Working as a mobile tech for the last 20 years, I have had a lot of strange but exciting repairs. Using the word exciting might be a little strange to some of you, but along with this job comes a lot of challenges. The challenge to be able to quickly and efficiently solve the problem is what makes it exciting. It’s also exciting that when a job is a tough one it usually is a job that I learn from. In a lot of ways, these jobs are how we as techs get our training. You really get to know the system and how it works when you spend a good amount of time analyzing, reading and testing the circuit or circuits at hand. Fortunately, making those sacrifices and spending extra time on jobs of this type is what will enhance your skills and knowledge for the next tough job.
| Vehicle Details |
| Year: 2003 |
| Make: Chevrolet |
| Model: Impala LS |
| Mileage: 120,639 |
| Engine: 3.8L V6 |
| Transmission: Automatic |
| Mileage: 129,025 |
| Customer Complaint: Will not pass state inspection |
Although the check engine light was not on, the customer’s car still failed because not all of the monitors had completed. Monitors are tests that the Engine Control Module (ECM) runs on the system and/or components it controls. The state-run inspection program, a software program that looks at all the monitors in the vehicle's ECM, checks to see what monitors have run to completion and passed. In New York State, a vehicle older than model year 2000 can have up to two monitors not completed and still pass the emission test. A vehicle newer than model year 2000 can only have one monitor not completed.
This Impala came in with the check engine light off, but with codes P0135 and P0141 stored in the ECM memory. The description for these codes is heated oxygen sensor (O2) performance bank 1 and bank 2, respectively. Along with the codes that were stored, the monitors for the heated oxygen sensor, the catalyst monitor and the evaporative emission monitor were not complete. Most of the time these codes are set due to the fact the O2 sensors’ internal heater isn’t working correctly. The shop replaced both of the sensors twice thinking they received faulty parts. After reviewing the freeze frame data set by these codes, I didn’t notice anything that stood out to alert me of a problem.
The next step was to grab a wiring diagram and look at the circuits of these two O2 sensors. These sensors each have a four-terminal connector. Terminal A is the sensor low signal circuit, terminal B is the sensor high signal circuit, terminal C is the ground circuit and terminal D is the power feed circuit. Looking at the sensor schematic, I first checked the fuse that powers this circuit. The fuse was actually labeled oxygen fuse. I hooked up my meter to the negative battery terminal and checked both sides of the fuse. I didn’t use a test light as I personally like to see the voltage being read. Remember, it doesn’t take a lot of voltage to light a test light. With today’s computer-controlled vehicles, using a Digital Volt Ohmmeter (DVOM) is much more accurate. So at the fuse I had 12 volts on the meter. Next I unplugged the O2 sensor connector and put my meter on terminal D, which is the 12-volt feed from the oxygen fuse. I had a solid 12 volts. It looks like my sensors are getting the power to turn on, right?
Dynamic beats static
Not so fast. This is where you can get in trouble. Testing the voltage on an open connector is not the same as testing it with the circuit operating, and if there is a voltage drop in the circuit somewhere, the only way I'll find it is to test with everything plugged in and "on." I performed a voltage drop test on the sensors’ ground circuit with the connector plugged in and found that I had about 1.9 volts on this circuit. For a ground circuit that’s way too much. After seeing this, I knew that our problem with this vehicle is on the ground side of the O2 sensor circuit.
I disconnected the sensor connector and hooked up a headlight in series with the sensors’ ground circuit. As I applied power to the circuit, the headlight lit up, but wasn’t very bright. That tells me the circuit is still making contact, but there is a high resistance in that circuit. Looking at the circuit diagram, it shows that the ground location is G113, which is located on a mounting stud at the transmission’s bell housing. I had the shop remove the stud and clean the mounting surface where the stud goes, along with the terminal ends of the ground circuit. After the work was performed, I again applied the headlight to the circuit and this time the headlight was brightly lit. One final test was to check the circuit with my meter to check that my voltages are correct. I had 12.3 volts at my sensor on terminal D and 0.01 volts on the ground side on terminal C.
