Todhunter

I've got a 1997 Honda Accord LX, with an F22B2 engine. Car runs great, but about two weeks ago the CEL came on. I pulled the code and it is P1457 (EVAP Emissions Control System Leak Detected). I already checked the gas cap, and the gasket looks fine and it was and is tight (I am allowing it to click multiple times when closing it).

I have an Accord HELM manual, but it is for a 1994, and the section on testing the EVAP control system is lacking some of the gadgets I am seeing. I started following the troubleshooting section on Page 11-123. No vacuum from the EVAP purge solenoid when cold and throttle is snapped to 3K, and there is vacuum when the engine is warm and the throttle is snapped to 3K (this is what you want, according to the HELM).

Now for where things get different. The manual shows the EVAP canister as having two hoses going in the top (one from the tank, one going to the EVAP solenoid), and the fresh "purge air" hose going into the bottom. My canister has the two hoses going into the bottom of the canister (one from the two way valve, one going to the EVAP solenoid), and what I have to assume is the fresh "purge air" hose going into the top. Also, the Canister Vent Shut Valve on top of the canister is not in my HELM, my two-way valve has some kind of solenoid on it (HELM doesn't show this), and the HELM also doesn't show the Vent Pressure Sensor.

Can someone help me out and hook me up with the EVAP control system section of a HELM for a 1997 Accord LX? If not, I can take some pictures and maybe someone can help point out what solenoids are what and how to troubleshoot them.

Personally I think it's something with my two-way valve. When I hooked my vacuum pump up to it to test it, it made some kind of moaning sound that I have heard before after I shut the car off. I couldn't properly test the two-way valve per the HELM I have, as my vacuum gauge doesn't read as fine as it requires.

Thanks in advance.

ETA - see picture below of what I have:

phootbag

The solenoid valve with question marks ( ? ) is probably the 2-way Bypass Solenoid. This solenoid is used during the P1457 diagnostic function by the PCM.

The 2-way valve has a diaphram that acts as a one-way valve. When you apply vacuum in one direction you should have flow. When you apply vacuum in the opposite direction you should be able to build vacuum ( no flow ) and the needle on your vacuum pump gauge should hold steady. If you have flow in both directions or the needle doesn't hold steay and visibly falls when it should hold, the 2-way valve OR the bypass solenoid is leaking. Determining which one is leaking ( they are connected to eachother ) can be a pain in the ***. I usually just replace them both. ( they aren't that expensive ).

phootbag

The system itself is split into two sections, P1456 ( tank side ) and P1457 ( canister side ).

I am convinced the system was designed with lots of "second thoughts". Current designs don't have a 2-way valve or bypass solenoid. Much easier to diagnose.

The system you are working on has a purge control solenoid ( the mouth of the system ), the canister vent shut valve ( the butt hole of the system ) and the 2-way/bypass unit ( a middle sphincter to seperate the system into two halves ).

The function of the PCS is to allow vacuum to enter the EVAP system for diagnosis ( and to control vacuum during normal purging ).

The function of the CVS is to shut off the opposite end of the system during testing.

The fuel tank pressure sensor is located on the tank side of the system.

You can test the system yourself by powering the CVS to close off the butthole.

phootbag

Power the bypass solenoid to open both sides of the system creating one large system. Remove the line from the system side of the purge solenoid and apply vacuum. You should be able to build and hold a vacuum. If not, you have a leak and need to find it.

If you can hold a vacuum there are no external leaks. You need to test for internal leaks and solenoid valve function.

The 2-way/bypass can be tested like mentioned above. The canister valve is a simple soilenoid valve and can be tested by powering it, checking for flow. Leave it unpowered and check for no flow. Pretty simple. Same goes for the bypass solenoid and purge solenoid.

I have replaced a few PCM's because the drivers for the bypass solenoid were burned up and couldn't operate the bypass solenoid during the PCM's test. This is something you can only confirm with Honda's scan tool by commanding the PCM to operate the valve and checking for the command at the solenoid.

My day off today and just woke up. Hope this made sense.

phootbag

BY the way, the p1456 is tank side and the p1457 is canister side. Since you have a p1457 you don't need to focus on the tank side, fuel cap etc.

Todhunter

Thanks phootbag - lots of good info. I will test the two-way valve again. My vacuum gauge is 0-30 inHg, and I need one with a finer scale (HELM recommends 0-4 inHg) to properly test the two way valve, as the gauge needs to be able to discern 0.2 inHg from 0.6 inHg. But when I did try testing it with the pump/gauge I have, it never even tried to build a vacuum (that I could see on my gauge), and just made that moaning sound.

I'll also test all my solenoids. I think my buddy has a Snap-On scan tool - it may be able to command sensors to operate.

Then again, it may be worth it to just go ahead and spend the $50 on a new two-way valve.

phootbag

The moaning sound is probably from pulling vacuum in the direction that normally allows flow. The diaphram is probably fluttering. Try pulling vacuum from the other direction and see if it holds.

It would suck to replace a part that isn't bad.

Todhunter

Just an update, and hopefully closure.

I bought a new two-way valve and replaced the old one yesterday. Upon removal of the old one, I am almost certain it was the problem. In the picture below, you can see that the part cracked around where both nut-inserts are cast into the plastic. I believe these cracks allowed the clamping force from the screws to be reduced, causing an air leak between the two-way valve and the solenoid that bolts onto it. Hopefully after the car goes through some driving cycles, the CEL will be gone for good.

Thanks to all that gave advice!

Todhunter

Still getting the CEL. I am going to go through all the hoses and make sure there are no leaks this coming weekend.

Todhunter

I ended up having to replace the battery, and after that, no more CEL.

bjgaz

I have had this code appear twice on my 1997 Honda Accord SE 2.2L. Both times I have needed to clean the solenoid valve that is mounted on a bracket by the Cruise control I thinks it's the canister purge valve. It has one line coming from below the canister and one line to the intake. It has Denso PCS3 on the label.

I verified that the solenoid coil was good with an ohmmeter. I then applied a piece of rubber tubing to one port and tried to blow thru the valve. It was closed (normal). I then applied 12VDC to the solenoid via the two terminals on the device and hear the valve click. Once opened I blew into the valve but it was restricted. I removed 12V power and shot some WD-40 into both ports and waited 30 seconds, drained off the excess and reapplied power. The valve opened and closed freely and without any obstruction.
Beats paying for a new valve. I have 200k on my accord and live in Tucson.
Hope this helps someone.

butch burton

P1457 problem solved! Was starting to get oil leak and thought the oil pan gasket was bad. Had my mechanic look at it and when I went in to pick up car - he had a rusty and greasy oil pan and it was leaking because the salt had eaten through the oil pan. It also was allowing the vacuum to leak out of the system. Had already replaced all of the evap system including rusted steel fuel supply lines.

So those of you who live in salt country - check out that oil pan. Oh only have 331K on my accord.

Butch