NaroEscape

Since we've owned this car (04 GT3) on long trips we get a slight fuel smell in the cabin. Just a wiff, but enough to notice. Then early this year (coming back from Daytona 24) we started having fuel fill issues. Could only put in a couple gallons until it would stop, the only 1/2 gallon at a time. That went on for a while, so I honestly drove the car a lot less. We went through they system and found the charcoal canister was full of fuel!! No clue how that happened. So we replaced it and it seemed to fix the fill problem, but still smelled fuel on longer drives. Then a few weeks ago, the fill problem came back with a vengeance- can only do 1/10 gallon at a time before it actually sloshes out the filler neck!!

So I took it all apart today. Canister has fuel in it again. There's a vent valve right on the filler neck I'm going to replace, and two other valves that come off the carbon filter to a little holder above and forward of the wheel.

I'm concerned that the vent valve at the top of the tank is bad...but you can't get to it without dropping the tank from what I see...

Anyone have similar issues and know a fix? Will replacing those 3 valves do the trick? Thoughts on the vent valve at the top of the tank?

Thanks for any advice or insight...

Macster

Liquid fuel must never be allowed in the canister. To prevent this there are couple of different ways. One way is the tank has a dome with a vent at the top. There is a restricter valve which limits air flow and once the tank is "full" the air in this dome with the restricter valve prevents the tank from filling any further. This leave a dome of air to allow for expansion of the fuel.

If this valve has failed then it can let all the air leave the dome and if the tank continues to be filled this can force fuel up the line into the canister.

I won't cover the other ways as it appears from the parts drawing the GT3 tank uses this system. There is a "bleeder valve" up high on the fuel filler tube that has some lines that I think run over to the canister.

That fuel sloshes out the filler neck suggests fuel is coming up this line as the line connects to the filler tube up close to the top. Normally this is to direct vapor from the tank filling up to near top of the filler tube so the gas pump nozzle vapor recovery system can remove the vapor.

Also, as the fuel tank contents heat up and they do as the car is driven from the heat of the fuel pump at least this can force fuel up the vent line and into the canister. Even if the car sits unused if the ambient temperature changes enough this can heat up the fuel and if the tank is full force fuel up through the vent line. Once the fuel is in the canister it can't leave. The canister and its load of charcoal is intended to only be exposed to fuel vapors which it collects and then upon engine start or shortly after filtered air is routed through the canister which removes the vapor collected by the charcoal. This air continues on to the intake where it gets routed to the combustion chambers and burned.

So, based on my layman's analysis I'd suggest you look at item 12 in this diagram as the likely culprit:

http://www.autoatlanta.com/porsche-p...section=201-00

Of course this is not a hard diagnosis, I'm not a Porsche tech by any stretch of the imagination. The part is $126 which ain't necessarily cheap especially if it is not needed, and I don't know what level of work it is to gain access to that area.

NaroEscape

Macster - I agree that liquid fuel should never get into the canister. What I can't figure out is...how is it happening? I don't 'top off' the tank or anything like that. In fact, it never really fills since this issue started, even AFTER we replaced the canister the first time. What caused it to happen to begin with??

Yes, I am replacing the #12 "bleeder valve" you indicated - p/n 996.201.143.01 There are a two other valves - 996.605.201.01 and 996.605.213.01 that I referenced above that seem to regulate air and flow to the canister, so I'm replacing them too. I already have them out (which is how I knew about the fuel in the canister). #6 and #9 on this diagram:

http://dcauto.gotdns.com/illustratio...ghlight=194982

I can still blow air in and out of each of the 3 fittings on the canister quite easily so there is no blockage or anything there.

Like I said, what concerns me most is that valve on top of the tank - which might be the 'restrictor valve' you refer to - 996.605.203.01 that I can't see how you can access. It's not on the image you provided, but it is here: #21 Fuel Tank Vent Valve

http://dcauto.gotdns.com/illustratio...ghlight=194886

And then comes the question - why have we always smelled fuel in the cabin?

Macster

The link you provided regarding item #21 is not a GT3 fuel tank diagram. It is just a plain old 996/997 fuel tank. The GT3 fuel tank is different at least based on what I found at the link I provided which was to a GT3 fuel tank.

If the fuel vent line/valving is messed up this can cause a build up of pressure in the fuel tank and most if not all the fuel/vapor lines. In some cases the cap can vent over pressure (or under pressure). In an enclosed space upon engine start the cabin air intake can pull in fumes that the cap may have vented.

If the system is experiencing over pressure at other times fuel vapor (possibly even liquid fuel) can leak out. If this is occurs inside any portion of the body smellng fuel in the cabin is a likely result.

My general rule with newer cars -- since fuel vent/vapor recovery systems became standard equipment -- is any odor of fuel is due to a leak of the fuel system, either a liquid line or a vapor line. (Pretty much the same rule applies to cooling system. Any whiff of an antifreeze odor is due to a coolant leak.)

