A California Dodge offers a few lessons in electrical troubleshooting
The culprit is a 2005 Dodge Stratus SXT, equipped with a 2.4 transverse 4-cylinder and automatic transmission. The complaint is an illuminated Malfunction Indicator Lamp (MIL), and the stored code is P0491: Air Injection System Performance. Only the California version is equipped with a secondary air injection system, and its purpose is to add air into the exhaust when the engine is cold to speed up the heating of the catalytic converter, bringing it on line as soon as possible.
The code is recorded when the Powertrain Control Module (PCM) determines that too little or too much air has been added, based on information from a dedicated MAF (Mass Airflow sensor) located on the intake side of the air pump. In operation, air is drawn in through a separate filter, then the MAF, by the air pump.
The air then exits the pump to a check valve located on top of the exhaust manifold. Behind the manifold is a small chamber cast into the head, with four individual ports leading into each exhaust runner, just downstream of the exhaust valve. The pump is controlled by a relay, which in turn is operated by the PCM. This code could be set by a mechanical fault or an electrical one, so a little homework was in order before testing began.
How a Control Module Operates
Diagnosing circuits operated by a control module is not as hard as many techs think it is. Like any other diagnostic situation, it all comes down to understanding how the module operates. Control modules are computers programmed to carry out specific functions by their designers. In order to carry out their programming, they need information.
In this first test, we verified that the PCM was in control of the system and it was working. The MAF signal, however, shows very little actual air flow.
This information is delivered by a variety of sensors. In the case of the P0491 code, the primary input is the ECT (Engine Coolant Temperature) sensor. The colder the engine, the longer the PCM will run the air pump, up to its maximum of 20 seconds. Input devices provide either an analog (varying voltage) signal or a digital (on/off) signal to the module. Analog signals are not well understood by the computer, and most analog signals are further converted in the module to a digital equivalent by an A/D (Analog to Digital) Converter.
When the module has the information it needs, it will carry out its programmed instructions and operate an output device. Output devices are typically internal switches that either complete the power path to a circuit load (a relay, solenoid, or coil just to name a few) or complete the ground path to the load. These internal switches are called drivers A power side driver is referred to as a high side driver, while one that completes the ground side is referred to as a low side driver.
In the second test, the check valve was disconnected from the exhaust manifold. The MAF signal now shows no restriction, at least up to this point. Time to pull the exhaust manifold and have a look.
These drivers can be simple on/off switches (relay control), operated at a programmed duty cycle to mimic a varying voltage to the load (alternator field coil) or turn on the load for a programmed amount of time (fuel injector). The latter method is called pulse width modulation, or PWM, and differs from duty cycle modulation in that there is no total cycle time involved.
PAGE 2At first glance, the corrosion on the valve led me to think it may be the culprit.
In many circuits, the module knows that it commanded the operation of the load but doesn’t know if that operation was successful. In these cases, some form of feedback is needed to let the module know if its commands were carried out successfully. These feedback inputs can be from a dedicated sensor like the MAF sensor used by this Dodge, or from an existing sensor that the module monitors for change in state as the command is executed.
For example, the oxygen sensor on our Dodge responds to the air flow injected by the secondary air injection pump by switching lean, providing an existing sensor feedback to the PCM that lets it know the air it commanded actually got there. If no change of state in this sensor was seen, the PCM would assume that nothing happened and record a code P0410: Air Injection O2 Sensor Monitor. The absence of this code in our problem child tells us that some air is flowing, and is one more observation to include in our “gathering information” phase of troubleshooting.
Testing Control Module Circuits
Control modules serve different purposes electrically, depending on their function in the circuit.
Until the manifold was removed, that is. Then you could see that there were no ports for the air to flow through.
First, they are an electrical load in their own right, and as such, need a good power supply and ground. These can be tested like any other electrical load, using the voltage drop testing method. The first step is to identify all power and ground supplies to the module, using the appropriate schematic. You’ll also need a connector pin-out diagram to further identify exactly which connector pin is which. Problems in either power supply or ground will affect how the module operates and can also affect the reference voltages and grounds the module provides to some of its sensors.
