Doubtful PCM, never heard of that as a cause with this code. I would concentrate under the hood where the squirrel was chewing. I bet your leak is in that area or a missed chewed wire. Thats where these lines are also, by the battery check them and the wiring, I wouldn't clamp anything off. Good luck.

 

Hey all! I have had a P0443 (Purge valve solenoid circuit error) code for about a year now.

I have done everything here, replaced charcoal cannister, the actual purge solenoid (twice, once with off-brand and recently with OEM), replaced the gas cap, and replaced the leak detection pump.

I checked the connection into the purge solenoid and it clicks on and off when I send power to it and I checked the wiring going into it to make sure there was power and there was.

Yesterday I did a smoke test running the smoke into the purge solenoid tube and back into the system.

I clamped off the hose going from the charcoal cannister to the fuel tank (maybe shouldn't have?)

The only place I got smoke was the vapor intake filter under the gas cap area inside of the tail light and as I understand it, that is normal to get smoke there. Maybe I should clamp that off, but otherwise I didn't get smoke anywhere else.

I did have Meineke check it out since I had it there for an oil change and they think the PCM is the next thing to replace, but I am not sure I buy it.

I did have a wiring harness issue (the one by the battery) where a squirrel ate it up not long ago, but that was fixed. Could there be another electrical issue I need to check besides the power going into the solenoid?

So questions...

1. Did I clamp off something I shouldn't have during the smoke test?
2. Should I clamp off the line to that vapor intake filter under the gas tank intake and smoke it again?
3. Is there another electrical connection I should test like maybe the one to the leak detection pump?
4. What are the chances it is the PCM and is there a better way to test and be SURE it is?
5. Any other ideas before I take it to the dealership with my head hung low in defeat?

Thanks!

Like Deepblu mentioned, check the wiring repair, a poor ground, or positive lead into the Evap sensor could be causing this.

Also, IIRC there shouldn't be any type of leak coming from the gas cap area, all fuel vapors should be filtered via the charcoal canister.

 

Thank you for these responses! @deepbluerubi & @dgsimick

So, I just asked JustAnswer this and according to the mechanic on there (who was not extremely helpful as he just pointed me to a doc) it says that 18/25 of mechanics reported that it was the ECM OR that by replacing it, the issue was fixed. (Wonder if a new wiring harness comes with the new ECM?) Somehow, I bet there was and since the squirrel chewing was on the other side of the wiring harness, maybe that is the case?

I hooked up my multimeter to the wires coming into the purge valve and it read a steady 13 or so with no fluctuations. Not sure how often it should be triggering?

But I am not sure what that means that it was a constant 13? The ECM sends a ground signal to open the valve I believe. Would it only read 13 grounded and therefore the purge valve is always open?

I just don't know enough about electronics, so I am usually guessing my way through life when I use a multimeter.

@dgsimick Are you 100% sure on the filter? I have done some research on this, and a few different places folks say it is normal for smoke to come out of the filter. There is a little filter on the inside and under the gas intake. It is a vapor intake filter as I understand it:

Here is the thread:

EVAP Leak (Fuel Vapor Canister FILTER???)

I have a 2015 Jeep Wrangler with almost 90K miles. I've been troubleshooting an EVAP leak that started as a P0456 and then became a P0457 with the characteristic GASCAP message. The dealership updated the software, which didn't fix my P0456 problem and then told me I'd be waiting at least a...

www.wranglerforum.com

From KiwiScott on that thread, but I confirmed it on another YouTube video as well.

"Yes, smoke from the filter below the filler is normal, just not anywhere else. Usually they leak at the top of the tank or the actual canister, or the solenoid valve fails. All things you've already sorted so all I can think is it's not being actuated correctly? I'm not sure how you'd test it."

Thanks for the comments and help!

Either way, I am going to double check the wiring that was repaired.

Thanks!

 

Ok this might be simplified. It looks like there is nothing between the ECM and the purge valve as far as other connectors, wiring harneeses or fuses go. From what I can tell there are two wires that are directly connected to the purge solenoid from the ECM. Which would make this quite a bit more simple as the squirrel chewing area isn't really affected by this.

And if that is the case, it probably is the ECM.

 

What about the tubing to the valve, maybe he gnawed a bit of it and you have a leak there?

