Fuel pressure thread
 

So, I am starting this new fuel pressure thread, adding some info that I have collected up to this point and formulating a few theories too.

In the next couple days I'll come back and clean this up.
Then I will follow up, as I'm working on solving my problem.
Attachment 418023Attachment 418024

Let me start with this, I've got a code.
I've first scanned with an OBD2 and found the P228F code.
then found the P228F thread.

https://mbworld.org/forums/c63-amg-w...ode-p228f.html

Strangely, in this thread, all shops continually started "from scratch", everyone started by replacing the fuel pressure sensor. I didn't understand why.

Then I scanned with Xentry and got the 3699 code. But Xentry is so misleading, forget about pressure out of limits. IT TELLS YOU its the sensor. How can you ignore that..
So At this time I had reset the code so just ordered the sensor. Surely my situation was going to be different. I even replaced the sensor without resetting the code and 3 startups later, the check engine light went away. The engine seemed to run smoother too..
Until it came back., because it's not the sensor.

Now, I managed to see the Master inquiry and the dealer that serviced my car when new, put in fuel additive A0009892545 16. They did this every time they saw the car.
Now, the car has 3 pumps: two on the right side and a non-powered jet pump that works on Venturi on the left side of the tank.
Now although the factory literature doesn't show this, the jet pump contains both a fuel pressure regulator and a fuel filter, as Thassos has shown in his video.
But the fuel pressure regulator doesn't have a vacuum sensing capability so it's not progressive/linear, it's just preset, a pressure clamp.

Also, this car like many others keeps adding fuel in the long term fuel trim (although not in Xentry) and it never fully lost the rotten egg smell in the exhaust.

So, if these separate facts were to be put together, I wonder if the fuel filter is clogged and/or the fuel pressure clamp is stuck, because getting way too much mildly corrosive fuel detergent from that dealer.
So my next move will be to replace the jet pump.
In case this doesn't work, I'll try replacing the two right fuel pumps too.

As a relevant story, in my Alfa Romeo 164 days, I was a Techron guy. I added a can every month sometimes. Well, I got a hole in my fuel tank 3 years later. This stuff is corrosive, that's how it cleans.

I was doing a bit of research on the forum, trying to see if the people that complained about the rotten egg smell were also complaining about these codes. I did not correlate the two groups yet.

If your car was sold in PA, you have a mild rotten egg smell every now and then in the exhaust, you may want to follow this, in case you start getting fuel codes.

Finally, the jet pump arrived.
There are two models. The early cars part number begins with 204, the later I think is 218.
Attachment 418070

Now to read the WIS again and to try to understand what these hoses are doing, then on to replacing.

The WIS indicates replacement of the fuel filter only. A simpler and easier procedure, only the left cap has to come off. The part would surely have been cheaper. I wonder if the pressure regulator could be part of the filter assembly. I should have read this before ordering.

What I think I see here is that the two right side pumps are installed in parallel, the two output hoses converge and en-route to the fuel filter, the smaller diameter hose that diverges is probably the venturi effect suction hose, that takes fuel from the left side of this saddle tank.
I see the hoses zip tied together, right now it is nor clear if I cut the zip ties and it's also not clear if all 3 hoses travel from the left to the right through the same channel.

there is a center hose, which could be a vapor pressure equalization hose.

On to the car now..

battery off, fuel level on "fuel low"...

Ok, I'm back and finished the work.
There are no particular symptoms.
I opened the left side and right side. I disconnected the fuel pumps pulled the left jet pump with a rope, pulled the new hoses in with the rope.
The car had not been sitting for a long time and the fuel was really low. And it was cool, was really not hot.
There was a lot of pressure relieved when I disconnected the outgoing hose from the jet pump. Also much fuel came out. like 2-3 cups.
You have to feed the new hoses sticking a hand in the left side and one pulling them with a hand on the right side and wiggle them, so that they can get seated.
The minute they are sufficiently stretched, the left side cover can be closed with no effort.
Attachment 418083Attachment 418082
Once I was done and started the car, it obviously ran like crap. It surprised me how long it took for the air to be bled out of I guess the fuel rail.
And of course I racked up a bunch of codes. Low fuel pressure, misfire and so on.
But then it started to run ok and hold a nice idle.
Then i reset the codes and took a ride to the gas station.
This is of course subjective, but it does feel like a fresh fuel filter and fuel pressure regulator. Kinda like new injectors, only more intense, the car is meaner.

