I have searched the forums and have not found a similar problem described so I will throw this out in hopes of some ideas.

My gently used (79K miles) '96 F150 has developed a problem which seems to revolve around the evap canister purge valve error code 0443 and the "Check Engine" light. I have done some troubleshooting and replaced the solenoid valve but the problem persists:

Once the truck is warm after driving for 5-10 miles, when coming to a stop, the engine will begin to idle rough getting worse until it will often die unless I pump the accelerator pedal. It will abruptly "clear" and seem to run OK for a while. If I turn the ignition off and restart, it will also clear up for a while. If I reset the code before starting out and put my reader in PENDING mode, I get a 0443 Pending when it begins to run rough and after several more miles, a "Check Engine" light and the normal fault code.

This problem has become more pronounced in our latest round of warmer weather and, yesterday, occurred while driving (35-45 MPH) in addition to the at-rest scenario described above. I had to pull off the road where I tried restarting. It did restart but would not idle without pumping the throttle. It suddenly cleared and I drove home with no further incidents.

Searching the archives shows a lot of knowledge and talent in this group. Hopefully, someone will have a suggestion on what to try next.

TIA

 

Is that P0443? '96-up codes have 5 digits, including a letter in the 1st position. Look up the definition on a website like this one:
http://www.troublecodes.net/OBD2/
Don't pick the name of a part out of the definition; read & post the COMPLETE definition. Look it up on several lists to read it worded a few ways, and it'll probably help you understand what it really means.

With the CANP unplugged & the key in RUN, the R wire of the CANP connector should show battery voltage. With the key OFF, CANP disconnected, and PCM disconnected:
- the CANP should have 30-38 Ohms resistance
- the CANP circuit should have less than 5 Ohms resistance from the PCM connector to the CANP connector
- there should be more than 10K Ohms resistance between any pair of these circuits: CANP (LG/Bk), BAT (battery +), & GND (battery -); there should be more than 10K Ohms between VPWR (R) and CANP or BAT.

 

My apologies for not noticing the protocol for posting codes. You are correct that it is P0443 "EVAP Control System Canister Purge Valve Circuit Malfunction" I haven't dug this deep into a vehicle since I was in high school and my '56 Chevy was only 3 years old so please bear with me....I'm trying to get some late smarts.

After performing all your suggested resistance and voltage checks everything checks OK EXCEPT for the connection to the PCM. The wiring diagram (unknown origin) I have shows it connecting to pin 67 but there is no pin 67 on my connector not even a hole for the wire. I ohmed out between CANP and all pins on the PCM connector and found no connection.

In looking at the various diagrams and drawings I have found here and elsewhere, all show a simple solenoid valve between the canister and the throttle body. However, my valve also has a white hard plastic vacuum line coming out of the bottom and running to somewhere I haven't searched as yet. Does this have a bearing on how or from where the valve is actuated?

Thanks for the help so far.

 

Pin 56 (191 LG/Bk) should have less than 5 Ohms to the same wire at the CANP connector. If it doesn't look exactly like the pic in my post above, you're not looking at the CANP.

 

There is nothing like that that I can see under my hood. This is what I have (3 photos). It agrees with the sticker for the vacuum system. The valve is located in the upper left corner of the first photo.







Or am I really getting blind in my old age?

 

Steve, reading that sticker indicates a "medium duty" vehicle (I thought a F-150 was a light-duty?) with California emissions. Could this be where the discrepancy is between what he sees and what your data indicates?

 

It is possible, I guess. I honestly don't know enough about emissions systems to even know where to look for an answer. Those idiots in Sacramento are always screwing up something good

 

That's not a CANP - that's a VMV, as your VECI label indicates. They were only used from '96-up, so I never think about them. AFAIK, all the voltage & resistance values are the same.

.

CA could have had different criteria for light vs. medium duty. This is a non-CA '96 4.9L F150 VECI label:

 

OK, it is a VMV and we are in the same chapter if not on the same page. And there is continuity (<1 ohm) between its oneector and pin 56 of the PCM so everything external to the PCM checks OK.

I connected my voltmeter to the switched side (grn/bk wire) of the connector at the VMV through an insulation piercing clip. With key on, it read battery voltage as expected. Here is where the results don't make sense to me. If I start the engine, the meter shows +13.7V so I assume that pin 56 of the VMV has not switched to ground. If I run the engine for a while the voltage will drop to about 11 V and then slowly decrease to about 9 V over a period of 3-4 minutes and then abruptly return to 13.7 V. I repeated this through three cycles. It seems the PCM is trying to ground the solenoid but never quite doing so. With the readings I am getting, the coil is only getting about 4.5 V to 5 V across it. That may be enough to make it work with the vacuum control port but I would expect to see < 1 volt on that side of the solenoid using a solid state device.

I was never able to generate a code but am wondering if it might need actual driving conditions to generate that.

Anyone have any thoughts? I'm dreading that it might be that particular control circuit in the PCM...ugh.

 

The PCM doesn't "ground" the VMV wire; it pulses it to ground, just like the injectors, IAC, EVR, and most others. The pulse-width changes to create varying output on whatever device is being controlled. Most handheld meters don't update fast enough to display this pulsing, so it can create a unique result on each meter. Try using a 12V LED (noid) test light - they don't draw enough current to affect the circuit, and an LED can respond much faster than an incandescent bulb.

 

Or find a friend with an oscope. I'd let you use mine, but yer on the wrong side of the country.

Adrianspeeder

 

I sort of suspected that might be what is happening. The descriptions I have simply say it gets grounded. I have a good oscilloscope in my electronics lab so will bring it home this weekend and take a look to confirm. However, if everything checks out OK, why the P0443 and check engine light?

 

If the PCM is setting a code for a nonexistent condition...