I have a 2014 E350 Coupe (207.388) with a check engine light with codes P0451 and P0452. I’ve checked the purge valve and canister shut off valve and both seem to be ok. I took it to a shop and they replaced a pressure sensor that is integrated with the fuel filter. My confusion is that I think this pressure sensor is for the fuel pressure not for the fuel tank pressure. I can find part numbers for both sensors but when I search and look at vendor descriptions the function of the sensor I get a mixed bag that either is both fuel pressure and fuel tank pressure.

can anyone help me determine which sensor measures the fuel tank pressure during the EVAP leak test?

thanks in advance.

 

The fuel pressure sensor is not part of the Evap so I would cross that out.
The tank pressure sensor should be on the top of the tank, or close to it like the charcoal can. So my guess the shop got it.
Basically the tank sensor is just looking for a bit if vacuum so it knows fumes aren't escaping. No vacuum and it assumes something is wrong. Is the computer smart enough to know what is wrong? Imo, no. For example; my Evap readiness monitor hasn't been ready since Feb 2022, so I guess I've had a leak since then, yet I don't have a code. If I do see a code, like on yours, I don't automatically assume whatever the code suggests. I'd look for more realistic reasons that set those code(s) off.
#1 suspect imo is leaks.

I'm in California and now we need all readiness monitors "ready" in order to pass smog. It used to be you were allowed one not ready because it's not easy to get all them ready, and the most common fail is the Evap because it leaks so easily. Even when it doesn't leak the readiness process is just plain stupid. So in my efforts to see what's wrong with mine, I was poking around today. What I found was a rubber elbow a few inches from the Evap solenoid that was basically half split open. Most of hoses connected to the solenoid are stiff plastic but there were a couple rubber elbows close to the valve. The rubber looked fine, but pulling on the hose a bit I noticed one elbow had a split on the inner radius as if someone cut it half way through with a razor. My guess is you have similar elbows on your car?. The gas cap can also leak with time, or I should say the caps O-ring. Any leak can cause an Evap fail, and at <10 years old I suspect rubber parts.
I was going to simply wrap the elbow with rubber tape, but decided to replace the elbow and small section of plastic line with rubber hose. It seems the rubber is glued on the plastic line, or maybe over time they just kinda became one because I had to shave the rubber off the hard line with a razor blade. Ever sharpen a pencil with a knife? If so, you know what my hard line looks like now. Knowing what I know now I would choose to wrap it.

If you find and fix the leak(s), then clear the codes and see if they stay gone.

 

Thanks for the response. There is a pressure sensor on the top of the tank but it is in the middle (right to left) and may not be accessible from the access points under the rear seat. I can see the sensor on similar tanks for sale on eBay. The mechanic removed the rear seat and replaced the fuel filter with an integral sensor. This sensor measures inside the filter. I hooked up my scan tool which displays live data for fuel vapor pressure and this signal seems to be bad (still) The reading is always the same. I would expect some variation

a couple of starts after pick up from the mechanic I have codes P0451 and P0452 showing again. What ever the mechanic did, didn’t change anything.

in the picture below the top sensor (brass colored) is the one the mechanic replaced. The bottom is the one I’ve seen in pictures mounted directly on top of the tank. Can some one identify which is the fuel tank pressure sensor?

 

I've never seen that brass one, but the plastic one looks familiar. The actual fuel pressure sensor is threaded in, because it's high pressure. There may be a fuel pressure sensor inside the tank, but I've never seen or heard o f it. so since those two just pop in I'd have to assume they are tank pressure.

When you see the pressure in you tool, is it ambient psi or something else? If ambient then I'd assume it's working and you either have a leak or the solenoid isn't opening.
A couple clues: If the Evap sys is working you should have a slight vacuum when you remove the gas cap. Maybe not after sitting for a while, but, for example, when you fuel up you may notice it. I can't recall my ever doing this, which I assume is that hose I mentioned? In fact, I can't recall any car I've owned maintaining that that vacuum, but I have seen it in many other cars.
Another is if the leak is of reasonable size your car may have trouble idling. Idle rpm will move around and may even stall thanks to the leak screwing up the expected air/fuel mix.

