Wednesday, March 30, 2011

2003 Chevrolet Impala

I always get the same comments from people when I tell them what I do. The conversation usually goes one of two ways. One is "Oh yeah I have this problem with my car.....what do you think?" When I explain that I only diagnose for the professional shop not the car owner the conversation either ends or the next line is "Well, don't these shops know what they are doing?" followed by "They just hook up the machine and it tells them what to change." This is the publics perception of how "easy" it is to diagnose cars today. When I tell people the shops I do work for in fact know what they are doing and it is not always that easy to diagnose today's cars it usually falls on deaf ears. Respect has never been part of the automotive repair business unfortunately. This case study is a prime example it is not always that easy as hook up scanner, retrieve code, change what it says on the scanner, and collect money.
So I get a call from Mr. D. He tells me he has an Impala with an evap code. He changed a part and it still has issues. Mr. D hates evap codes. So I get there and pull a P0446 code (Evap vent system performance). I look at two items when addressing trouble codes. One is description and code set criteria. Here is the description.

SYSTEM DESCRIPTION
This DTC tests the evaporative emission (EVAP) system for a restricted or blocked EVAP vent path. The control module commands the EVAP canister purge solenoid Open and the EVAP canister vent solenoid Closed. This allows vacuum to be applied to the EVAP system. Once a calibrated vacuum level has been reached, the control module commands the EVAP canister purge solenoid Closed and the EVAP canister vent solenoid Open. The control module monitors the fuel tank pressure (FTP) sensor for a decrease in vacuum. If the vacuum does not decrease to near 0 inches H2O in a calibrated time, this DTC sets.
Now the code set.

CONDITIONS FOR SETTING THE DTC
  • The Fuel Tank Pressure sensor is less than -10 inches H2O.
  • The condition is present for as long as 30 seconds
 To put this in plain english. The evaporative system sees something other than 0" vacuum when it should be 0". How it sees this is the fuel tank pressure sensor. See below. Remember, click to enlarge pictures.




The normal voltage I see for a fuel tank pressure sensor on a GM vehicles at sea level is 1.49-1.52v with the gas cap off and during a leak test the voltage will rise to about 2.5 volts. This is the opposite of most manufacturers. So I am really interested in what our fuel tank pressure sensor voltage is on this vehicle.
That is right-0.22v on the scanner. This reading is saying this evap system is under pressure! So, my next step is to unscrew the gas cap and recheck. I do and the voltage stays the same. A new vent solenoid has already been installed by Mr. D.
At this point it is time to backup my scanner readings with DVOM (Digital Volt Ohm Meter). As much as I trust scan data whenever possible I like to see raw readings.

So I hook up my Dvom and.....
Dvom agrees with scanner. My next thought is what if the sensor signal wire is partially shorted to ground causing the low voltage. That is easy enough to check. We will jumper the 5vref to the sensor signal and see if the voltage at the scanner changes. Lets first see if we have a good 5vref.
No problem there. Lets jumper.

Now lets recheck our fuel tank pressure pid on our scanner.

A little fuzzy but it is 5.00 volts. So now we know we don't have a wiring issue. What could we have? Given the fact that the voltage didn't change when the gas cap was removed my next suspicion is the actual fuel tank pressure sensor itself. GM is kind enough to give us an access panel. So off it went.
Lets look inside. A bit grungy in there, but no chafing issues which can happen.

At this point I am pretty confident a new fuel tank pressure sensor will solve this issue. So a new fuel tank pressure sensor is popped in.

Now to recheck scan data. Remember, what we should be at.....

It is real fuzzy but it says 1.49v. I run an evap service bay test on my scanner and everything passes. At this point I guess you are wondering. Why didn't it set a fuel tank pressure sensor code? Good question here is why. Look at the code set criteria for P0452 (fuel tank pressure low).
CONDITIONS FOR SETTING THE DTC
  • The FTP sensor voltage is less than 0.1 volt .
  • All conditions are present for more than 5 seconds .
  Were we meeting the code set criteria? Nope. Almost. This is a great example of how important code set criteria is. Most techs love to look at the tree. I like to look at the code set criteria. Remember, the scanner doesn't tell you what part to change, proper diagnosis is still needed. 

9 comments:

  1. Great fault shooting!! You got what it takes to fix any car.Thank you for telling the world that you are a talented automotive tech and also for sharing your career experience.tell you what, your troubleshooting pattern is similar to ScannerDanner's channel in UTube Do u know each other?

    ReplyDelete
  2. Thank you for the kind words. I do not know ScannerDanner. I will have to search for him on YouTube.

    ReplyDelete
  3. I'm assuming this is something you do with the car running? Just making sure because I have the same fault code on my 02 Camaro. Thanks in advance.

    ReplyDelete
  4. This is awesome! I'm assuming you do these tests with the vehicle running? I have an 02 Camaro with this code and the pressure sensor is the only thing I haven' checked. Thanks!

    ReplyDelete
  5. Sorry I didn't think the first one went through.

    ReplyDelete
  6. No, these were all done key on engine off.

    ReplyDelete
  7. Ok. I was getting 4.8 vdc from the signal wire with gas cap on and off. I went ahead and put a new sensor on it and it's reading 1.48 vdc cap on and off.

    ReplyDelete
  8. John I have a 2006 impala and the gas gauge is not working properly...Could this be the issue also. I put in $30 and says I have less than a 1/4 of a tank...

    ReplyDelete
  9. If the signal wire would it been short to ground and when you jump the 5v to the signal wire wouldn't you burnt your pcm/5v reference

    ReplyDelete