Macster

I agree that liquid fuel should never get into the canister. What I can't figure out is...how is it happening? I don't 'top off' the tank or anything like that. In fact, it never really fills since this issue started, even AFTER we replaced the canister the first time. What caused it to happen to begin with??

Assuming you are not purposely overfilling the fuel tank with the vapor recovery system probably not working right the normal air space that would remain in the fuel tank may not be there. Also, the fuel nozzle shutting off may if continued to be ignored -- and I don't intend this to sound mean it is a natural reaction to when the nozzle is beleived to have shut off "early" to try to add more fuel to the tank ( I do this myself when this happens with one of my cars, though often I actually walk around and turn on the key and check the fuel level to make sure the nozzle shut off prematurely) may result in an overly full fuel tank.

Normally the canister should receive only vapor. But obviously something is wrong and is letting fuel pass through a valve/line that should only carry vapor. It is not clear to me if this vent line and any valving in it in the event it does get liquid fuel in it if there is a mechanism that lets this fuel drain out again.

If not then even if at first if the fuel is not high enough to reach the canister it can as the vehicle is being used be forced higher and on into the canister.

NaroEscape

While that may be true, the link I provided shows exactly what is on our GT3. The link you provided does not have the valve shown, but it exists as I was trying to take it off earlier today but cannot access it (it's under the framework of the chassis that's above the tank). Additionally, the diagrams from AA show that the carbon canister does not have those additional two valves, which ours does (I have them sitting on my desk right now). I noticed also that there are multiple diagrams for fuel tanks for the 996 models...GT3 or not.

This is probably more respresentative of what we have.

http://dcauto.gotdns.com/illustration/index/2160710


Agree on all points - if you smell fuel (or any other odor), there is a leak of some sort. I don't believe it's a liquid fuel leak because you don't smell is all the time. And while I generally agree with your assesment that any sort of overpressure may allow small amounts of fuel vapor to escape, thus be able to be detected through smell, I am just not sure what is CAUSING the overpreassure. Over pressurizing the tank will also lead to the difficulty of fueling also, so I beleive that is the root cause here.

I know what the operation of that vent valve on the fuel filler does, so hopefully replacing it will fix the problem. We'll see...

NaroEscape

Well....when the tank is empty and it stops after only putting in 1/2 gallon...yes, I am guilty of trying to add more fuel.

I'll report back after replacing the valves.

blake

I'm suffering from the exact same two issues after longer track days, but don't plan to address it until the long winter months... Looking forward to your update!

TurboCup87

Replace the rubber gasket where the pump/sender attaches. They don’t seal well after being taken apart a time or three. A worn one allows persistent fuel smell in the cabin. Not saying this is the cause of your canister filling up.

Burger

I recently replaced my vent valve (996-201-143-01) and am now able to fill the tank, and was also experiencing the fuel smell which is no longer present.

Hope you have the same luck!

NaroEscape

Update: I replaced all three valves and it SEEMS to have fixed the fill issue - I can put 12+ gallons in it - but when it fills, the gauge does not say it's full. And I don't want to add any more for fear of overfill issue and this whole thing happening all over again.

I've now filled it twice and had the same outcome

however...the fuel smell is still there after long drives. Running around 15-20 minutes at a time, no smell. But after I visited a customer about 45 minutes away, I smelled a whiff of gas once I got close to home after the meeting. You can't smell it outside, when you open the front trunk or sniffing anywhere around the trunk or fill area...just inside the car. Very weird.

Macster

Kyle touched upon it. Consider the seal/rubber gasket where the pump/sender attaches. However you report you do not smell any gasoline odor when you open the front trunk. If this were leaking I'd think you'd know it.

My only other thought would be gas or vapor is leaking from around the tank fill tube and you catch a whiff of this when the car's cabin ventiliation system pulls in the fumes.

NaroEscape

Follow up: yesterday I filled the tank for the third time since I replaced all the valves....and its not filling all the way. Drove it down to the light coming on, and could only put 10 gallons in. After that, less than 0.1 gallon at a time, so I stopped. Gauge says 3/4 full. So while replacing all these valves somewhat fixed it and it's better than before, there's still a problem.

sp0da

I recently replaced my fuel pump in an attempt to troubleshoot what eventually turned out to be a dead ECU in my car. Once my car fired up and I filled with gas, it wouldn't go over 3/4 tank. The issue turned out to be that when I reinstalled the fuel pump/sender, the fuel gauge float was tangled with some of the fuel hoses. After removing the sender, and then pushing the hoses towards the front of the car, then reinstalling the sender, it free'd up space for the fuel float to operate properly and I can now fill the car and the gauge reads full.

Agree on Macster with the sender gasket - I used a brand new one when I did my fuel pump R&R and it was extremely difficult to use it for round 2. The fix for me turned out to be sticking the gasket in the freezer for 10 minutes before reinstalling. YMMV

NaroEscape

While I understand your guys point here, the sending unit or anything on the top was never removed or disconnected, so not sure how/why an issue would have occurred there?