I cleared the codes out of th ECM’s memory and took the vehicle on a test drive. The heated O2 sensor monitor set rather quickly, but some of the other monitors did not. The evaporative emissions (EVAP) and catalytic converter monitors did not run to completion even after a long test drive. Based on experience, the EVAP monitor takes a long time to run its internal tests but the converter monitor should have completed by now. I parked the vehicle and looked at the information system for Technical Service Bulletins (TSBs) on this vehicle, but did not have any luck. I went back to the car and looked at scan data. I noticed that there was a P0420 code now set in pending codes. This is one reason why it is critical to test drive and recheck every car you perform a drivability repair on. With the O2 sensors offline, the converter monitor couldn't run. One more drive cycle and the check engine light would have come right back on.
Not quite done
The description for code P0420 is catalytic converter efficiency low.There wasn’t any heated O2 sensor codes pending or otherwise. Now it dawned on me! With the vehicle being driven with both heated O2 sensors not working for what I assume was a long period of time, could they have damaged this cat? With the O2 sensors offline, this vehicle ran in open loop much longer than it was supposed to, which over a period of time is what probably damaged the cat. I had the shop replace the catalytic convertor, clear the codes and perform a road test. This time all the monitors set, and the car was able to get through the state inspection.
Sometimes when working on a vehicle, you come upon a problem that can cause another problem to be hidden. Fixing the cause could possibly show the effect. Make sure you thoroughly road test every vehicle that comes into your shop with a drivability problem. Had a good road test not been performed after the O2 sensor repair, this vehicle would have been a comeback. Another great tool to use for a drivability problem is at mode $06 data using your scan tool on the generic side. Looking a mode $06 data is a great way to check and see if your repair was successful or not. Working on today’s vehicles can be like going to war. Use all the weapons you have in your arsenal to win the battle.
An intermittent stall
While I was still at this shop, the owner asked me to look at another vehicle. This vehicle was a 2007 Chrysler Pacifica with a 3.8L V6. The customer complaint was an intermittent stalling at idle. There were no codes stored in the vehicle's ECM. This problem would happen randomly, but only when the vehicle is at operating temperature. The question I was asking myself was how am I going to find the problem with this vehicle when the problem only happens infrequently. With very little information to go on from the shop and the ECM not giving me a whole lot of useful hints, I decided to look i n some of my online resourcees to see if this was a common problem with these vehicles. After doing a search on this, I came up empty. Since this stalling problem doesn’t happen all the time and I can’t find any useful information on it, should I give up or go to war? Being the stubborn tech that I am, I decided to do a little out-of-the-box thinking. I decided to hook up my scan tool and put it in record mode and see if I could tell what sensor, component or module was shutting down. I let the car run at idle for about 15 minutes while recording the scan data. Finally I was able to witness the vehicle’s stalling issue. It acted like someone had just reached in and turned the ignition switch off. No stumble, rough idle or hiccup was noted. Was this car losing spark or fuel? Could it be an electrical issue? I didn’t think it was losing fuel becaue most of the time when a vehicle runs out of fuel, it slowly dies out. But in this case it was an instantaneous shut down.
Based on the movies I recorded and the testing I had done, I suspected the EM had an internal problem. I told the shop to get a replacement ECM, and I would come back and program it. A couple days later I returned and programmed the ECM and retested the vehicle. The car ran well and was test driven, along with letting the car idle for almost an hour. This car was fixed and the battle was won!
Both of these vehicles were tough ones to nail down. The hard part was each vehicle had an underlying problem that wasn’t staring me in the face. Sometimes you have to exercise a little more patience and diligence to extract the culprit. Many techs get frustrated when chasing down hard-to-find problems. Sometimes you have to take a step back and look at the weapons you have to fight this war. Choosing the right weapon will help you win the battle and hopefully the war as well. The best weapon of choice is your brain.
