New Old thread- gas fumes in cabin 16V

[MDE3]

I have been chasing a gas fumes problem in the cabin for some time...my techs and I have identified several problems but still not soved the issue.

To Date:

Replaced line from tank to pump
Resealed fuel filter and lines
Replaced purge valve
Replaced Carbon canister
Replaced fuel cap (3 different ones..all new)
Removed tank and pressure tested/smoke tested to 16 psi (twice)

What is very annoying is that each of the parts replaced above except the fuel caps, were in fact defective ..loose carbon falling out of the canister, blocked vaccuum lines (with loose carbon), plugged/jammed eha valve, leaking fuel hose from tank to pump, questionable fittings at the fuel filter..replaced. There were leaks evident in some cases. all this and still more gas smell. None of this solved the problem.

I have not changed the eha since the smell started, but relaced it in August this year for different reasons. Gas odor started a month or so later. However there is no odor of gas in the engine housing, nor any visible leakage at the FD, EHA, or Injector coonection...I have had these before with no gas odor in the cabin also. It seems to be coming from the rear......

Now when leaving the car ovenight there is a strong smell of fuel in the cabin in the morning. When driving with the windows closed and the fresh air vent open..no odor. When driving with fresh air vent closed and heater/ac circuit blowing..no odor when driving with any of the above conditions open or closed, but if the drivers window is cracked an inch or more...strong fuel odor coming in...reduced cabin pressure from the window cracked creating lower pressure in the cabin..sucking fumes in. Assume through the rear vent which is right above the fuel door.

Am stumped.

[bbarcher]

My 86 190E had fuel odor, turned out that the tank was slightly collapsing due to a bad charcoal canister not allowing it to "breathe" in air as fuel was pumped out. The constant flexing of the tank eventually ruptured it on the front of tank, where it could not be seen. The leak was on the passenger side, about midway up the tank.

To confirm my leak without removing the tank, I stuck a paper towel under the tank, and moved it around, when it came out wet, I knew. There are some small depressions that collect little puddles of fuel directly under the tank where you can't see.

I replaced my tank twice, because I missed the canister problem the first time and it ruined the first tank I installed. No problems since I replaced the canister. Don't tell anyone I used a BMW canister because it had a nice open bottom with a filter instead of having a small tube to breathe in vent air. Plus all the used MB canisters were bad too, all the carbon pellets were pouring out and that's what clogged the lines.

I know you did the smoke test, and that is fine, but if the leak point is submerged under the fuel level, no smoke will escape. At this point it is worth it to pull the tank for inspection. Order a new screen filter for the bottom of the tank where the hose attaches, since you will have easy access to replace it once the tank is out. And the replacement tank may need one. You may be able to repair the tank if you are safe about it. But I would recommend replacing it if it is distorted. The seal around the fuel sender may not seal well because the tank gets distorted there as well.

[MDE3]

Have had the tank out twice to inspect..no collapsing was visible..tank holds pressure fine as well...The smoke test was also done as well as a pressure test...I am told to to 16 psi....This was done twice with the tank removed from the car. I have a spare tank in another car sitting at the same shop, and were the tank showing any signs of leaking, they were more than prepared to change it...might still come to that, but hard to believe no signs of fuel leaking under pressure or visible smoke either if the tank is cracked.,

There was at one point fuel under the tank which was actually from fuel running down the outside of the fill pipe (upper rubber seal was bad) through the trunk, and down to the bottom of the tank. This has now dried up. Forgot to mention that it also had a fuel leak from loose fittings at the fuel filter...all of which had been fixed during previous episodes.

The fittings were all checked and tightened on the bottom of the tank as well. Yesterday afternoon, just before I wrote this, I took the car on a run from being up on a lift...got the fuel odor nice and strong and went straight back up on the lift...no fresh fuel under the tank or on the tank, or under the rubber bale that supports the bottom of the tank..in fact no signs of fresh fuel leaking anywhere...but lots of cabin odor...but only with the window down and drawing air into the car from the rear vent.

At one point there (a few weeks, and multiple repairs ago) was fuel under the car...but very much forward of the tank and pumps...however yesterdays inspection showed no fuel leaking there either. Will try again to see if the eha is leaking..but was dry to the touch after a hard run and strong cabin odor when I looked again yesterday.

[MDE3]

With all these other problems now eliminated, and still having this odor in the cabin, we once more put the car on a lift, and smoke tested the entire tank and purge system once more. We found some vacuum anomalies..but no leaks. The fume issue is constant but also intermittent...meaning at times during a trip or while testing we will get no fumes..even with the window cracked. However sooner or later these fumes appear again...however always and only with the window cracked..or either after the car has been sitting for a while.

It seems like we may have a sticking vent valve..the one located in front of and under the fuel tank (not to be confused with the purge valve located between the carbon cannister and the intake vaccuum line as I previously did).

This valve works two ways...to allow built up fumes pressure in the tank to vent out to the carbon cannister, or to allow fumes and fresh air back into the tank when the pressure in the tank becomes negative briefly under load. Our feeling is that the valve is sticking intermittently..but frequently, allowing the tank pressure to rise to where it now vents out of the gas cap pressure relief vent, and then is drawn up into the cabin vent right above the fuel fill door..when the car is moving with the window cracked.

We have ordered a new vent valve, but while waiting, we cleaned the old one and put it back. Test driving the car today and late last night, we are still getting the same thing intermittently, but with very much reduced intensity...again only when driving with the window cracked. Hopefully we are finally on the right track.

