Assume nothing, test everything

We have a 2011 Dodge pickup truck on loan so one of my students can perform research on Active Regeneration of the Diesel Particulate Filter (DPF) and the NOx absorbing catalyst (NAC). I was told the truck had a new Engine Control Module (ECM) to cure issues with hard starting, low power, and multiple Diagnostic Trouble Codes (DTCs) intermittently showing up. I was assured the truck was now fixed.

Like this article? Sign up for our enews blasts here.

Being happy to have the use of the truck, I did not question anything. Three students were performing an active regeneration after they had fooled the ECM into thinking the DPF was loaded with soot. They were analyzing scan tool data, taking exhaust gas readings and testing temperature changes when the truck went into limp mode and the Service Engine light came on.

I first thought they had done something to cause the problem but looking at the codes, I decided differently. I instructed them to start diagnosing what had happened. When I got back to them, they had the wiring diagrams and diagnostic procedures printed. They had taken some readings with their Digital Volt-Ohm Meter (DVOM) and already had concluded that the new ECM was bad. First thought here was, “How am I going to explain this?” but then I started thinking about how the ECM had already been changed. I made a phone call to get more history on the previous repairs. I then logged in to look at the truck’s repair history.

This is where the story got interesting. The truck had a history of multiple electrical and drivability issues. The truck seemed fixed for a week or so then something would reoccur, forcing the truck back to the dealership. After another call to talk about the truck’s current problem, they were glad to let me take a stab at diagnosing it.

After looking at the diagrams and DTCs, I wanted to start the truck and scan over the data to see if something jumped out. But when we started the truck the DTCs became inactive. Wonderful! Intermittent faults are so much more fun! We tried to get the fault to reoccur, but had no luck. I told the students to go back to running the regen procedure. I would soon realize why the

truck was such a pain. The truck was still doing fine even after running another regen. The next day, another group of students started researching the Air Fuel Ratio (A/F) sensors. After running the truck for about an hour, the DTCs returned.

We now had five DTCs present. The first one listed was a P0405 EGR Position Sensor Circuit low which meant all three EGR valve position sensors read lower than 0.40 volts for more than five seconds. One student asked, “How can you have all three sensors fail?” I was unsure how to answer him doubting they all failed at once. The next DTC was a P049D indicating the EGR Control Position exceeded learning limits.

Now, I am getting confused. How can the position exceed limits if the sensor circuit was low on voltage? I moved on to the next DTC, a P02E8 indicating the Intake Air Flow Position Sensor Circuit voltage was low. The big kicker to this problem was the next DTC of P02E9, Intake Air Flow Position Sensor Circuit high. How can the circuit be high and low at the same time? The last DTC was a P0652 Sensor Reference Voltage 2 Circuit low, which really made sense later.

While it’s beginning to sound like an ECM, I can’t swallow that. The P02E9 was the real key to the problem, but at the time I was not thinking clear while students were rendering their own opinions and asking questions.

The first step was to view the diagnostic flow charts and wiring diagrams. After looking at the wiring diagrams, I always like to draw simplified wiring circuits. This helps me to get through all the distracting parts of a diagram so I can focus on the system layout. At the time it was useless to take any readings as the data looked correct and the codes were inactive. We let the truck sit over night to cool down so we could make our test connection to the EGR Valve and Throttle Air Flow Valve.

We hooked our DVOMs to each wire at the connectors for the Air Flow and EGR Valves. We also backed the DVOMs up with scopes connected to each circuit to capture the fault as it occurred. Sometimes meters are too slow to catch glitches, and with a good scope I can take recordings and play them back when looking for that “needle in the hay stack” type problem. The students started the truck to warm it up. No problems occurred after running it for 20 minutes so they restarted the regen process. They were backing it on the dynamometer when the MIL (Malfunction Indicator Lamp) came on.

Confusing Development
We scanned the data for the EGR and Intake Air Flow Control Valve. The voltages were low on the scan tool so we compared them to the readings on our voltmeters. They showed the same voltage on the sensor signals as was reflected in the scan tool data. But the meters showed the 5.0 volt reference was down to about 2.4 volts. So why do we have a code for a high circuit voltage on the Intake Air Position Sensor?

We looked through the wiring diagrams for the reference voltage routing and then drew a simple circuit of just the reference voltage from the ECM to the sensors. Nothing jumped out at me so I had the students start following the diagnostic flowcharts. I went to check on other students while

they supposedly followed the charts. I came back about 30 minutes later to be told that the ECM was bad. The students claimed they had followed the charts exactly, but had found nothing. I detest the last statement on a DTC chart: “Replace ECM and retest” Now it was time for me to sit down with the wiring diagrams and do some studying.

