Tackle other-worldy vehicle problems by relying on the data

Oct. 1, 2016
Every so often some of the vehicle problems we encounter can seem sort of paranormal. Even though we keep telling ourselves that there must be some logical explanation to what is causing the fault, the data we are observing is incomprehensible and our usual tried-and-true testing reveals little or no guidance.

With Halloween coming up, I have lately been reminded of the short-story horror series, Tales from the Crypt. That’s because every so often some of the vehicle problems we encounter can seem sort of paranormal. Even though we keep telling ourselves that there must be some logical explanation to what is causing the fault, the data we are observing is incomprehensible and our usual tried-and-true testing reveals little or no guidance. With that theme in mind, here’s our first story.

2007 Toyota Camry with Code P0606 ECM Processor Error

On this vehicle, I needed to program a new ECM that had been diagnosed and installed by another tech. Maybe programming isn’t the proper term, however, since pretty much all that was needed was a VIN Write, but I was happy to oblige. When the tech got back from the test drive, code P0606 was already set, and I was asked to go over the diagnosis that led to the ECM needing to be replaced. Now, even though there needs to be some verification of the controller’s power(s) and ground(s), when you get an internal processor code it is highly likely that you will be replacing the ECM, so what went wrong on this diagnosis? 

After a short test drive with a new PCM installed, the same Code P0606 returned. This is the only code that is set which is why the technician ignored the code flowchart that had the O2 sensor codes combined with the P0606. They instead followed the code flowchart for P0606 Only, which simply states to Replace ECM.

When looking at the trouble code flowchart, I noticed that there were two options for diagnosis; one for a P0606 stored with other codes and one for a P0606 stored by itself. Well, only the P0606 is stored with nothing else even in pending status, so according to the flow chart: REPLACE ECM. No testing, no checking, nothing, just replace. Ok, I can see where the tech called for the replacement ECM, and he was justified in making the call. But there had to be a reason the same code would reset with a new OEM ECM. My first thought was a poor ground connection, so I used an ECM connector pin out from ALLDATA and my LOADPro dynamic test leads to check the integrity of the grounds for the ECM. If you’re not familiar with the LOADPro, it is a device that replaces your regular voltmeter leads and allows you to perform a dynamic voltage drop test with a push of a button on the circuit you are testing. Well guess what? Grounds test fine with virtually no voltage drop. Since I’m here, let’s check the voltage supply, too. Same thing. The terminals of the harness connector also passed a pin drag test.

At this point, I am getting the eerie feeling that we received a defective ECM, but what are the odds that a new ECM would have the exact same problem as the one we replaced? Maybe this vehicle is possessed. Remember when I looked up the code and there were two options listed for P0606? A lot of times, if there are multiple codes, it would make sense to ignore a certain code if other codes are set that have a higher priority, or it could be that they could cause that code to set. Well the other codes are oxygen sensor codes. Could an O2 sensor cause an ECM processor malfunction? Yes, according to the code set criteria. A malfunction of the Air/Fuel Ratio Sensor transistor or the Heated Oxygen Sensor transistor could cause the ECM to believe there is an internal processing error. After hooking up the scan tool and graphing the data from the oxygen sensors, I couldn’t believe what I was seeing. Looking at the impedance or the Rear Oxygen Sensors, it was easy to see that while Bank 1 Sensor 2 was reading in the range of 90 ohms, Bank 2 Sensor 2 was exhibiting spikes up to 21,247ohms! I guess that is something that can trick an ECM. 

Something I have not monitored before but definitely looks out of place is the impedance for the B2S2 Oxygen Sensor. Note that is as several spikes reaching over 21,000 ohms, ultimately this is what was setting the ECM processing error.

After replacement of the Bank 2 Sensor 2 Heated Oxygen Sensor, we found that both rear bank oxygen sensors’ impedance came down as the engine warmed up and stayed relatively flat. Keep in mind when monitoring the impedance on a cold engine, the resistance will be higher, but will drop as the sensors warm up. A long test drive confirmed that the poltergeist that was possessing the Camry had been exorcised.

After installing a new rear O2 sensor on Bank 2 the readings were back to normal and the code did not reset. Note how the impedance of the rear O2 sensors drop as they warm up to operating temperature.