 

@deepbluerubi @dgsimick Ok Maybe I fixed it? I unplugged the harness that is plugged into the ECM and has the wires that connect to the purge valve. Then I inspected it, moved some wires around and plugged it back in and the check engine light is gone. LOL! I am still getting the code though, so I guess I just have to wait til it clears? But the light is off. Could it be that simple?!?!

 

Time will tell, if it doesn't come back on in one hundred miles or so it might be fixed. I would be feeling around the tubing for a nick/ bite mark. Either way good luck.

 

Yeah the positive cable did look a little ragged almost like it was nicked, but that doesn't make much sense as I am getting power via that cable to the valve. Sure feels good to have the needle move even just a little bit and even for just a moment though.

 

You're on the right track. One wire should read voltage when measured at the canister from the battery neg post. The other shouldn't unless grounded by the ECM. I.e if there's any other ground on the second wire she'll be powered up full time.

If you have wire you could back probe the pcm connector and shunt it to the canister as a proof. Test both wires from battery neg at the pcm too. One should be live, the other should have no continuity to ground unless pcm activated.

Cheers,
🥝 Scott

 

@KiwiScott Ok, so question about that as I am not totally sure of something. If I hook up my multimeter to the positive and negative cables that go into the Solenoid from the ECM, what should it be reading after I start the engine? Right now, it reads consistently at around 13 volts which a little fluctuation from time to time down to 12.5 or something. That's it. I have left the car running for quite a while thinking that it would fluctuate at least at some point when the ECM triggers it. I have also felt the solenoid at startup and while the car is running and don't feel any clicks (but should I?)

One mechanic I asked said it is super random as if I could leave it running a long time and it wouldn't trigger. I figured once the car was warmed up, it should at least open once.

I did test the solenoid again by just hooking it up to the battery and it definitely clicks so it looks to be working.

I am beginning to wonder if it is the ECM or if it is something else that is causing the ECM not to trigger it. I ordered an affordable iPhone based bi-directional scanner that will be here tomorrow that will allow me to send the signal for the ECM to open it or close it, so I think that will confirm whether or not it is the ECM.

I talked to an ECM repair/refurbish company and they said that this issue is almost never the ECM, which again is weird since that other mechanic suggested that 18 out of 25 times mechanics fixed this issue by replacing the ECM.

Oh, also to make sure there wasn't an issue in the wires between the ECM and the solenoid, I also just cut the cables near the ECM and re-rerouted them with some test cables and the same thing was happening, so I ruled that out. Unless it could be the pin connection itself, but the pins look ok too.

Thanks for your help!

 

@KiwiScott Ok, so question about that as I am not totally sure of something. If I hook up my multimeter to the positive and negative cables that go into the Solenoid from the ECM, what should it be reading after I start the engine? Right now, it reads consistently at around 13 volts which a little fluctuation from time to time down to 12.5 or something. That's it. I have left the car running for quite a while thinking that it would fluctuate at least at some point when the ECM triggers it. I have also felt the solenoid at startup and while the car is running and don't feel any clicks (but should I?)

One mechanic I asked said it is super random as if I could leave it running a long time and it wouldn't trigger. I figured once the car was warmed up, it should at least open once.

I did test the solenoid again by just hooking it up to the battery and it definitely clicks so it looks to be working.

I am beginning to wonder if it is the ECM or if it is something else that is causing the ECM not to trigger it. I ordered an affordable iPhone based bi-directional scanner that will be here tomorrow that will allow me to send the signal for the ECM to open it or close it, so I think that will confirm whether or not it is the ECM.

I talked to an ECM repair/refurbish company and they said that this issue is almost never the ECM, which again is weird since that other mechanic suggested that 18 out of 25 times mechanics fixed this issue by replacing the ECM.

Oh, also to make sure there wasn't an issue in the wires between the ECM and the solenoid, I also just cut the cables near the ECM and re-rerouted them with some test cables and the same thing was happening, so I ruled that out. Unless it could be the pin connection itself, but the pins look ok too.

Thanks for your help!

If you have cut the cables from the pcm to the solenoid and replaced them both rather than back probing and continuity/ground testing, and you have tested the solenoid then yes, you have proven the pcm driver fet fault.

My method suggestions were to check the pcm switched low side was infact being switched after identifying it by testing from battery negative while the solenoid was disconnected. One wire at the pcm should be live. The other wire should have switched continuity to ground. I'm not completely sure what the switching criteria are but I imagine it would be based on sufficient intake vacuum to maintain evac V's solenoid closed at wot (@Digger84 will likely know more of the Nitty gritty on that). It's not recommended to test solenoids at full battery voltage without a current limiting device (simple low wattage 12v light bulb will do) in series (in one of the test wires). However with this one I think your probably fine if you only did momentary test to see if it clicked.

