Lessons learned for DTC P0440

I have a 1997 Bonneville SE (not supercharged). About 3 years ago, my gas tank developed a small leak. IMO, there’* nothing worse than working on the gas tanks so I had a local garage replace the tank with a new one. Note: the internal components such as the fuel pump, fuel level sender, roll over valve, and pressure sensor were transplanted from the old to new tank. Initially all was well with no leaks and no DTC’*. Then after about 2 months I started getting DTC=P0440 which is listed as the EVAP Emission (EVAP) System. For about a year, I lived with this code. Since I live in NY, passing the emissions test is required. I found that during the real cold months, the periodic test that precipitates the P0440 is not run during extremely cold weather and usually if I time the inspection right, I could get through inspection. After about a year of having the gas tank replaced, my fuel pump died. This time I did the job myself and replaced the pump and pump filter. While I had it all apart I tried to locate what might be causing the P0440. I did locate some pin hole leaks in the fuel tank filler pipe which had undergone quit a bit of corrosion. While I had the filler pipe away from the gas tank, I cleaned, brazed, pressure tested, and then painted the filler pipe. I assembled everything back together and all worked well…… at least for about 3 months, until the P0440 once again reappeared. It was frustrating.

I do have the factory service manuals for this car and I also accessed a lot of the posts on the Bonneville forum for more helpful info. If you ask my wife, she would tell you I spent way too much time on this problem.

IMO, the FSM could have been much more clear on this EVAP system layout. What is not made blatantly clear to the uninformed is that there is a difference in the layout between the SE=VIN(K) non-supercharged and the SSE=VIN(1) supercharged configurations.

The FSM shows a diagram of the EVAP layout, but fails to identify this particular figure as unique to the SSE. Since this was not done, this error made making sense of what was really going on with my SE much more difficult.

The following link shows two excerpts: 1.) is the EVAP layout ( FSM pg 63-552 ) ( which is not labeled as unique to the SSE) and 2.) Engine controls Component Location (FSM pg 6E-45 ) which was what finally contributed the clue to this confusion.




Bottom Line, the EVAP Layouts are grossly different between the SE and SSE… but the manual only tells you this indirectly via the Engine Controls Component Location table. Here I am looking for this “ EVAP Vacuum Switch” on my 1997 SE and …. There is not one on it ! Boy, did I waste a lot of time on that one.

So, the FSM does not give an informative and useful layout diagram for the 1997 SE EVAP system. Below is my perception of the 1997 SE EVAP Layout… please excuse the poor chartsmanship :



If you compare the EVAP layout of the SSE, which is shown in the FSM, to the hand drawn layout of the SE, you can see that the SSE does not have a pressure sensor in the gas tank as compared to the SE. Likewise, the SE does not have the EVAP Vacuum Switch that the SSE does have.

From what I could find on the Bonneville forum, I think one of the best descriptions regarding the EVAP system was :

https://www.gmforum.com/showthread.p...=evap+canister

Look for the SSEBONNE4EVA post….

My only comment regarding the SSEBONNE4EVA post is from what I can tell from looking through the FSM, the P0440 is also defined as a “ Large Leak Test Failure DTC “…. See the following to photos of the FSM… note these two photos are from the same FSM page.



So with all that,…. Let me cut to the chase. It was the rollover valve in the gas tank that was causing all my problems. Look at the poor chartsmanship picture. If I was to leave my gas cap loose, and the evap system test was to run, the PCM would detect that it was incapable of acquiring the required vacuum which I think is around 5 inches column of water relative to ambient. OK… lets say the rollover valve is stuck closed but the gas cap is on OK. The test would run and again the required vacuum would not be attained thus throwing the same DTC = P0440. That is what was happening in my case.

Initially it was unclear to me what was really going on. I had this image of the EVAP system attaining the desired vacuum but tapering-off too quickly. That was not the case… if it was, I should have gotten a DTC = P0442. But for my case, my 97 SE couldn’t even do that. The system could even attain the desired vacuum regardless.