I will need no more than 2 days to see if the code comes back. Because it becomes stored pretty quick, like in one day and then the second day it lights up the check engine.
But checking with Xentry for stored codes, I should be able to see it stored as early as tomorrow, if it's still there.

If it is, no big deal, I'll replace the pumps too, now it will be easy.

Now I'm off to cut up the old fuel filter. It was much heavier than the new one, so I'm excited to see what kind of crud is in it...:)

Attachment 418081Attachment 418079

Attachment 418084

The weight difference was a cup of fuel still enclosed in the filter can.
I would say the condition of the filter is what one would expect at 60K miles. Which is also when a fuel filter can be replaced. What a filter that is beginning to be loaded does is to muffle acceleration inputs. And the new filter changes this.
I'm going to take a look at idle fuel pressure.
I looked at it before and it was hovering up and down around the prescribed value.
Looking at it may tell me if the fuel pressure regulator may have had an issue.

I could not find anything about this filter, googling the part number comes back empty. Not that important because one can't remove it without cutting the filter can open.

Now while WIS talks about replacing the filter only, the EPC doesn't seem to have a filter only part number.
They either sell the whole thing, like I bought it, or they can sell separately the fuel level sender, or the "coupling housing", which the parts outlets call plug and sell for $3. Maybe this is the housing of the electrical connector.

So, as I wrote, I don't really expect 100% for this to make the code go away, may have to replace the pumps too.

One forum member had to go beyond replacing the fuel pumps, to make the code go away. He had to replace the fuel tank, but this was done under warranty.
The fuel tank is a bit expensive, $689 and I expect there will be quite a shipping price too.
Now, from my experience with buying factory new large and expensive sub-assemblies, they come many times with surprising items already included.
It could be that they sell the tank with pumps already on board. From the picture I see that it includes all filler neck hardware.
The problem is that the poster that changed the tank said that the dealer that was doing the work under warranty, worked with AMG and they tried a somewhat different tank, that they said was known to fix this in several instances. I don't know the part number of this different tank. And the original part number does not show substituted.
Of course I'd have to check with the dealer parts guy with the latest parts software revision, to see maybe the new part number shows there.
Since then he stopped posting about this, this has fixed it for him.
I wonder if Mercedes provides some kind of hotline for situations like these.

The fuel pressure in idle is about the same. I didn't check how unchanged it stays under heavy acceleration.
The OBD2 trims are a bit different, meaning they're still rather large, the long term, but the short term do negative and stabilize the long term.
Previously, the short term stayed close to zero but always positive and the long term seemed to always creep up very slowly.

So, our cars have returnless fuel system. I am reading that in this system, the fuel pressure stays constant. This is consistent with the regulator being non vacuum adjusted. The flow is regulated by the ECU accelerating the pumps instantly as needed.

The following blue section is only on a By The Way base and you can skip it if it seems complicated.
In any fuel system you can change the AFR (Air to Fuel Ratio) in two ways:
1 injector duty cycle
2 fuel pressure.
You may ask how to change the fuel pressure when there's clearly a fuel pressure regulator there, to prevent the pressure from going over the prescribed value.
The answer is by intentionally overwhelming the regulator. For instance in my Subaru, that has a fuel return, at cold start up when you need a rich "choke" AFR, the ECU shifts the pump into high speed and provides such a flow of fuel, that the regulator cannot dump the excess fast enough through the return line into the tank and results in an increase in the fuel line pressure over the regulator value and a rich cold start up AFR. Once the ECU sees that the start up condition is met, shifts the pump back down and the regulator takes over.
With returnless its easier to overwhelm the regulator, just run the pumps more. So pressure control is also available to returnless systems, in some cases, but most of the time, pressure should be considered constant.


This is heavily dependent of the pressure sensor, because the reaction time of the ECU is minute. A tired pressure sensor may work,. but begin the respond slightly slower or a little slightly slower. So a new sensor will always be felt.
Now I'm trying to figure whether in that fuel filter can, the regulator is upstream or downstream of the filter. It is at the bottom of the can. So maybe it is in the constant pressure downstream of the regulator section. Filters are usually in the low pressure high volume "fuel circulation area", upstream of the regulator.
Thisone would see higher pressure then. If I'm right, a worn high pressure operating filter probably looks different. The particles are jammed deeper into the filtering media. It seems cleaner but it isn't.

