Paul Waterloo

When I bought the car, had a fuel smell in the car, when it sat in the sun, it was crazy bad. Changed the gasket on fuel gage sending unit, 95% went away (I thought it all went away at the time).

On very hot days when I run the AC, I have finally figured out that I get a fuel smell in the cabin from under the hood. This only happens when the car is not moving or I am running slowly, 40 MPH or less. If I'm on the highway, there is not enough vapor being brought into the cabin to notice.

I finally figured this out yesterday by changing the HVAC to recirc and the smell would go away after 2 or 3 minutes. Take it off recirc, and it comes back.

Looked under the hood, and I can see the two flaps that close off the fresh air intake (just look under the plastic cover in front of the windshield from the fender) when in recirc, and then open when the switch is in its normal position.

QUESTION: What should I be looking at? I looked at the two fuel line connections, but don't see any white residue on them from leaking fuel. It only happens when it's above 80 degrees F or so. I have not looked at anything in detail as of yet.

What other components/area would you be looking at for a ever so slight fuel leak? When I open the hood just after shutting the car off, I can't smell anything.

Thanks!

Paul Waterloo

Just doing a quick google search, sounds like a need to inspect the charcoal canister system. From Clark's Garage:

The fuel tank has a ventilation system which is designed to prevent pressure buildup in the tank by removing fuel vapor. The system has hoses that run from the tank along the bottom of the vehicle on the right hand side, into the engine compartment, along the back of the firewall, to a charcoal filter canister in the left fender well (must remove the interior fender well cover to locate canister).
If any of those fuel lines is loose, cut, or leaking, you will get a strong fuel smell inside the passenger's compartment. Also, if the charcoal cannister itself is split or otherwise leaking you'll get a raw fuel smell in the passenger's compartment.

maniaquede968

Also have a look at the filler neck/tube because sometimes the collar gets loose and it smells in the car. There is a 944 tread on the subject. Is the smell worse after you fill up thecar?

Paul Waterloo

No, the smell is definitely coming from the air vents with the windows closed. I took all the fuel filler stuff apart last year, and yesterday I filled the car all the way until the pump stopped (never did that before) and no smell at all. I usually stop at 15 gallons.

I could always use my nose back there to smell fuel, and yesterday there was nothing, I could pull the carpet back and nothing, where it used to smell terrible before I changed the fuel sending unit gasket.

Jfrahm

--edit-- just saw your new message... you might be able to pressure test at the top of the sending unit.

I had a cracked nipple on the top of the fuel level sensor unit, and a rotted hose due to that leak. JB weld and new hose sorted it. I do not know if your recirc test eliminates this or not (or just changes the cabin pressure and airflow) but it's worth a look back there.

You might be able to pressure test the system at various points also.

Paul Waterloo

So here is what I did today:

Took the inner fender off to expose the charcoal canister, I know the clamps on the hoses are OEM, but I replaced them. It's easy to get to, just pull the inner fender out, has about 8 or 10 fasteners.









I leak checked the canister with the old "blow into it while plugging up the other two hole" method....looks to be good, I believe it was replaced just a few years ago.

Put these regular hose clamps on top and bottom....the old ones were very loose.



On the inside, I removed and inspected everything. The electric purge valve is not working, it is powered from the ECM, it's open when it is de-energized and energized. It picks up electrically, I can hear it, but the valve does not change state. I bought a used one off ebay this afternoon. OEM is $200 plus.



I also checked the vacuum valve, it was working, changing states when vacuum was applied and when it was removed.

I changed the hose clamp on the connection to the intake manifold also, which seemed to tighten up the connection. It's down low on the manifold, you are seeing the left hand fender and just behind the alternator where the connection is in this picture.



I also pulled the beauty cover off the engine and inspected for leaks at the fuel rail didn't see anything.

Then I drove it around this afternoon with the recirc OFF. This would normally produce a fuel smell in the cabin, it was 88 degrees F out, but I didn't smell any in my 30 minute drive. I don't know if it's fixed or not, but at least I feel better right now.

I don't know what the electric purge valve does (I know it opens the line to the canister), but I do not know its normal state. I would guess it's closed with the ignition off (0 VDC), and open with the ignition on (12 VDC). I'll see once I get the replacement.

I'll update this with the long term test over the next week or so.

biosurfer1

Can anyone elaborate on a fix if there is fuel smell after a fill up? I get mild to moderate whiffs of gas after a fill up and I'd like to fix it if possible. Neck filler?

Paul Waterloo

I would start here....would also help to know if you have a coupe or a cab.

https://rennlist.com/forums/968-foru...in-my-fix.html

biosurfer1

thanks....coupe

MB968

Paul, thanks as always for a good writeup with nice pics.

I fixed an issue a year or so ago that was from the metal filler pipe that comes out of the top of the sender unit. The tube is swagged into the metal cap, but the interference fit on mine was no longer tight. With a fill-up, 24 hours would net a teaspoon of fuel sitting in the relief under the cap.

But, lately when it has been hot here, I've noticed a smell in the car. I'm going to run your tests with/without recirc to see if I get similar results. I'd assumed my smell was coming from one of the hoses etc. that are in the filler area. Since they are so difficult to get to, I think I'll give the charcoal canister a look, as well as the control valve. Maybe I'll be as lucky as you.

fejjj

I had a similar problem and found my evap line disconnected.

Paul Waterloo

Which one is the evap line?

Paul Waterloo

Okay...so I got the new solenoid valve, posted this on the main forum....no response, does anybody know what this does?

I've been troubleshooting a fuel vapor smell in the cabin, and I think I resolved it with putting some new hose clamps on and tightening others in the carbon canister system.

However, when I was testing everything, I found that I could get the vacuum valve (item 15 below) to change state when applying/removing a vacuum from it, but I could not get the electric solenoid valve (item 14 below) to change state when it had 12 VDC applied/removed.

This was based on my "blow/suck" oral test.....it has 46 ohms, you can hear it pick up when you energize it (when the ignition goes on, it is powered from the ECM), but it really has no effect on being able to blow or suck through it, it still seems open.



So I bought a used one on ebay. New ones are over $200. It also has 46 ohms, I can hear it pick up when energized, but once again, it really has no effect on being to blow/suck through it.

What would be the purpose of this valve? I have no idea and have no idea if mine is working correctly or not, would think it would change states.

There is an arrow on it (see adjacent to electrical connector) which I would guess means it would allow vapor to go from the canister side back into the intake manifold, but not allow vapor to go from the manifold to the carbon canister.







It plugs into the intake manifold on left hand side in the picture above, and goes off to the carbon canister on the vacuum side. The vacuum is connected right at the throttle plate, so only when you are off throttle and vacuum is present behind the throttle plate does the vacuum valve open.

Thanks for any input!

MB968

Don't know the system, but have you considered that the elect connection is not power to the valve. Check resistance across it with and without vacuum and with various levels of vacuum. Maybe it is a switch that is activated (powered) by vacuum not powered by electric.

Paul Waterloo

That is an interesting concept as it is powered from the ECU, it could be more of an input into the ECU that changes on vacuum or no vacuum. It doesn't sound correct to me....but I've been wrong before! Thanks for the thought!