strum456

I have a P1456 code. I do not have any vacuum leaks (internal or external). CNS, PURGE, BYPASS, all work by grounding wires on the ECM.

When driving, the FTP is not showing any sudden changes to indicate that vacuum is being pulled on the tank. However, with the CNS valve closed and the BYPASS valve open, the PURGE DOES PULL VACUUM ON THE SYSTEM.

So, the purge is opening, but the bypass is not (unless I make it). What would cause this? Is a bad ECM the only possibility?

el crapitan

try testing the ftp sensor. testing would be similar to testing a map

strum456

Thank you for the reply. The fuel tank pressure sensor works. It is around 2.5 or 2.6v at atmosphere. If I apply vacuum it goes down and if I apply pressure it goes up. The system holds vac and pressure for a minute at stays within a hundredth of a volt.

el crapitan

i'm just having a little trouble here following your terminology. what is CNS? and by bypass, do you mean vent valve?
evap systems are pretty simple. but we need to be on the same page here to get this sorted
also, the year and model please

strum456

1999 Acura Integra LS 5 speed

CNS= Canister vent valve

Bypass= valve that connects to the 2 way valve at the left bottom of the tank. It is normally closed and must open to allow the engine vacuum to bypass the 2 way valve and pull on the tank.

Sorry for the confusion and thank you for your post.

strum456

Well, I went ahead and just ordered a used ECM. I was hoping that someone on here would know how to diagnose these kinds of problems.

el crapitan

alright. let's go back to the original post. during monitoring, the purge opens, the vent shuts, the bypass opens, and vac is pulled through the intake. vac is then monitored through the ftp. so you're saying the bypass is not opening unless you manually open it? the bypass not opening shouldn't trigger a 1456. i don't think. when you performed your datalogging, the code was not present at that time was it?

strum456

I have cleared the codes multiple times. It always throws the code after it is warmed up and I'm driving at a constant highway speed for a mile or so.

I know for sure that the purge is opening, because if I ground the bypass wire at the ECM, then I can see the FTP reading drop suddenly when I accelerate or if I'm driving at a constant highway speed. If I don't ground the bypass to hold it open, no sudden changes occur to the FTP reading.

el crapitan

what i'm thinking is that the ecm is not opening the bypass because of the code. have you leak checked the system? you need to solve for the 1456.

strum456

I may have an extremely small intermittent leak. I read somewhere that if the voltage on the FTP changes less than .05v in 20 seconds, that the system should be ok. Sometimes I see virtually no change in a minute, and other times it might change .01v every 5 seconds or so.

If I have a leak, it is darn small. I've already had the tank down and checked everything out. I put vacuum grease on all the hose connections when I installed them.

Do you think a leak that changes the FTP .01 every 5-10 seconds es enough to worry about? Like I said, it doesn't do it all the time either.

el crapitan

you have a leak. that is what is causing the 1456, and therefore, the inoperative evap system. these systems can detect a leak as small as .04". this will not be visible with the naked eye. you need to use an evap leak tester. this will first pressurize the system to determine the size of the leak. with the system pressurized, you want to isolate by pinching off lines to zero in on the location of the leak. the tester can also introduce smoke into the system to detect the location of a leak.

strum456

So, because I have a leak, the system is not functioning. How would the system detect such a leak if it never pulls vacuum on it? I'd like to learn about how the system tests itself.

strum456

Another question. Why would the purge be opening but the bypass is not? Does the purge just open all the time, regardless of what everything else is doing?

el crapitan

the purge must close during self test. that is how a leak is determined. if the purge does not close, the system is not sealed and the pcm will consider this a leak and throw the 1456.

strum456

Does the bypass have to open for the system to complete a test?

el crapitan

yes. vent valve shut, the bypass first needs to open when vacuum is being drawn. then the purge shuts and bypass shuts and vacuum is monitored through the ftp. if the system does not hold vacuum, 1456 is set.
this test will run immediately upon start up of the vehicle, btw. by the time someone pulls out of the driveway, this test will been performed.

strum456

When the test is performed how much drop should i see on the ftp? I am not seeing more than .03 v or so.

el crapitan

do you mean when vacuum is drawn how much should voltage drop?

strum456

Yes

el crapitan

well, you should see a helluva lot more than 30mv.
the ftp should operate exactly as the map does. and with a map you would see 3v at wot. that is, 0 vacuum. and about 1v at idle. which is 21"hg. so since the system uses manifold vacuum, it should drop down to about 1v and stay there.
the ftp won't go all the way to 3v, but that is it's range. it'll probably be more like 2.5v
so, you should see about a 1.5v drop.
i have a feeling the purge or vent isn't closing. manually close both and apply vacuum to the system. if it doesn't hold, pinch off the hose going to vent valve. if it now holds, the vent valve is the problem. if it doesn't, try the same with the purge solenoid. you just want to isolate here

strum456

All the valves open and close and hold vacuum as they should (when I actuate them by grounding the respective wires at the ECM). The problem is that the ECM doesn't seem to be telling the valves to work during driving (with the exception of the purge).

I ordered a used ECM. According to UPS tracking, it will be arriving tonight. I'm going go driving with the new ECM and my multimeter hooked to the FTP wires. When and under what conditions should I expect to see the voltage drop?

