I've got:

P0300 - Engine random misfire
P0446 - EVAP performance
P1441 - EVAP flow during non-purge

I think the EVAP codes are the relay being grounded constantly...I'm not sure though.

Are these related in any way? I've long suspected my MAF is going on the car, but no hard evidence yet.
It flashed the CEL when I took off hard from a light today. It was a 5 lane road and I had to dump the throttle on this ricer Honda next to me... It stumbled, flashed, and then was alright. I still managed to leapfrog the Honda for 100ft before letting off, because street racing is really dumb. It was around 4k (IIRC) when it flashed the CEL.

Well, this is strange. P0446 means it can't purge to the EVAP canister, while P1441 is the opposite. Hmm.

P0300 is a random misfire, and may be an indirect result of this. A failing MAF would set the check engine light, perhaps P0300, but I think your EVAP system has an issue — a grounded relay would definitely do that.

Replace that questionable relay, clear the codes, and see what happens.

Alright. Where is the relay for the EVAP solenoid?

The Haynes manual says it'* relay #5 in the relay center, which is located above the passenger floorwell. You'll have to take the black cover off, which is simple.

Of course, finding #5 may be tricky, as I don't think they're numbered. (There should be a diagram on the back of the cover.)

Switch that relay with any other (e.g., your fuel pump relay). If some other system fails (e.g., your engine sputters and dies because your fuel pump isn't pumping), then you have a bad relay. I picked up a couple during my last junkyard run.

Alright. Thanks. I'll do what you said and I'll keep you posted

Why do you guys think a relay is grounded?

Both 1441 and 0446 point to checking the lines and the canister. Knowing how a 97 has a purge valve and purge solenoid separately, we know that the lines to and from then can be an issue. As well as the lines under the airbox at the canister...and the u bend that goes into the throttle body.

I'd suggest checking your vacuum lines first before the electrical. The vacuum lines are a much more common problem area.

I agree, check for vacuum leaks. Spraying them with carb cleaner really don't work in this case. You need to actually look at them. The 300 code is nothing to worry about, it'* a misfire because of the vacuum leak.

Update bump...

I had 2 fittings that were loose\worn\cracked etc... Replaced...

I also noticed some odd things... Trials and tribulations, if you care to read, follow...
I appreciate your comments afterward.

The vent solenoid is always open unless powered, in which case it is shut. I think this is done so the canister can vent when the car is not running without using any battery power.

The purge solenoid is the opposite. It is shut unless powered, in which case it is open. I think this might be done for the same reason. If it operated the same as the vent solenoid, there would be air leaking from the canister vent into the throttle body.

I also noticed that, for both solenoids, voltage from each connector was low. With the car running and 14.8v at the battery, I had 9v-11v at the connectors. I also noticed that the solenoids were not moving in response to power.

So, to test them, I removed both from the car and wired them directly to a momentary switch and a spare car battery. They were both partially stuck open\close. After cycling them a few times, they both freed up and performed perfectly.

I reinstalled them in the car only to find once again, they did not operate when power was applied. I tried wiring a solenoid to the positive side of the connector and then grounding to the chassis. This worked. I measured voltage from connector to chassis. This read only 0.3v less than battery voltage on both connectors.

I assumed that whatever ground the solenoids shared was damaged in some way. Therefore, I came to the conclusion that I had to run new grounds for both solenoids. To keep it simple, I followed both solenoid grounds in the harness until they met, and then ran a new 12ga wire to my shock tower (the same location of chassis ground, much better than the stock fender).

Now, when power is applied to the solenoids, they operate correctly.

When the relay activates, it should remove power which ever one it operates, and should work correctly. I'm assuming the correct operation of the relay is "normally-closed."

Now with no vacuum leaks and no EVAP codes, the motor is stronger than ever. For a test run I took it out on a deserted road and it climbed up to 100 MPH effortlessly.

I am very pleased.

I tossed the same response to the thread at BPro for you.

Due to the way the system works, your post is a bit confusing. The solenoids get ignition power all the time, the pcm then grounds or removes ground to them to actuate them in the desired way.

Below is a 98 diagram as I don't have 97 available. Difference being 97 has no tank pressure sensor and a second item. Technically 97 has a purge valve and purge switch. The diagram shows us the ground controlling pcm.


Next thing to metion is that the strut tower isn't technically better than the fender ground. The battery supplies ground to the chassis via two connections. The main connection is a heavy gauge battery cable that leaves the battery and goes directly to the engine block between the AC and starter. The other cable is thinner gauge and connects to the fender behind the battery. Our engine/pcm get their grounds from a connection at the back of the engine block by the oil filter adapter.

I would think by grounding them all the time, you have closed them and removed control from the PCM, which once it see'* and runs the needed tests will give you a code for them not functioning properly.

Good show, Bill!

Nothing like a some light electrical engineering over my thousand cups of morning coffee.

Because you seemed to get a better ground when you bypassed your vehicle'* grounding system, I'd check your negative battery cable, the main ground Bill mentioned, and perhaps the solenoids' connectors themselves. You might find some corrosion along the way.