I started looking at the diagrams to see what the students may have missed when one of the students discovered the five-volt reference circuit fed seven other sensors. Why weren’t we getting DTCs from the other sensors? This reference voltage circuit is the supply for the Fuel Rail Pressure Sensor, Crankcase Pressure Sensor, Camshaft Position Sensor, Inlet Air Pressure Sensor, Exhaust Differential Pressure Sensor, Boost Pressure Sensor, and the Radiator Fan Control. As they started the truck, I noticed an extremely long crank time like a low-pressure fuel supply problem such as air in the line. But we did not have any new DTCs, so I am starting to wonder about the ECM myself.

I then remembered that was one of the reasons the ECM had been previously changed. We loaded scan tool data for all nine sensors and sure enough all were reading incorrectly. This corresponds with the low reference voltage and makes sense, but why no codes for them? I am really wondering about the Cam Position and Fuel Rail Pressure Sensors being off so much without codes. I was trying to gather my thoughts while I scrolled back through some of the oscilloscope recording when I noticed one of the channels had spiked at about 4.95 volts. This was the scope channel that was hooked to the Intake Air Flow Position Sensor circuit. Then we noticed the reference voltage on the other two channels spiked up to 5.10 volts at the same time. I am convinced now that we have an electrical circuit problem, a short in the harness somewhere, but where?

The Missing Link
Do you ever get those feelings that you are missing something, but can’t put your finger on it? I knew it was staring me in the face, but was unable to put it all together. I went back to the simple circuit diagram I had made and then scanned over the diagnostic charts looking for the answer. Most of the diagnostic procedures have you testing for shorts and resistance between circuit connections. I decided to start over from the beginning with the diagnostics. Sometimes it is best to not assume all the steps have been followed nor followed correctly.

The data for all the sensors on the 5-volt reference was my first stop. I looked it over and then it hit me. The reason the ECM was not setting DTCs for all the sensors is that the readings were all within the normal voltage limits. The ECM had not had time to rationalize the readings were inaccurate. The fuel rail pressure reading was off causing the hard starting. The Cam Position Sensor still was able to produce a signal, as were the other sensors that were on the same reference voltage circuit. The problem had to be intermittent and related to temperature, but what else? I had an idea. I looked back at the oscilloscope recordings and when the Intake Air Flow Valve moved the voltage spiked. The signal voltage went high at over 4.9 V then it went below .4V. This should and apparently did set codes for high and low Air Flow Sensor Circuit voltage. I then realized the Air Flow Valve moved as a result of the ECM commanding it to allow for EGR flow.

Here is where I asked the students if they really did follow the diagnostic charts through each step properly. Naturally, I got a yes, but wondered. I decided to start over with the diagnostics. Playing a hunch, I asked which one they did last. They said the P0652 Sensor Reference Voltage 2 Low. I asked why they picked that one for last. They said they wanted to eliminate the other codes as they still believed the ECM was bad. I had them connect their volt meters again at the EGR Valve.

With all sensor connectors connected, the voltages still were well below normal. The diagnostic chart for the P0652 code has you unplug the connector for each sensor and measure the reference voltage at that point. If one of the sensors is shorted, when you unplug it the reference voltage should return to normal on the rest of the sensors.

That is why I had them hook up at the EGR valve as they had already replaced it, along with it being the easiest to get to. Again playing a hunch, I had them unplug the Air Flow Control Valve connector. The five volts immediately returned. After the looks on their faces returned to normal, I quizzed them about following the diagnostics precisely and thoroughly. They admitted they had assumed the ECM was bad and did not want to go through all of the steps. We replaced the Air Flow Control Valve and the truck has run for weeks now without any problems.

Important Tech Tip
We found several references in the diagnostic procedure steps saying to “Power Down” (key off) the ECM for 10 minutes. If you ignore this step the codes will not clear. We found another twist also. If we disconnected and reconnected the connector at the ECM (even with the batteries disconnected) we would get something like 20 codes upon the next key on. Those cleared normally. If you have phantom codes, before doing anything else, unhook the batteries (both) for at least 20 minutes, clear the codes, and then go after what codes might return.

Subscribe to Motor Age and receive articles like this every month…absolutely free. Click here