The Honda Manual lists a procedure for testing the purge valve where the engine has to be up to operating temperature and the RPMs need to be quickly brought to over 3000 for the purge to draw vacuum. Wouldn't the same conditions need to be met for vacuum to be drawn on the tank? I am confused because I have heard from you and other sources that some kind of test is performed within 30 seconds of start up. If so, I'd like to know the conditions of the tests performed and what I should be seeing on the FTP voltage.

el crapitan

yes, cold start test is performed initially. this is done because tank pressure changes if the vehicle has been run for any period. the initial test first checks to see if vacuum can be drawn. ie, vent shuts, purge opens. if vacuum can not be achieved, a code is set (large leak) and no further monitoring is performed by the system. if vacuum is achieved, the purge shuts and the ftp monitors any pressure increase (vacuum decrease) to determine if a small leak exists. if a small leak is detected, the system will run through the small leak check several times to qualify the fault. if a leak is qualified, the system will usually now release any vacuum by opening the purge, then shutting it, then monitoring pressure to ensure that the leak was not a false alarm caused by naturally occurring tank pressure increase. that's pretty much a typical controlled evap system. certain criteria needs to be met for certain portions of the tests. that's evap readiness monitors take the longest to reset, along with the o2 readiness monitor.

but don't worry about all that. and don't worry about the manuals procedures. you know how the system works, and the code setting criteria for 1456. you can perform your own testing. keep it simple
tee a vacuum pump in between the canister and the purge. shut the vent valve. shut the purge. apply vacuum. does the system hold vacuum?

strum456

This is what I have learned, thus far, about the evaporative emission (EVAP) control system checks, and what conditions must be met to get DTC P1456 light to come on (i.e. "Tank Side" leak).


General Description

The evaporative emission (EVAP) control system stores fuel vapor from the fuel tank in the EVAP canister temporarily to prevent it from escaping to the atmosphere and to enable it to be drawn into the engine by intake manifold vacuum and burned. In addition, the onboard refueling vapor recovery system (ORVR) stores fuel vapor generated when fueling in the EVAP canister. The EVAP control system leak detection method improves the accuracy and frequency of the detection by determining faulty components and vapor leakage for each part.

The EVAP control system (from the purge valve to the fuel tank) is divided into two parts. One is the "EVAP Canister Side" (from the EVAP two way valve to the purge valve) with DTC P1457, and the other is the "Tank Side" (from the EVAP two way valve to the fuel tank) with DTC P1456. Each side uses different detection principles and methods. The tank side 0.04 inch (Hg) leak (DTC P1456) detection procedures have been modified since the ‘98 model year, and the same detection method is used in the first drive cycle and the second drive cycle.

• Three steps are performed for leak detection during a valid drive cycle (DTC P1456).

Step 1: Start the engine from the specified engine coolant and intake air temperature. Compare the output value from the fuel tank pressure sensor right after start-up with the one after 20 seconds have elapsed. If there is a change between them, the tank side is (OK) considered free of leaks. Otherwise, go to Step 2.

Step 2: Monitor the changes of absolute pressure in the fuel tank for a specified time period after starting the engine. If it changes by a specified value toward negative pressure, or the difference between the maximum and minimum absolute pressure in the fuel tank is large, it is interpreted as no leakage in the tank side and the detection is complete. Otherwise, go to the next Step 3.

Step 3: If the difference between the absolute pressures in the fuel tank right after the specified time period has elapsed and after a while is small, it is interpreted as leakage in the “Tank Side”. If it is in the first drive cycle, a temporary DTC is stored. If it is in the second drive cycle, a DTC is stored, the MIL illuminates, and the detection is complete.


Malfunction Threshold (MIL illuminates); All conditions, below, must be met.

Step 1: The change of fuel tank pressure is 0.2 kPa (1 mmHg, 0.07 in.Hg) or less.

Step 2: The change of fuel tank pressure is 0.2 kPa (1 mmHg, 0.07 in.Hg) or less. Or the difference between maximum and minimum value is 0.7 kPa (5 mmHg, 0.2 in.Hg) - 3 kPa (22 mmHg, 0.8 in.Hg).***

Step 3: The change of fuel tank pressure is 0.8 kPa (6 mmHg, 0.3 in.Hg) - 0.9 kPa (6 mmHg, 0.2 in.Hg).****
*** : Depending on initial engine coolant temperature and elapsed time period.
**** : Depending on initial engine coolant temperature.


Driving Pattern

1. Start the engine at an engine coolant temperature and intake air temperature as specified under Enable Conditions, and let it idle until the radiator fan comes on.

2. Drive the vehicle immediately at a speed between 25 - 75 mph at least 6 miles (10 km).

• If you have difficulty duplicating the DTC, retest after turning off electrical components such as the audio system and A/C, and try a different gear position.

• Drive the vehicle in this manner only if the traffic regulations and ambient conditions allow.

strum456

I found the above post on another thread on here. I am trying to decipher all of it. It looks like the system builds pressure in step 1. Then Vacuum for step 2. Then pressure again for step 3.

I checked for leaks, like you have suggested before I posted on here. With the canister vent shut, the canister side holds vac and the FTP remained steady. Then I opened the bypass (connected to the 2-way valve) and the FTP reading dropped (the tank was under vacuum). The whole system held vacuum for a while until I unscrewed the gas cap, and the FTP reading equalized at 2.57v.

The only discrepancy I have noticed is that sometimes when I pull vacuum on the tank, I will notice the FTP voltage increasing about .01 every 6-10 seconds. I thought this was a leak at first, but then last night I noticed that the voltage continued to increase beyond 2.57. It kept climbing up to 2.8 or 2.9v. Note: the car was not running. I believe what I'm seeing is just the gas evaporating and causing pressure to build.

We'll see what happens with the new ECM too. So far, I don't have any codes pending.

el crapitan

keep us updated, please
thanks