Assigning Sensor Authority
(or hierarchy of needs)
By Brian Manley
I took the call one evening. My friend, Tony, had a question regarding an '88 Mercury Marquis that started, died, and would barely run - even while coaxing the gas pedal. I agreed to head out to his home to help diagnose the Merc, so I made a mental list of tools I thought I'd need, loaded them into my trunk and headed over the next day. I rarely get a chance to do a house call and I thought this just might be a challenge!
A thorough visual inspection revealed no major issues. The car had 72K miles, was literally driven by a little old lady, and had been well-maintained. The air filters, plugs, fluids and belts were all new, and no repair to the vehicle preceded this case of the "no-runs." All vacuum lines and wiring harnesses were intact and appeared to be routed correctly.
I decided to use our New Generation STAR (NGS) tester to run a key-on, engine-off (KOEO) test to probe the mind of the EEC-IV processor and see if it had recorded any glaring faults. After the scanner booted up, it gave me a code alright, except this code was one of its own: "FATAL SYSTEM ERROR 7" (Figure 1).
I was a little puzzled; I was hoping the NGS would give me a code "X" that would point to part "Y," which I'd then test to confirm or deny its ability to perform its assigned duties, and move on from there. I also wasn't sure what that fatal error meant. Could it point to a possible processor failure? I'm aware that these old EEC-IV systems had TSBs that recommended processor replacement for excessive spark knock, hesitation after cold start-up and rough idle. It was then that I began forming a list of possible suspect systems and components.
Ignition and Fuel
Time to begin with the end in mind, so I hooked up the fuel pressure gauge to the fuel rail test port, a spark tester to a plug wire, and a noid light to a fuel injector harness connector. With the key on, the fuel pump pushed 41 psi; good enough for a 5.0 liter (Figure 2). I also observed a strong snap at the spark tester and a consistent pulse from the noid light.
Hierarchy of Needs
Psychologist Abraham Maslow once proposed a hierarchy of needs for human beings as it relates to how our brains relate to the world around us. He proposed a pyramid where the most important needs for humans - physiological and safety needs - were at the bottom and deemed primary above all others. Physiological needs (satisfying hunger and thirst) and safety needs (the need to feel that the world around us is organized, predictable and safe), are the first requirements of a normal human brain. Only once these needs are met, Maslow believed, were we prompted to meet other needs such as achieving independence and respect from others, and fulfilling our own unique individual potentials.
If I shift this analogy to our Mercury's power control module (PCM), then I must determine which inputs to our car's "brain" are primary and which are secondary. This next step, one which I take when faced with a no-code or a no-scan-tool communication issue, is to develop a hierarchy of inputs to the PCM beginning with the most important needs. The base pulse width in our Merc, like most fuel injected systems, is controlled by two major inputs: rpm and input from the load sensor; in this case, our manifold air pressure (MAP) sensor. These two inputs are critical to calculating base pulse width for start-up, and then the pulse width is modified by other inputs such as coolant temperature, throttle position sensor (TPS), oxygen sensor and adaptive fuel trim memory.
Rpm and load input are critical and, of course, without rpm there would be a no-start. We verified with our noid light and spark tester that the PCM was receiving a signal from the distributor pickup and was capable of using that input to fire the injectors. The EEC-IV ignition module was also capable of understanding this input, then using it to trigger the ignition coil. We still weren't sure if the spark was occurring at the correct time, but I put a loose timing chain low on my no-start list.
Hertz, Doesn't It?
Following my logic, I back-probed the three-wire MAP sensor. With my digital volt/ohm meter (DVOM) on the VDC scale, I probed the MAP ground wire with the black lead, and the reference voltage (VREF) wire with the red lead. I read just under 5 volts; correct for our 5-volt system. Next, I moved my red lead to the signal return wire and switched my DVOM to the Hz scale so I could test the frequency output of our sensor. I read 102.6 Hertz. Compare this reading with the values in the MAP Frequency chart.
Other car makers use analog MAP sensors that output a variable voltage based on the pressure in the intake manifold. On this Ford product, however, the MAP sensor is frequency-modulated, which means the output signal is a variable frequency rather than a variable DC voltage. To test these sensors accurately, you must use a frequency counter, which appears on most of the better digital automotive multimeters. You can also use a lab scope, but did I remember to throw my Fluke 98 into my trunk? Grrrrr!