Many parts sites, like Pelican Parts or FCP Euro, will have an option to "Confirm Fitment". I assume you noticed that. I've learned to take these as "likely", not "confirmed". So if in question I will check several sites. If they all say the same thing I'll assume it's correct. If not, then I have research more, or get off my lazy *** and physically look at the part I have.

I just checked ebay and this popped up first thing: https://ebay.com/itm/366033082781
I'm not certain it's your exact tank since it says W212, but we share a lot of parts with those cars so odds are very good. Maybe you saw the same pix but this is great info. Sensor is in the middle towards the front, with the little orange shield around it. I assume the shield exists so the sensor doesn't get knocked out during tank install and the factory?
With any luck you could simply reach up from under the car and pop it out, and of course unplug it. I hate all the different plug locks they have on these cars, but thanks to this great junkyard pix we can see it's a very simple and easy lock.
The pix also helps understand it if you need to lower the tank a bit to access it. Seems to me like you could just lower the front of the tank a bit? However it goes down, please post how you did it so the next victim can learn from your experience

Also, odds are good there's a bunch of dirt and what not around that sensor, especially since that orange shield might as well have a sign that says "rock and dirt collection unit". You can see all the dirt on the ebay tank, but I have zero doubt they turned it upside down and dumped all the rocks and sand off it after they pulled it. If it were me I'd blast that area with shop air before removing the valve. Bend a tube or whatever you need to hit it from the top with the air. If not air I'd use water. I suppose it may not hurt to get dirt in the tank, but it makes me cringe, and if it did do something bad you have a much bigger problem. Plus dirt may get all over the seal on the new valve seal during install, maybe even prevent it from fully seating, but either way it could leak and you're back to square one. Just seems obvious to blast it clean first.

I bought a cheap borescope on Amazon and the image is pretty good. Much better than the $40K borescope I have a work. If I had to do this job I'd use it to scope out the area up there. If you don't have a borescope, you just might consider this a good excuse to get one. It seems my exact model is no longer made but this, same brand, looks basically the same:
Worth it kinda tool imo, plus has uses for many other things around the house and what not.

 

Thanks for the response. The brass sensor is held into the fuel filter plastic housing with a metal clip so it won't pop out. Both of the parts are confirmed to fit my car. my impression is they serve different purposes. I haven't found a good diagram of the fuel delivery system, but I think the fuel filter is on the fuel pump inlet. The brass sensor may be used to determine that the fuel filter is clogged or something like that. The sensor you found in the picture from eBay is the one I think is the fuel tank pressure sensor. As you can see from that picture, it is not near either of the two fuel tank access point (circular flanges). The WIS I have indicates the fuel tank must be dropped to remove/replace this sensor. Unfortunately, my mechanic replaced the brass one with the fuel filter and it didn't seem to fix anything (shortly after we got the car back, the check engine light came on with the P0451 and P0452 codes again.

I have a OBDII scan tool which will also provide live data. There is a data set for "fuel tank vapor pressure" which I assume is the pressure in the vapor space of the fuel tank. The number I get is -3999.75 Pa (-0.58 psi). I suppose this might be an OK number, however, the number never changes. tonight I had the engine running at idle and the value was -3999.75 Pa. I then took off the gas cap which should vent the tank to atmosphere. the value didn't change.

I will try to blast the sensor with air--that's a good idea. And I do have a borescope, so I will try to look at it. I need to go back and study the wiring diagram, but I think there is a connector that is accessible at one of the access ports under the rear seat. So I guess I'll try that too.

I originally took this to the mechanic because I'm not sure I can drop the rear axle to drop the fuel tank to get to the sensor with just a floor jack and a couple of jack stands in my garage. .Unfortunately I believe the changed the wrong part. I tried to tell the service manager, but he was believing his "master mechanic".

Thanks for listening.

 

Oic, I thought the brass one was one you're unsure of, because I sure was.
I never knew there was one inside but I suppose it sounds like something MB would do. Seems to me more like a part that causes more problems because they exist and fail than actually help, but I'm guessing. I've never seen an in tank fuel filter clogged. In fact, the only fuel filters I've ever seen clogged were clogged by their own element.
Maybe someone who knows for sure what that guy is can chime in?