I've never actually heard of a pcm based evac solenoid actuation fault such as this one. I'm thinking the live must've shorted to ground at some point and blown a pcm fet. Usually that only happens to injector driver fets. I guess at the end of the day they're both solenoids being low side switched (usually) so it's probably just luck on my part that I haven't come across it yet myself. It sounds like your mechanic is correct.

The good news is that fets are cheap to have replaced. That's if your new tool confirms it's not switching. If it is actually switching and not able to maintain an evac vucuum then you still have a vacuum leak somewhere causing your fault.

Another possibility would've been a pcm ground disconnected, but if you're getting battery voltage on both new wires at the solenoid from battery negative (literally at the battery) then the pcm is definitely grounded.

I hope this makes sense, kinda thinking out loud?

Cheers,
🥝 Scott

 

@KiwiScott Yep that is what I have concluded. I was able to buy a super-cheap bi-directional scanner off of Amazon and command the solenoid to open using it and it wouldn't open.
So I am hoping it is the ground connection. I hooked up the positive of the solenoid to the PCM and the ground of the solenoid to the battery negative terminal and commanded it to open using the scan tool.

When I did that it opened. So, I am hoping it is the ground connection in the ECM harness. Someone else reported this and they opened the harness and slightly squeezed the female pin for the ground connector and that solved the problem.

Going to try that today and I will report back. If that doesn't work, going to have to have the ECM replaced or someone repair it I think.

Thanks for the responses!

Craig

 

@KiwiScott So, an update. I thought I had this sorted (or hacked.) I just hooked the negative cable going into the solenoid up to the ground terminal on the side of the ECM and was going to call it a day. Check engine light finally went off. WOO HOO! The Eco light came on which I haven't seen in aeons.

When I tried the bi-directional scan tool to command the actuator to open it worked. So, I thought all was good!

I drove it to wal-mart and then on the way home, the light came on again. NOOOOOOO.

I think possibly that now, the solenoid is open all the time instead of closed all the time.

Ugh... so I am still down to the receptor on the harness being bad or the ECM being bad.

Checking with an ECM repair shop to see what they think.

C

 

Kinda sounds like it's still a connection fault.
Fets don't usually die intermittently. Also my understanding (I must be wrong) is that the low side, so the ground side, is what's being switched by the pcm, with the highside staying live. If that's correct, then when you made a ground side to battery neg connection that would've made the solenoid live, bypassing the pcm. You wouldn't have been able to actuate the solenoid via the pcm with your tool. Since you did, and it worked, that means it's the high side being switched?
For it to respond to your tool at all the fets must be okay. So you either have an ongoing connection issue, or the pcm is not reliably getting the trigger it requires to fire the solenoid?
Sorry I can't help more, I need to learn more about what parameters it's switched and what might interfere with that.

Have a look at this:

I seems these are pwm solenoids so the voltage read on a DC multimeter should vary rather than simply switch.

Cheers,
KiwiScott

 

I thought about what you were saying about activating it with the ground connected and here is my theory. My scan tool has 0% to 50% settings to actuate it. I am thinking that the solenoid was constantly open (because of the direct connect to ground) and when I was "activating" it I was actually just changing the signal each time. I would assume that the ground "activates" it, but the amount of voltage sent via the positive wire would control the amount the solenoid is open.

So, when I was testing, the test wasn't turning it on or off, just modifying how much signal it was getting. Do you think that makes sense?

Also, can you let me know what a Fet is?

Thanks for the video. Learned some from it, but some of it is over my pay grade. Dang I wish I understood electronic testing better.

Thanks for your help!!!

 

Yes that makes sense. I'm not sure though. It would be unusual.

Mosfets (fets) are like a solid state switch, the can make or brake a circuit based on a signal they receive. There are n and p channel mosfets for use in high or low side switching. They can switch fairly high voltages and currents extremely fast. Their signal is provided by a 'driver' circuit pulse width modulated.

So it seems I was wrong assuming they work in essentially the same way as an injector circuit. I have some learning to do. I hope you manage to find the cause of your fault, and I hope it's just a connection, signal, or wire rather than the pcm.

Cheers,
KiwiScott

 

Any luck ?

 

What was the fix