Searching on Hondatech has been fruitless so far, so here's the basic outline of the situation. A friend of mine swapped a D16Y8 (OBD 2 vtec civic ex motor) into his 96 Civic LX sedan. It's a daily driver and he wants to keep full OBD 2 functionality/legality if at all possible even though a lot of people just give up and go OBD 1. He installed the P2P ECU, which is from I believe a 96-98 5 speed ex coupe with the Y8 motor, and has successfully taken care of the VTEC wiring and IACV wiring that most people run into (no other codes or problems).

The problem is, after a few hundred miles it's throwing a P0453 code --> high voltage/open circuit on fuel tank pressure sensor. I looked through the actual Honda service manual and it turns out that my friends 96 LX sedan is one of the only models that did not come with this sensor, while the vehicle this ECU came from obviously did. I'm guessing that installing the sensor would not work because he would probably have to swap the gas tank and/or fuel lines. Finding what I presume would be the correct ECU could be very difficult considering how specific of a model it would have to be from, a 96 Ex manual transmission sedan.

Looking at the output diagram of the fuel tank pressure sensor:



It says in the service manual that, with the gas cap open, the output voltage should be 2.5 V, which would be atmospheric pressure (0 relative pressure). So I'm guessing the normal operating range with everything working right is 3 - 4.5 volts, with the tank being under pressure? Does a fuel tank ever draw vacuum under normal operation?

After adding a pin to that slot in the wiring harness, could I tap it into a 5 volt reference wire and then wire a resistor in-line? What resistor would I need to order (what ohm rating) if I want it to read around 4 volts continuously? If it sees straight 5 volts I'm sure it will still throw a code. If it sees 2.5 volts it will probably think the gas cap is open and eventually throw a code. In the service manual it also says that "When fuel vapor pressure in the fuel tank is higher than the set value of the EVAP two way vale, the valve opens..." So I'm guessing if this value isn't in a proper range, it will yet another evap code. I just don't want to solve one DTC and create another.

I'd appreciate any input at all. I know this is a really obscure question but there don't seem to be many Honda people who attempt this exact swap and I can't find information anywhere.

[Edited on March 16, 2009 at 1:09 PM. Reason : voltages]

3/16/2009 1:06:14 PM

Quote :

"What resistor would I need to order (what ohm rating) if I want it to read around 4 volts continuously?"

V=IR?

3/16/2009 1:18:20 PM

I'd start with some kind of variable voltage power supply and see what voltage actually works first.

3/16/2009 1:21:25 PM

Ray, go to Majestic Honda's website and see what and where that sensor is. Jeremy is very familiar with their site, and they've got complete diagrams covering this shit. Putting a sensor inline isn't going to be hard, given the right hardware. Might be the best thing to do.

Being that the ECU's inputs are most likely of high impedance design, putting a resistor inline simply won't work. You'll end up with a full 5 reference volts regardless. You need to use a voltage divider circuit to actually vary what the ECU sees. Very simple, but to be completely variable, gets a little more complicated.

Another thing you might run into is this: if the ECU reads differential between Key On/Engine Stopped and Key On/Engine running or something like that, then it might still flag a malfunction if it sees the set voltage immediately when the key is switched on. I vote for robbing a sensor off a parts car or buying one and figuring out how to incorporate it.

Another thing: you said performance was lackluster, especially when compared to my K24A4, which is only rated at 160 hp. Are you guys SURE that the VTEC solenoid is energizing? Is VTEC oil galley pressure where it needs to be?

3/16/2009 1:30:58 PM

So the VTEC is kicking in? I think that's all that really matters.

3/16/2009 2:29:19 PM

^i dont think it is.

^^i would def look at that.

3/16/2009 3:25:29 PM

^ i said the performance was lackluster probably because I daily drive a V8 and I have trouble feeling a difference between like 80whp and 100whp. I think it is hitting VTEC. I don't drive it every day, and my friend says the vtec is definitely working. The only time I've ever been able to feel vtec was on a vtec d series turbo car.

my friend looked into it a little more, and it turns out that the ECU is dumber than we thought. It will accept a straight 2.5V signal, which is weird because that's the signal when the gas cap is open... So he is going to use a potentiometer hooked to a 5 volt reference voltage and just set it so it outputs 2.5 V. problem solved.

[Edited on March 16, 2009 at 3:30 PM. Reason : vtak yo]

3/16/2009 3:29:13 PM

you can't hear the VTEC changeover?

http://www.youtube.com/watch?v=CeQ5inIaPSw

3/16/2009 3:44:58 PM

yeah, i was going to say i see no reason why a set input value within the "happy" range of the ecu wouldn't be just fine. while under normal/real world operating this sensor might be sending constantly changing/variable values to the computer, what the computer actually needs to not throw a code could be any set value as long as its within that range of 2.5v to whatever. basically, it'd be just like the iat resistor mods some people do.

3/16/2009 5:25:28 PM

^^

DOHC sounds and SOHC sounds are very different.

3/16/2009 7:25:53 PM

Try the pot and see what happens. If it doesn't work, let me know, and we can rig up a simple voltage divider.

3/16/2009 7:30:44 PM

hell, screw a pot even... just do the math and go to radio shack and get the right resistor for a few pennies.

[Edited on March 16, 2009 at 7:56 PM. Reason : you've got the fsm, right?? should have the numbers you need and you already know what the ecu wants]

3/16/2009 7:56:00 PM