Imo instructions on how to do things often make the job far more complicated than needed. Maybe to justify labor costs? Not saying it's wrong, just that I don't believe it until I get to the point I realize it's actually correct. I'd wager that most of the time the job is easier than claimed.

I'm not certain how the sensor works but I'd assume it measures psia and compares it to the ambient sensor to get tank vacuum. There will be some error there, but .5psi seems like a lot.
Have you looked at the ambient sensor? You should also have an air filter sensor and MAP sensor. When the eng is off they should all read the same, more or less. On mine I have two filter sensors plus a boost sensor, and this is how I check to see if any are bad, because if one reads different, then...
Speaking of, be sure you plug the sensor in and check it before you install it. Would hate to get it all back together and ooops.

I wonder if you could use the sensor wire it to your advantage. Like a ship inna bottle, get the sensor out and in with a stick or something, with the help of the wire. I've done similar in the past in tight spaces.
Another thought, which you probably won't like but is fair game for me, is drilling a hole in the car to access something. Of course I'd seal it after, considering what's under it, but totally an option if it were me.

Axle? I'll assume you meant driveshaft. The driveshaft isn't a big deal really, annoying, but not a biggie. I would consider the driveshaft to potentially be a great place to support the tank once I unbolt it, assuming the tank lowers enough to get my hand in there. I'll take your word on how much space is in and around there because I can't remember the slightest about that area because it's been years since I've been under that part of my car, assuming I looked at that area at all.

I generally don't believe mechanics either. You have to get one that is both intelligent and experienced with whatever it is. Last time I was at the stealership they said there was nothing wrong with the car. I *****ed so they had the shop foreman come out to "explain" it to me. We gently locked horns for a bit, each testing the others knowledge. It seems we each approved of the other and he agreed to let me take him on a test drive. It only took a few seconds for him to agree something was wrong. Unfortunately he too could not figure it out, which he admitted with zero bs.
Come to think of it, that's basically how all my dealer visits have gone, of which I think I've had I five in my entire life. No dealer has ever fixed a single issue I have ever had because they don't know how. Granted I tend to have weird issues, but it's still sayin something. I haven't been to any mech other than a dealer so I can't say about indie shops, but I did work at one for a short time. No I would not take a car to them! I'm certain some shops are good, or at least honest, but honest and know what they're doing I'm not sure there there are all that many?
The dealer just goes by what Xentry tells them. If Xentry can't spell it out for them, apparently they don't know what to do. An indie might have Xentry, but I'd imagine Autel or something is more likely. Doesn't matter because a code is a code and different software still gives the same code. Those codes point directly at that tank sensor, or a leak tricking it into that code. So if he changed that brass one I'd be asking wtf and to please explain the thought process. Perhaps you can get them to admit the mistake and replace the correct sensor for free. Maybe, if you're lucky, because I would not look fwd to replacing it, or paying to have it replaced.

You know, I'm not much for jack stands. I know they work well but F me they make me nervous. Not so much that they'd break, but I fear the car tipping over, the jack slipping off, or people not chalking the wheels. Why do I worry, because I've seen it. Heck, I even dropped my car not long ago, I wasn't going to get under it, and I knew it might fall, but I wasn't worried. What I was is too lazy and just wanted to see something real quick. I jacked it up by the rear diff, which people often do, but you have be wary doing that.
Since there's no way I'd get under this monster to do an oil change using jack stands, I made wood ramps for it. I also have zero trust in the plastic ramps they sell, which you can verify with online pix of ones that broke, but the store bought ramps are too steep, at least for the front end. I used 2 x 12" boards, stacked, each layer shorter than the one below, with a 45 cut on the leading edge to better drive up. I can post a pix if you want, but the angle is such that the front of the car just barely clears it as I drive up. Just an fyi since you seem like the type that just might consider making them. I love them and don't think I'll use stands ever again.