So, before I really understood all this, I started farthest away from the real problem… Murphy’* Law….. The Canister Vent and the EVAP Charcoal Canister. I removed the PCM and then the EVAP charcoal canister. This canister has Qty=3 hose nipples. 1.) goes to the vapor line that trails back to the gas tank. 2.) one line that goes to the purge valve that sits on top of the forward most rocker arm cover gasket. 3.) A very large diameter hose, approx ID = 3/8” which goes to the Canister Vent Valve / Filter assembly which sit directly on top of the EVAP Charcoal Canister.

See photo below:


After removing everything off the canister, including the vent valve assy, I plugged up 2 of the 3 lines and then put a vacuum on the last line to see if it would hold vacuum…. And it did. It also did not exhibit any liquid gas inside the canister… which was determined solely by feel .

Next I moved my attention to the vent valve which should be open with no power applied to it. Using an extra piece of vacuum hose I had laying around, I tried blowing through it using the vacuum tube and my mouth. A good flow was observed. Next using a couple of test alligator test clips, I put 12 Volt DC from the battery to the terminals of the vent solenoid. Using the same test approach, now no flow was observed. Additionally, I tried to suck air back through is like one might try to suck air out of a soda bottle. I was convinced that this valve was working properly.

I next did the same procedure with the purge valve… The only difference here is that the purge valve is normally closed when no power is applied to it… which is the opposite of the vent valve. The conclusion was that the purge valve was also working properly.

Next all the rubber hose connections were replaced.

Next I removed the gas cap, so that no pressure could build in the tank and then I tried to blow back from the vapor hose which attaches to the EVAP Canister back into the gas tank. That is when I really started getting somewhere. It was like it was blocked. So I went underneath the car just in front of the passenger side rear tire where the fuel filter is.
I disconnected the steel line from the plastic line, for the vapor line, which is the smallest steel line of the 3 steel lines… in that vicinity. ( 1.) the biggest being the Pressurize fuel from the fuel pump in the gas tank, ( 2) the middle size : the return line from the pressure regulator on the fuel rail on top of the engine, (3) the smallest size: the vapor line from the gas tank rollover valve to the EVAP canister .

I could get air flow from the steel line back to the EVAP canister, but almost no flow from the plastic line back into the gas tank. Applying a vacuum rather than a pressure also yielded no flow…. At that point, I knew the culprit was somewhere between this junction and the interior of the gas tank. So in a desperate attempt to dislodge whatever could be causing the blockage, I used an air compressor on the plastic line at about 120 psi which still yielded no flow.

So I removed the gas tank from the car and removed the sending unit assembly from the tank. I thought I might find the vapor line kinked or crushed somewhere, but all looked fine. I even used a piece of mechanic wire to feel up the tube for a blockage, but everything was nice and open. The only thing left is the rollover valve.

The purpose of the rollover valve is in case the car is in an accident and ends up on it roof, then the top of the gas tank now becomes the bottom of the gas tank. The liquid gas is now at the opening of the vapor line which terminates at the normal top of the gas tank….
Without the rollover valve to now block this passage way, pressure in the tank could push liquid gas through the vapor line down to the EVAP canister which it then could make its way to the vent valve, which could be open, especially if the ignition was turned off and exit to the atmosphere through the vent valve filter and land on something hot like an exhaust manifold causing a fire.

Here is 2 pictures that show this rollover valve:




Although I’m sure it’* listed a non-serviceable, it is easily disassembled.
Here is my rough drawing of its internal components :



Here’* what happened…. The rubber component which has a needle point on it which meshed with the valve seat which is the approx. .040 “ orfice in the plastic housing had swelled in diameter so that is would no longer fall downward away from the valve seat.

As the car would hit bumps in the road , etc. the ½ “ ball would rattle around and occasionally impart an upwards impact on the brass cylinder and also the swollen rubber valve components. This would make the rubber valve move upwards, but now with the increased diameter of the rubber component, it would not move back down afterwards. Consequently, the result is that the vapor line would remain blocked under all circumstances.

If I had to guess, I would say maybe the change in chemistry of the gasoline over time has caused this rubber part to swell.

I hope this helps some of you out there.

For my situation, I made a replacement part for this rubber part. But the consequences of doing this wrong could be back is in a rollover accident.

I hope this helps some of you guys out there.

Jack