2007 Hyundai Sante Fe with Code P0133 B1S1 slow response

Our next story turns out to prove that sometimes we are our own worst enemy. A 2007 Hyundai Sante Fe with a 3.3L engine had showed up at our shop with an illuminated MIL due to a P0171 Bank 1 Lean and P0174 Bank 2 Lean codes currently present. The technician who diagnosed the problem, a very competent one at that, smoke tested the vehicle and found that the intake gaskets were not sealing, causing the lean condition the customer was experiencing.  After removing the intake and installing new gaskets, the technician test drove the vehicle to verify everything was good. However, horror struck when they found a pending code P0133 Bank 1 Oxygen Sensor slow response. The technician took a quick look at the scan data to confirm a problem with the Bank 1 Front Oxygen Sensor and added it to the ticket to get replaced. 

The data obviously matches the code description for a slow responding Bank 1 O2 sensor, but the graph of the sensor looks eerily familiar to a post cat O2 sensor.

A second test drive was performed with the new oxygen sensor installed and déjà vu, a pending code for the Bank 1 O2 Sensor returned from the grave. He asked for some advice on what could be causing it so I suggested checking for exhaust leaks. It was just in for some lean codes and maybe the condition wasn’t all caused from the intake leaking. Both a visual inspection and a smoke test of the exhaust system were performed and no problems were found. This time, I was asked to take a look at the vehicle myself.  The first step was to check for any TSBs, as anyone who has gotten burned by not doing this before can attest.

The Bank 1 Front and Bank 1 Rear O2 sensors are located right next to each other in the darkness of the back of the intake plenum and the firewall.
Notice that with the front and rear oxygen sensor connectors unplugged, the locating tabs are identical so the technician did not notice any problem with reconnecting them in reverse locations.

I like to graph data so that I can see the change of the PID over time, especially if it’s something like an oxygen sensor. Using the Hyundai GDS scan tool, I graphed both front oxygen sensors so I could compare them to each other, and what I found was creepy. The oxygen sensor for Bank 2 was switching normally and within a good operating range. Bank 1, however, was definitely slower, about seven times slower than Bank 2. So I was seeing why the PCM would flag the code and also why the other tech made the call on a defective sensor. I politely asked if he was sure he put the new sensor on the correct bank (notice I said politely, since asking a frustrated technician a question like that can have painful side effects). He assured me he did. Creating a vacuum leak did show a response from both sensors, but again, Bank 1’s response time was lagging way behind Bank 2. Could it be a bad sensor? It’s possible, but how could it fail in such a way that voltage fluctuation peaks are OK, but they just occur at such a slower rate than the other side? Almost like a pre- and post-catalytic convertor oxygen sensor, like a lightning bolt it came to me. The Bank 1 pre-cat sensor is on the side closest to the firewall and a little difficult to get to. It just so happens that the connector for the Bank 1 post-cat oxygen sensor is right next to it. Well when the tech replaced the intake manifold gaskets, he disconnected the Bank 1 front and rear O2 sensor connectors and wouldn’t you know it, both connectors have the exact same locating notch. Since the connectors are right next to each other, there was no stretch on the wiring and although the sensors connector ends do have different colors, without good lighting in the tight, dark area between the back of the intake and firewall, he probably just put them together thinking there was no problem since they both snapped together easily without any resistance. Reconnecting the front and rear Bank 1 oxygen sensors into their correct corresponding harness connectors corrected the slow responding O2 graph and fixed the vehicle.

After swapping the Bank 1 Front and Rear Oxygen Sensor Connectors to their correct locations, the Bank 1 Front Oxygen Sensor was reading at a similar frequency to the Bank 2 and the vehicle was ready to be returned to the customer.

2010 Ford Mustang running lean and mean

Our last story is another vehicle I received from a tech who had been struggling with a rough running and hesitation condition on a 2010 V-8 Ford Mustang. While using freeze frame to see when a code sets is an important piece of information, an entire diagnosis should not be based off the readings. This is what I believe happened when the tech made the decision to replace the intake manifold gaskets on this vehicle. The freeze frame showed that the engine was just warming up when a lean condition was occurring, so instead of suspecting leaking manifold gaskets and confirming the problem, the trigger was pulled to replace and he shot himself in the foot. As many have said; test first, replace later. The tech continued to struggle with the vehicle over the next couple of days testing fuel pressure and volume. He smoke tested the intake and exhaust system for leaks, but found none. He also replaced the Mass Airflow Sensor with an OE one from the dealer, but the problem remained unaffected by the repair attempts. 

While the volumetric efficiency test showed great results, there was definitely a problem with fuel control on this Mustang. Note that the short term fuel trims were fluctuating wildly from -7 to over +30.