At the same time however, for the same number of miles total travelled, because of no fuel circulation, a lot less galklons of fuel travel through the filter in a returnless system, because no fuel is returned to the tank then circulated back through the filter. The only fuel going through the filter is the one consumed by the engine. SO from this standpoint yes the filter would last longer. But with probably double the pressure that is reduced.

Now I'm reading that the whole point of a returnless fuel system is for the fuel in the tank to remain cool. This way controlling fuel vapors in the tank better.
But what does this do at the fuel rail...?
In Mechanical Engineering, you take the thermodynamics course. In it, you study Phase transitions: vapors to liquid and so on.
Now, there are the obvious things, that you don't need to be an engineer to know, but also the way less than obvious things..
Vaporization depends on pressure. If yo increase the pressure, a liquid won't vaporize under increased temperature at the same temperature. You can push up the temperature of vaporization, depending of course .. on which liquid.
So fuel can be prevented from vaporizing inside the fuel rails and creating vapor lock in the return less fueling, by increasing the pressure. So returnless runs at higher pressure than normal, to prevent vapor lock. This is why so much focus on fuel pressure.

The only thing is that if they did all this, why not put the filter and regulator on the maintenance schedule and why not make it a bit easier to access, although with the proper tool for the lock ring it's not difficult.
And yeah, the tank really had cool fuel inside almost weird.

With the Subaru, I would not work on the fuel pump in front of the house, because the whole neighborhood would stink like gas and the neighbors would freak. This car.. nah, very little smell cause no vapors.

I am reading that the injector flow rates on a returnless system need to be particularly close. Don't know why yet.
And that there can be benefits from replacing the fuel rail, because the far end dead end of the fuel line becomes a gunk and debris accumulator to the point to where it can't even be cleaned.
I didn't even attempt to clean inside the fuel line when I had it out.
And maybe this is why the expensive AMG branded injectors with tight flow tolerances are needed, but let me check why.
Also the thing with the injectors stuck open is not necessarily an injector quality problem, but probably a returnless fuel system problem. SOmething tells me that the people with fuel hydrolock, the leaking injector was by the dead-end where the dirt accumulates inside the line.
We should not worry about injectors but replace the fuel rails regularly.


And here's an idea: Because the only fuel that leaves the rail is the fuel swallowed by the engine, it's easy for the ECU to hyper-accurately know how much fuel has been gulped, therefore ... the gas mileage.

So the gas mileage indicator on the dashboard should be... very accurate. But it is not.

Ultill you use the Xentry to reset the fuel adaptations, then it becomes accurate to the 1/2 gal and stays accurate. A second reset got it within 0.3 and by that I mean i compare the displayed average gas mileage that I show in the "from last reset" screen, with the reading in gallons from the fuel dispenser pump at the gas station, divided by the number of miles since the last odometer reset, which I do every time i fill the tank. For all I know, maybe the car's number is right and my average based on gas station gallons may be wrong by 0.3 gals. Maybe the Gas station scores 0.3 gals for every tank full.

Anyway, so why this inaccuracy solved by Xentry? Because there has not been a Xentry reset, but also because the fuel system performance has slowly degraded in 50 K miles. The ECU's knowledge of how much fuel has been gulped is now skewed until you do a reset. So how skewed could be a measure of how much maintenance the fueling system needs.

Ford Motor Racing fueling white paper

Here's an interesting article. Still no mention about why tight fuel injectors tolerances are needed for a returnless system. So I can't confirm this.
Interestingly, this article does mention using fuel pressure to overwhelm the regulator at high loads, to achieve the needed AFR, but they don't want to get into the details of why the injectors were selected in that way, why not selecting larger flow rated injectors.

Ok, I'm done with this for now, I'm going to drive the car now.

No stored code yet.

Interesting thread. Probably unappreciated and went over the heads of many, but was interesting to read and I'm sure it will help somebody in the future if they come across it.

Fingers crossed the code stays away now.

Thanks.
Still no stored code yet.
I'm starting to think about what does it mean, if my code doesn't come back.
Why for my car with the same symptoms as the others who changed the pumps, changing the filter and regulators fixes this?
Am I doing a partial fix, of were they doing a partial fix?