 

I’ll try to climb underneath this weekend to see if I can spot the tank sensor and assess how to get to it. But my WIS does say to drop the rear axle suspension to lower the fuel tank. I guess the tan sits above the rear axle

The tank pressure sensor (aka EVAP System Vapor Pressure) measures absolute pressure in pascals (see below). The reading (-3999.75) has been that value for some time. The scanner also picks up atmospheric pressure which is about 100,000 pa. This sensor seems to be working (not shown).



at some point I’ll remove the rear seat and poke around the access holes. The pictures from eBay confirm there is a connector for the tank pressure sensor near one of the access point. I’m hoping I can use this to confirm the sensor is bad.

thanks for the help.

 

I was able to spend some time with the car this weekend. I also got much better at ant searching and printing from my e-manual.

i could not see the sensor from underneath the car. I couldn’t even see the fuel tank with all the plastic covers and heat shield, etc. I removed the back seat and opened the right side access cover but I couldn’t snake my borescope much beyond the access port.

I did find the electrical connector to the sensor underneath the access port. I was able to confirm I have +5V at that connector but I only have a mV signal coming back out. I also had my scan tool hooked up. Most of the time the pressure reading was the -3999,75 Pa which I think is the lower limit of the measurement range. At one point the sensor started reading again and I even went for a drive. While driving the sensor was reading -100 to -200Pa. But then the sensor went dead again.

apparently my charcoal canister shut off valve has also gone bad. The shop reported this but I told them not to fix it—that I would take care of it. I previously had the canister out and bench checked the valve and it seemed fine. But I took it out again the weekend and the valve wasn’t working. If I have questions about this failure I will start a new thread.

my next step is to go back to the mechanic and see if I can convince them to replace the fuel tank pressure sensor.

thanks.

 

Hi, thank you for posting detailed information. I have a similar issue with a w204 that started with 441 451, and more codes kept piling on. I replaced the sensor under the seat and got nothing. I then started getting a 2422 as well which ended up being my canister vent valve. Unfortunately replacing the valve only cleared 2422 and im back to 441 and 451. I understand that The pressure sensor on top of the fuel tank can cause incorrect readings and throw codes.

So I am wondering if you have replaced it yet, and if you did are your codes gone?

It would be great to be able to test the sensor with a meter from the harness end which is accessable under the back seat on the passenger side, so I am trying to find anything I can about that. If anyone knows please chime in.

Cheers!

 

I too am curious just to see what he has to go through to replace that thing.

Meanwhile, here's some thoughts that may or may not help:
You won't know what the raw data means unless you can find info on that exact sensor. I don't know anything about the tank sensor, but with other pressure sensors I've learned that they vary. So I'd imagine you'd need to find data for that exact sensor, or the MB data for your exact car, and by exact I mean VIN #, because they might have changed sensors mid year, or ***** nilly.
For the sensor you could find one online, like Pelican Parts etc, and see if you either find anything online or email the mfg for a data sheet.
If it reads 0V or 5V then you can probably consider it hozed, but something in between you'd need to know exactly what it means. I'm sure you know this, but fyi.
Other than that I'd use a tool to read the sensor, like the OP did, which seems easier than the above. With a car that old I think you need a good tool because this won't be the last issue you have. Not sure what all tools can see/read that sensor, but I'd imagine most cannot, so be wary. I use Xentry, and while I never looked, I assume it can read it.

Since you have multiple codes for the same basic thing, and they keep returning, I'd assume it's a real problem. I'd also assume that tank sensor would need to be out of range to trigger 451, and if so I can see it triggering the other codes. The only thing I can think of that might cause this issue, and not be the pressure sensor, is if the Evap valve is stuck closed and the tank builds pressure, as they tend to do, and exceed the range of the sensor. So I'd check to see if the tank has pressure, like after a drive in the day, or does it have a slight vacuum. You can check for vac with it idling and see if the gas cap wants to stick to the tank. If you can't feel that then stretch latex over the hole. I'd assume from a latex glove but there is one other item that comes to mind...
If the Evap is working it should suck right away. If the sensor is reading low then I suppose it won't, but that too is a clue.