Fred!

[Dave M]

A while back I had a 1441 code and you pointed me to the purge solenoid or flow detector as the likely culprits. When I reset the codes, I would run without getting the 1441 until the computer "relearned' the sensors again, then the code would activate again (usually after about 100 miles). I replaced the flow detector and got no change. I then replaced the purge solenoid and that corrected the problem for a few weeks (a few hundred miles). Now I got the code again. The problem was definately fixed for a while, the computer had relearned again and was running fine, now the code's back again. Can you give me any insight on why it would be back?

Thanks!

[Injuneer]

There's no "learning" involved here.... the PCM actually "learns" very little - long term fuel corrections, IAC startup setting and a flag for continued use of low octane fuel. It doesn't "learn" from sensors and switches. Some codes will only set after the problem has happened a couple of times - that might explain why it takes a while to come back.

Did you check your hose connections on the purge solenoid to make sure you have them in the correct inlet and outlet positions - a lot of "uncured" problems track down to "switched" hoses. Shoebox has the hose routing diagram on his website.

I have seen a few cases where nothing cured the problem. I don't know the details of the setup, since my 94 doesn't have the flow detection switch. Something that causes a vacuum in the fuel tank might possibly pull backwards thru the vent line.... but that's sort of off the wall. There is a pressure relief valve that should prevent excessive vacuum in the tank.

[Dave M]

I'm pretty sure I routed the hoses correctly, but I'll double check, I have the diagram. The reason I say the computer learns, is because when I pulled the codes with my scanner, it automatically erases the codes from memory, resetting everything, like I had unplugged the battery. I took the car to be emmissions inspected shortly after having pulled the codes, and the emmissions machine wouldn't let it be tested because the cars computer was coming back "not ready to be tested" because the sensors were not providing any input to the computer. The shop told me that's what happens when the battery is disconnected, and that I should drive it a few hundred miles to get the computer "ready". I did, and it was. So, what I think was happening, is that the computer wasn't "seeing" this sensor for the first hundred miles or so after I would reset the codes. Once the sensors started talking to the computer again, it would trip the code again, like clock work.

[Bond Hawk]

When my car was under warranty, I had the same problem. After many replaced sensors and other parts, they found that a piece of charcoal from the cannister had become lodged in one of the lines, thus blocking the line (or at least obstructing it) from purging correctly. Try looking for that as your problem.

[Dave M]

When my car was under warranty, I had the same problem. After many replaced sensors and other parts, they found that a piece of charcoal from the cannister had become lodged in one of the lines, thus blocking the line (or at least obstructing it) from purging correctly. Try looking for that as your problem.

Thanks for the tip!! Any idea how charcoal would have made it out of the canister and into the line?

[Bond Hawk]

I've never taken the canister apart, so I'm not sure how it could have happened. The car had less than 10K when all of this was happening. Have you removed the canister? It looks like you can get to it from inside by removing the molding next to the rear passenger seat on the driver's side. I could tell they had done that after they finally fixed the problem.

[Injuneer]

It isn't unusual for the charcoal to break down into smaller particles, and make it into the purge line. However, that would plug the line giving you the code for "no flow during purge". But your problem is "flow during non-purge". It is possible if you are experiencing a breakdown of the charcoal, that the purge solenoid has got gummed up again.

If you have a vacuum pump, connect the hose that normally connects to the throttle body to the vacuum pump. Try and pull a vacuum on the closed purge solenoid. If it won't hold a vacuum when it is supposed to be closed, its stuck open. If it does hold the vacuum when the solenoid is closed, open the solenoid by supplying a ground (that's how the PCM opens it) and you should be able to pump a decent amount of air through the system with the vacuum pump, and no approeciable buildup in vacuum.

If all checks out, place a vacuum gauge in the purge line, near the flow sensor. Start the car and look at the vacuum gauge to see if there is any vacuum in the line when there shouldn't be. Possibly, as I noted earlier, the fuel tank is developing a vacuum (no air coming back in through the pressure relief valve, as the fuel is used up) and that vacuum could be tripping the flow sensor switch. It might take a long drive to develop the kind of vacuum required to trip the switch.

As far as the sensors learning, I don't think thats what your mechanic was explaining. What he was indicating was that immediately after you "clear" the PCM, it is showing "not ready for test".... and I suspect that is done to prevent people from clearing all the codes a few minutes before they pull into the emissions inspection station. All the sensors are working during this period.... the PCM is simply not letting anyone do an emissions level scan on a PCM that has just been cleared.