2005 Boxster will not fuel / spits fuel back / no error codes - SOLVED

[Jim Long]

2005 Boxster 2.7L Manual transmission

 

I am trying to diagnose the problem of not being able to fill more than 2 gallons of fuel without it spitting back at me. Just to start, I've tried:

 

1. the "turn the filler nozzle sideways / upside down" trick ---> does not work

2. multiple gas stations with different types of vapor recovery nozzles ---> does not work

 

Background: I've owned the car for approximately 10 months. It has always been difficult to fill. Lots of stopping / restarting of the fill process but recently gotten worse whereby fuel is making its way back up the fill tube and spraying out of the neck filler. Not cool.

 

High level diagnosis: air is not being evacuated from the tank during fueling.

 

I have no DTC's and no issue in starting, drivability, etc.

 

I have read the previous discussion on this subject here:

I have also read similar issues on other sites including two official tech training videos on the EVAP system including the subject of ORVR (Onboard Refueling Vapor Recovery).

 

Most issues discussed on other sites relate to trouble code concerns usually landing on the common purge valve issue where it fails in the open position (relating to too rich, too lean, fuel trim issues, etc.).

 

None of the training was 987 specific as each manufacturer has to package their own systems of parts to meet the emission requirements but the basic concepts remain the same.

 

Here's where I have examined / tested my system.

 

1. Removed the tube from the tank vent to the charcoal canister and blew air through it - clear, no issue

1. Blew low pressure air down through the filler neck and witnessed air coming out of the top of the fuel tank vent - clear, no issue

2. Removed the charcoal canister and blew air into the inlet - clear, no issue, no charcoal bits or pieces found anywhere, smell of canister was very faint of gasoline (so I assume it is not saturated)

3. Examined the LDP (leak detection pump) and wiring, appears clean with no issues, air filter is like new clean.

4. Separated the purge valve from the forward cabin (connects at rear of cabin low, driver's side USA) and blew air from the carbon canister outlet connection to this rear port - clear, no issue.

5. Removed the purge valve and its related tubing from the air intake down to the rear port

5a. Tested the purge valve actuation via Durametric before disassembly (clicked open, no issues) and on the bench under vacuum (same, no issue), electrical resistance measured 24.4ohms

5b. Ordered new purge valve for replacement anyways (due to pain of removal, not risking a future failure of a 17 year old component and having to go back in there)

5c. Completely disassembled the purge valve and noted no trouble found in any physical internal components, retested electrically and watched the unit operate correctly sans plastic housing.

5d. Blew air into the remaining tubing leading from the rear port to the air intake - clear, no issues.

6. Replaced charcoal canister connecting it only to the fuel tank and went to the gas station - filled fine up to 2 gal and then continued to fill "full blast" until a total of 6 gal was added to the tank - no issues.

 

Since this last step worked with an open atmosphere vent in an otherwise closed system I now suspect that I have not identified all of the venting that occurs during the ORVR process. I believe this vent, generically called the vent solenoid, is a part of this part on the top of the fuel tank:

 

 

Does anyone have any more info on this component or how the ignition off, power off refueling sequence works to move tank pressure and allow fill-up on the 987 platform?

 

Thanks to all who have gotten this far!  Jim

Edited January 17, 2023 by Jim Long
solved

[Jim Long]

Ok, I figured out the answer:

 

During fueling the air escapes the top of the fuel tank into a hose connected to the bottom of the charcoal canister. There are two exits the air can take after entering the charcoal canister:

 

1. through a small diameter line to the back of the car which is normally closed at the purge valve (or regen valve if you prefer). 

--> this is only open (vented) during normal engine operation per ECU instructions, not at all related to fuel filling at the pump.

 

2. through the leak detection pump (LDP) system port attached to the canister.

--> this was the confusing part because I thought the LDP was only a one-way system, allowing flow into the canister for pressurizing the system to check for pressure drop (i.e. leaks)

--> turns out, when not pushing air into the canister per ECU request (for example, during engine off refueling at the pump) it allows air to flow back through the LDP and out the air filter attached to the LDP. THIS is the primary way the 987/997 vehicles vent fuel vapors to the atmosphere during refueling.

 

So, after further tests, the flow rate through the charcoal canister has to meet a minimum flow rate to allow for continued fuel filling operation. For reasons yet TBD*, the flow rate of vapors through my charcoal canister were insufficient to keep up with the needs of the escaping vapors from the fuel tank. 

 

*I plan to cut open the canister to see how it has become restrictive to flow. 

 

Solution: replaced charcoal canister (currently NLA Porsche) with a used one found on eBay. Fueling function fully restored.

[Jim Long]

I cut open the old charcoal canister after replacing it and found that the flow of vapor was being restricted by the moldy, hard charcoal of the outflow section of the canister. The inflow section charcoal was clean and perfect hinting that the moisture entry was from the other side.

If you have difficulty filling your 987 or 997 with fuel (pump shutting off quickly or fuel spitting back at you) without any recorded DTCs, consider the charcoal canister at fault. This system is identical on 2005-2012 Boxster/Cayman/911.

How moisture entered into this section and turned it into a semi-solid block over time is unknown, but it clearly affected the fuel filling capability by restricting the outflow of fuel vapor. Measurable flow was present in the old canister but when the new part came, the difference in flow was obvious.