Am hoping three other issues which have come up recently may be solved by this...stumbling when cold with what feels like fuel starvation (too much vaccuum in tank), premature fuel cut off (6300 rpms)..again too much vaccuum in tank, and idle surge when backing off the accelerator..idle will stay briefly at 2200 rpms, and then slowly drop to normal (900 rpms)..maybe due to excessive pressure in the tank..which bleeds down as fuel is comsumed, but continues to push fuel into the system due to excessive tank pressure for a couple of seconds...

Well time will tell...I actually had the fuel shut off climb to 6600 last night after cleaning the vent valve....twice, but then it shut off at 6300 again during another test run...we shall see.

Hubris is a terrible thing...

[junqueyardjim]

All those words in a couple of posts, but no indication of year, make or model.

[lkchris]

Verify EHA isn't leaking.

[MDE3]

Originally Posted by junqueyardjim

All those words in a couple of posts, but no indication of year, make or model.

And the thread title is?

Since they only imported the "legendary 16V" for two years, I didn't see the need to be much more specific..Of course I assumed Mercedes "aware" readers of these posts..my bad...

If you feel inspired to contribute...please do.

[latief]

just curios, what is the nature of the smell? raw fuel? or somewhat processed/burnt fuel/ rich mixture?

I know this might sound naive, but maybe you are getting a different smell? i.e oil leaking at the back of the head .....you seem very experienced, so I am sure you can differentiate, but just throwing this out there........

[MDE3]

Like sticking your nose in a gas tank...lol very strong odor of raw fuel..definitely not exhaust. Think we have found the cause now....have not yet recieved the new valve, but have an elbow in place..not ideal..just a temp fix, but the odor is gone.

[MDE3]

replaced the temporary elbow with new vent valve...gas fumes strong as ever...

I am stumped...as there was almost no fumes at all with the elbow in place instead of the vent valve, I can only assume that the pressure build up in the tank is causing a leak somewhere in the tank itself. We have removed the tank twice and pressure tested it, with a special cap to seal the tank...plus the tank always seems to be under pressure when we remove the fuel cap after running..so it would seem that the tank integrity must be ok.

Summary of symptoms


gas fumes in cabin when getting in the car first thing in the am..car cold...gas odor is slight but present.

strong gas fumes come in the car when driving but only with window cracked...no fumes when using heater with fan on, or fresh air vent blowing

no visible gas leaks under car when sitting (now)

at one point there was an apparent liquid fuel leak because I did have a visible fuel leak under the car. However since tightening all fuel lines, there no longer appears to be any live fuel escaping when sitting..have had the car on a lift multiple times...fuel was running down the fill pipe, then down the outside of the tank and collecting under the rubber at the bottom of the tank...this no longer appears to be the case, as it has always been dry the last few times we put it on a lift. This fuel appeared to be coming out of the fill port, and leaking by the rubber grommet that seals the fill pipe from the frame...however when we pull the car in after test running and having fuel odor, and inspect the fill port, there is no visible sign of fuel escaping..no visible wetness in the fuel fill chamber either

no gas odor under hood with hood open

Have done the following:

changed gas caps at least 4 times with new caps

smoke tested entire fuel system with everything in place..no obvious leaks

removed fuel tank twice and pressure and smoke tested it..no visible leaks

replaced carbon cannister

replaced vaccuum lines to carbon cannister

pressure and smoke tested lines from tank to carbon cannister

replaced purge/regeneration valve

checked eha for leaks multiple times

replaced vent valve

tightened fittings at strainer and fuel pump (or was told that was done)

Am stumped...thought we had it licked when we replaced the vent valve with an elbow and fuel odor stopped...figured bad or sticking vent valve...maybe still true....

We found loose carbon had plugged the line from the cannister to the purge/regeneration valve (the one inside the fender), as well as the valve itself..Although we tested it for flow and it seemed ok, we maybe have carbon in the vent line(goes from the tank to the carbon cannister),Maybe we have residual carbon in this line which got sucked back into the new valve causing it to stick too? Will open up the old valve to see if any carbon in there, then blow out the vent line to see if any carbon comes out. If I find any, will replace the two way vent line, as well as the new valve.

[MDE3]

....well as so often happens the soution was "right before our eyes"..at least so it would seem.

Apparently with all the smoke and pressure testing done, this leak escaped us. It was leaking at the union of the copper elbow and the tank outlet spiggot..a flared coupling...We replaced this hose in june with a used line from my donor car when finishing the engine/trans/diff swap in this car, because the previous hose was rotted out from the car sitting, and leaked profusely. I guess it's tough to get a flared fitting from one car..already used once, to fit correctly on a different male union.

We pressure tested the tank itself while it was removed (of course the hose was disconnected during the removal of the tank), and smoke tested the entire system while intalled, and saw no leak. Although there was evidence of previous leakage there at one point..no "live fuel" was ever seen until Sat. morning past (no real pressure with the smoke test, and the leak was below any air in the tank..so no visible leakage..and it was not constant, except when driving.

Now I am just hoping a new hose alone will solve the problem...the outlet pipe and male union has been "bonded" to this tank from some previous repair...we may have to swap tanks in the final fix, if this new hose and flair does not do the trick.

[Randy]

I filled up the '86 420 SEL today and I swear I could smell gas; I had the sunroof open so I closed it and they went away. Not exactly strong enough to be sure, but I really am going to check mine out after reading your posts.
I had Grand Marquis once, whcih woudl smell of gas when it was full. I finally found a tiny leak in the tank, so small that the gas would evaporate and never make a wet spot on the garage floor. Replaced the tank and problem solved.