When I was asked to look at the vehicle, I first thought I knew what the problem was before even looking at it, since I had seen other vehicles with lean conditions caused by the quality of fuel. Not that I have a crystal ball, but here in Las Vegas we have a large amount of gas stations that carry E-85 and it is fairly common for someone to just put the cheapest gas offered by a station because they assume it is 87 octane. Most of the time the vehicle still runs OK, but the check engine light is on due to a lean condition. A quick tech tip to determine if the fuel contains 85 percent ethanol is to monitor the fuel trims at idle and 2500 RPM. Vacuum leaks will usually show positive trims at idle, but improve at high RPM, while fuel delivery issues are just the opposite, usually reading OK at idle but trending positive at higher loads. E85, however, is positive at both idle and 2500 RPM. In other words, you may have +22 percent fuel trim at idle and +25 percent at 2500 RPM.  If I see fuel trims move like that, the first thing to check is the fuel quality. A quick peak at the fuel trims and I realized that this was not going to be that easy of a diagnosis. The short fuel trims were all over the place, ranging from -7 to +32 and long term was ranging from 0 to +12. I kept thinking that I must have bad scan data since this is not making any sense. I still did take a fuel quality sample, but it  showed the normal 10 percent ethanol content. 

The clue that led me to the cause of the problem was looking closely at the scan data from the Mass Airflow Sensor. Notice the erratic movements of the sensor while the vehicle was idling.

I needed to determine if I was having an airflow measurement problem so I decided to perform a volumetric efficiency test. So with the VE calculator app on my phone, I entered the numbers from my test drive. The results showed 87 percent efficient and on a naturally aspirated vehicle, that is better than most. So that rules out an engine breathing problem and the MAF was already replaced. On the test drive I had noticed that the vehicle was struggling under acceleration, like it wasn’t getting enough fuel, so that’s where I shifted my focus. I confirmed the fuel pressure by attaching a gauge and watching it as I made some hard accelerations and pressure remained rock solid. Scan data showed that long-term fuel trims were actually perfect at higher RPMs, but short term from slightly rich (-7 and-7) to extremely lean (+32 and +31). At this point, I was ready to phone in a medium because something unworldly was occurring —great volumetric efficiency, steady fuel pressure, good quality fuel and fuel trims that are making no sense whatsoever. With the file of the test drive saved on my Snap-on Solus Edge, I loaded it into Snap-on’s Shopstream Connect software program. I decided to think about it overnight while trying to pick out something I was overlooking on the data. Since the mass airflow sensor was a new OE part and I had good volumetric efficiency, I never paid much attention to it until I zoomed in. Looking closely at an idle portion of the data revealed a lot of oscillations of the sensor, which in turn created havoc with both front O2 sensors. Something is causing a disturbance in the airflow and it’s being picked up by the MAF sensor. Since the MAF sensor does not require the air filter box to be removed to change, the previous tech never removed it. When I did separate the air box from the air filter, I found a piece of heat insulating material that was glued to the top half of the box had partially fallen down and was flapping in front of the opening that led to the MAF sensor. This was causing all kinds of turbulence in the air stream. 

The adhesive that secured the heat liner to the underside of the air box had given up the ghost and allowed part of the material to flap in front of the MAF sensor creating all kinds of turbulence in the air stream.

Gluing the piece of insulating material back to the top of the air box reincarnated the Mustang, as its engine wailed when test driven on the road. 

Sometimes we encounter vehicles with problems that make little or no sense, and it can drive us to a point of exhaustion trying to decide what to test next to get a grasp on what may be causing the concern. What I have found is it is best to step back from the vehicle and look deeper in the scan data for answers before we lose our heads.

Sponsored Recommendations

Best Body Shop and the 360-Degree-Concept

Spanesi ‘360-Degree-Concept’ Enables Kansas Body Shop to Complete High-Quality Repairs

ADAS Applications: What They Are & What They Do

Learn how ADAS utilizes sensors such as radar, sonar, lidar and cameras to perceive the world around the vehicle, and either provide critical information to the driver or take...

Banking on Bigger Profits with a Heavy-Duty Truck Paint Booth

The addition of a heavy-duty paint booth for oversized trucks & vehicles can open the door to new or expanded service opportunities.

Boosting Your Shop's Bottom Line with an Extended Height Paint Booths

Discover how the investment in an extended-height paint booth is a game-changer for most collision shops with this Free Guide.