The following table shows the correct reading for a KOEO test:
|MAP Sensor Frequency Data|
|Altitude (feet) ||Frequency (Hz) ||Barometric Pressure (in. Hg)|
|0 ||159 ||30.1|
|1000 ||156 ||28.9|
|2000 ||153 ||28|
|3000 ||150 ||27|
|4000 ||147 ||26.6|
|5000 ||144 ||25.4|
|6000 ||141 ||24.2|
|7000 ||139 ||23.0|
According to this critical input, the PCM was calculating base pulse-width and delivering fuel based on its present location - the top of a very tall mountain. The PCM believes itself to be parked at around 21,000 feet; how much fuel do you think is actually squirting from the injector pintles at this altitude? Not much, to be sure, and apparently not enough to start the engine.
This particular sensor is dual purpose; it sends a barometric pressure reading with the key on, engine off, and then switches to a MAP input with the engine running. The lower the barometric input number, the lower the fuel injector base pulse-width will be.
Other Critical Inputs
Not one to rely on one condemned sensor, I also tested the next two sensors on our "needs" list, a decision based more on personal experience and frequency of failure than their order on the list. Since a TPS that is returning over 4.0 volts to the PCM will elicit a "clear flood" output from the PCM, shutting down the injectors, I always test this out of habit. The TPS swept just fine and settled at .80 volts at idle. I pulled the CTS connector to test its resistance, which indicated it was, indeed, in a 75 degree garage.
A new MAP sensor gave us a KOEO reading of 145.7 Hz. The 5.0 liter fired up and ran well (Figure 3).
1984 Chevy Camaro - Code 32: Baro Sensor or Circuit Problem
How much authority does a BARO or a MAP sensor have over a carbureted 2.8 liter V-6 engine? Well, this '84 model came to me soon after its engine had been replaced - a fact that prompted me to look very carefully at all of the vacuum and wire routing before jumping right into sensor-failure mode. The Check Engine lamp was illuminated at all times. My first once-over of the engine compartment did not reveal any glaring errors in routing, so I grabbed my scanner to probe the PCM. I found a code 32 - Baro sensor or circuit problem, and a reading of 30 kPa and .82 volts, which would indicate that this Chevy was at 20,000 feet! I went back to perform a closer visual of the BARO sensor and discovered this Chevy was equipped with both a BARO and a MAP sensor. The MAP sensor, located on the left side of the firewall, has its vacuum hose attached, and it had sufficient vacuum reaching the sensor. The BARO sensor, located on the right firewall, also had a vacuum hose attached that originated from an intake manifold source; not what a BARO sensor requires. If you look at the photo, you can see the hose had been cut and pushed over the sensor's port, where a small foam filter should be (Figure 4).
Removing the hose brought the BARO reading back to a believable level - 82 kPa and 3.66 volts. This reflected our altitude of 5,800 feet, and extinguished the Check Engine lamp.
How much authority does a MAP or BARO reading have as an input to the PCM? If the engine is fuel injected, a faulty sensor can, and often will, cause a no-start condition. On carbureted engines, most have a fixed main fuel orifice that will still allow sufficient fuel into the engine to keep it running, and most engines will still run well even if the mixture-control device is unplugged. Most sensor inputs are used by the PCM to trim the mixture, and customers only bring them in if they fail the enhanced emission test for excessive CO, or if the bright orange lamp on the dash annoys them long enough.
Brian Manley is a vocational automotive instructor for the Cherry Creek school district in Aurora, Colo. He is an ASE master certified automobile technician and a former member of the National Automotive Technicians Education Foundation (NATEF) board of trustees. He can be reached at email@example.com.
AutoInc. Web Site |
ASA Web Site |
Repairers Suffer Used Oil Regulatory Gap |
Making Sense of Sensors |
Computerized Measuring |
In the Beginning, I Had a Dream |
Guest Editorial |
Tech to Tech |
Tech Tips |
Around ASA |
Shop Profile |
Net Worth |
Stat Corner |