Here's what I mean:
If my repair, (by replacing the filter-regulator) is incomplete and I get the code back, then depending on how long it takes to get it back, it can mean that the people that only replaced the pumps would benefit from replacing the filter. For instance, if the code comes back in 1 month, then my incomplete repair can actually be the cause for which the pumps need to be replaced in the first place. Then if you invested in new pumps, you should protect them with a new filter-regulator.

If my repair is complete and the code doesn't come back in a month, then possibly the people with pump replacement partially fixed the problem. Pumps suffer normal wear and their flow in time degrades. A set of fresh pumps jolts the fuel pressure regulator, which after that falls back on normal fuel pressure. However, in time it gets clogged again and a year later the code comes back.

This brought something to my mind: The regulator sees unfiltered fuel, I think.

The hickups should come with some codes.

So after I replaced the jet pump, I had a very rough restart and there were 4 codes, which were low fuel pressure and misfires on 3 cylinders. After 30 seconds, the air was finally bled out in my case, the engine started to finally run very very smooth, I deleted the codes and restarted 4 times to make sure they're not reoccurring and everything was fine.

So using this as an example, I think hickups should generate ... some.. code. Now don't forget that the code that started this whole thing reads different in an OBD2 reader than in the Mercedes factory scanner Xentry. So maybe you're not scanning with Xentry and if you do you WILL see some code, which will give you additional clues.
There is something called "freeze frame" which tells you all the conditions at the time of the code. My code only occurred in idle or low load and RPM.
There is a strange coincidence, which is that one week before this code starting, I had run my car unusually low because I didn't trust the gas stations that were available in my path.
Normally I refuel couple miles after the low fuel message but now i ran maybe 20 Miles after that.

Now once I was curios to see what happens, 2 years ago and I really ran it low with cleaner in the tank. So after the low fuel message, like 30 miles later, there is a second message "go to gas station" or something like that which made me laugh.

It's been a bit more than 2 weeks and yesterday I ran the fuel really low, down to the go to gas station graphic, to see if the code comes back or at least becomes stored again.
It did not.
I think that the fuel pressure regulator normally doesn't work, stays closed because the ECU uses the pumps with feedback from the rail sensor to establish the flow and pressure.
I think the regulator works when the rail is loaded with fuel for a WOT run and then you take your foot of the gas.
Then it dumps the excess gas.
But this is when we notice the bad smell from the converters in idle after a hot run.
So this is when the code pops up too in idle, fuel pressure too high.
The regulator becomes partially stuck, the pressure is too high and the car runs rich making the converters stink.
But why did replacing the pumps work then?
Maybe by replacing the pumps the dealers replaced the jet pump too. In the other thread some people had the work done by the dealer.

Interesting read and findings, especially regarding the fuel rail and injector flow tolerances!

Thank you for sharing, Vladds :y

Please let us know if you find anything regarding discrepancies in tolerance of AMG stamped injectors vs the Bosch generic injectors.

very interesting thread now you should see what I did with my fuel system to run bigger lines and much bigger injectors :crazy:

So I think I took care of the rotten egg smell in idle after hot runs, with this.

And thanks guys.

Did you have to change the 2 pumps on the right side vladds?

No, it's been quite a long time, the code didn't come back.
Also, I ran the car so low on gas, that the normal low fuel message went away and was replaced by the Pull into gas station message.
I wanted to see if the code comes back, it didn't.

This thread is very interesting. I had the P228F code for about a year now with no noticeable affect on performance or gas mileage. I replaced the FPS on the fuel rail, and of course, it did nothing. I had seen someone recommend putting a bottle of techron in the tank. Since I only run Chevron or Texaco gas I didn't even try it. Finally 2 more people who had experience with the P228F code recently suggested I try it so I did. Boom, the code went away!

That was about a week ago.

I fire up the beast today and get a P2540 code, Low pressure fuel system sensor circuit range/performance.

Ah the irony of it all...

Hey, I know its been a long time since you opened this thread, but did you have the exact part # for the jet pump by any chance? I tried ordering it online with the local dealer but they said this part doesn't match with my VIN#.. The part# I ordered was 204-470-46-94-80

Thanks in advance

Hey man I see that you bought the 2008-2009 model jet or “left” fuel pump. This won’t fit your car because u have a 2010 and they made a diff part. I just bought the wrong part last night and reordered what u posted. U need to go to https://www.autohausaz.com/pn/MB-2124703294 or switch the part numbers and make sure it’s not 204 or it will be for the earlier models.