacuum switch which is on top of the vapor canister, which is unfortunately on top of the gas tank with no access panel to get to it (I had to get that in). The back seat cushion in a 300M is 99% foam, but there are metal rods that help the seat hold its shape. So what happened? Someone, perhaps me, sat in the passenger side back seat, and the weight of the person pressed the cushion down, which pushed the metal rod down, which broke both the NVLD solenoid and vacuum switch wires. Since the interior metal of the cabin is painted, the wires did not short to ground and did not go high; therefore, none of the checks the PCM does to see if the wires are either grounded or high show up. And the PCM just thinks the NVLD vacuum switch is not closing when it is supposed to and throws the 440/441 and perhaps other codes. I have had this problem on and off for years in this car, and I do not recall anyone finding this type of problem on the internet (perhaps not many people sit in the back seats of their LX cars). I do remember that old VW bugs had the battery under the back seat cushion, and you could short the battery out and cause a fire, but that is another story.

So here is a new diagnostic for troubleshooting the "evaporative system problem" 400 codes: Take two minutes, pop the back seat cushion out, and check the wires going to the NVLD assembly on the passenger side. Fix as needed, then protect the wires with a secondary cushion to prevent the seat frame from crushing the wires again.