Second, they are control devices, completing circuit paths on either the power side or the ground side of the controlled circuit. These can be tested at the controlled load, also using the voltage drop testing method. Here, though, your Digital Multimeter (DMM) may be useless for verification. Some of these controls are operated at speeds a DMM can’t keep up with, and the use of a Digital Storage Oscilloscope (DSO) is preferred. Using a DSO also can help to catch intermittent losses of control that you’d never see on a typical DMM. Be aware that some low side drivers use an offset ground called a floating ground that you will see on your scope capture when referenced to battery ground. This is normal and should not be confused with a ground side voltage drop problem unless it is excessive.
Last, they act as a source to a variety of sensors, providing power and ground independent of the battery. These are referred to as reference voltages and can be full system voltage or any amount designed in by the engineers. Most common is the use of 5 volt references to sensors like the Throttle Position Sensor (TPS). When testing these sensors, it is best to reference your meter leads to the supporting module, treating it like you would the battery when performing a normal voltage drop test.
PAGE 3This is the correct head for the car. Notice the ports just above each exhaust port.
Control modules can only perform well when the information they receive is correct. You can check the associated Parameter Identifier (PID) for these inputs on your scan tool, but be aware that substitute values may be displayed if you are using your scan tool in enhanced, factory specific, mode. You can verify the truth of the information by testing at the sensor input pin directly at the control module. If you find the measurement is incorrect, you can then back track to the sensor itself to see if the sensor is the culprit or the problem lies in the connection between the two.
Time to Fix the Dodge
Now that we know how the system is supposed to function, we can design tests of our own to determine what the underlying cause is. In this case, there is either a mechanical problem affecting air flow or an electrical one affecting the actual operation of the various electrical components.
There are a few different approaches we could use to narrow it down. We could use a scan tool with bi-directional control to manually operate the pump while monitoring the MAF signal, or we can test the system the same way the PCM does by watching everything at once while the system operates normally. Since we only have about 20 seconds to gather this information, a scope capable of recording the events was the tool of choice.
What do we want to know? Is the PCM receiving the correct information it needs to carry out its mission? Are the output drivers working? Are the controlled loads working? What type of feedback is the PCM seeing? Channel 1 (blue) of the DSO was connected at the PCM to the relay control pin. This one would tell us whether or not the PCM was commanding the air injection pump on. If so, then any input information the PCM needed in making this decision would be proven correct and we would know that the PCM was acting according to its programming.
Channel 2 (red) was connected at the PCM to the MAF signal return. This connection would tell us what the MAF was reporting as actual air flow through the system and help determine if there was a mechanical problem with the system.
Channel 3 (green) was connected to a low amp probe clamped over the power side of the pump motor circuit. This channel would verify whether or not the pump was running, and help determine if any fault existed in the relay or pump motor circuit.
Last, channel 4 (gold) was connected directly to the battery to let us know when the vehicle started and when it was shut off. With all 4 leads in place, we would be able to see what was really going on, just as the PCM would.
By monitoring the inputs and outputs directly at the PCM, we can see what the PCM sees and determine if it is acting according to its programming.
On the first run, the PCM does indeed complete the electrical path to the pump relay. The 12-volt signal means this is a high side driver. The pump is confirmed running by the current flow detected by the low amp clamp. So far, the system is working just as it is supposed to. However, the signal from the MAF is telling the PCM that very little air is flowing. This could mean one of two things. Either the MAF sensor or its circuit has a problem or there is a mechanical restriction to air flow. Which would be the easiest to check? Let’s disconnect the check valve from the exhaust to see if we can feel the air flow through it, and see what happens to the MAF signal.
There was just enough cold left in the engine to run the test again. With the valve disconnected from the manifold, air flowed through the valve with no apparent problem, and the MAF signal read normal. This left only one possibility — a mechanical restriction to flow in the exhaust ports. With this information in hand, it was enough to get authorization to remove the exhaust and take a look.
One point I failed to mention. This vehicle only had 34,000 miles on the odometer and I was somewhat surprised that the tests indicated a restriction. Until I removed the exhaust that is.
You see, the reason there was a restriction was because there were no ports in the head. This engine had had the Federal head installed for some reason in its past. The Federal head does not have these ports machined into it, so there was nowhere for the air to go.