Adapted from Gary Goms’ article in Underhood Service.
A client who owns a heavy-duty diesel shop called me to help diagnose a no-cranking condition on his father-in-law’s 2006 Chevrolet Tahoe. Because his work is mainly with heavy-duty trucks, my client knew he was lacking both in tooling and technical background when he didn’t hear the familiar click of a starter solenoid as he turned the ignition switch to “start.” Not anticipating a major problem, I loaded my mobile digital volt ohm meter (DVOM) and scan tool kits, picked up my client, and headed to the mountain cabin where the Tahoe was located. Unfortunately, it was a cold and windy fall day and the Tahoe was parked in a small, dark garage cluttered with a variety of do-it-yourself projects.
I usually begin a starting system diagnosis by connecting a scan tool and, on this Tahoe, the scan tool should indicate when the ignition switch engages the starter relay. But it didn’t take long to discover that although my scan tool would communicate with all other on-board modules, it wouldn’t communicate with the Tahoe’s powertrain control module (PCM). Next, I checked the fuses and relays located in the underhood fuse box to make sure the PCM was powered up. The battery tested 12.6 volts at the terminals, indicating a fully charged battery. As I suspected, I couldn’t feel or hear the starter relay click when the ignition switch was turned to the “start” position. Although most of the battery-activated fuses tested at 12+ volts, we discovered a number of ignition-activated fuses testing in the area of 4.5 to 5.2 volts in the key-off position.
The design of Chevrolet’s large underhood fuse box makes an excellent hotel for small rodents, so, with that thought in mind, we quickly removed the fuse box and found no traces of rodent activity.
Attempting to diagnose the no-cranking problem in a cold, dark home garage located miles away from my own shop would be difficult and time-consuming. For that reason, I recommended that the Tahoe be towed to my client’s commercial diesel shop for further diagnosis of what I call “The Case of the Missing Starter.”
Catching the Bus
The major problem I experienced during the initial diagnosis was determining whether the vehicle’s communications bus system was operating correctly or not. Sure, I could test each pin on the diagnostic link connector (DLC) with my DVOM, but other than verifying power at pin No. 16 and grounds at pin Nos. 4 and 5, I had no idea if the communications bus was working with the scan tool connected. In addition to the Tahoe, I had also recently experienced the same type of problem diagnosing communications bus issues with Chrysler PCMs.
After returning to my shop, I ordered a new DLC connector breakout box (BOB) to be shipped as soon as possible. Unlike a pin-out test, the DLC BOB allows a diagnostic tech to observe the electrical activity of all power, ground and applicable communications circuits through an LED light display while the scan tool is connected to the DLC. Each of these circuits can be easily tested for voltage and waveform activity by connecting banana-plug test leads commonly supplied with most automotive multimeters and lab scopes.
After the Tahoe arrived at my client’s shop, I connected the DLC BOB and scan tool. Surprisingly, the scan tool immediately communicated with the PCM and the BOB, which indicated that all bus circuits were functioning correctly. But upon attempting to start the engine, we discovered that the starter drive would suddenly disengage, as if the starter drive was defective, and then I’d lose communications with the PCM. This was the critical clue I needed to solve The Case of the Missing Starter.
Clearly, we had an intermittent ground problem at the PCM. Since a number of ignition-activated fuses were testing in the area of 4.5 to 5.2 volts in the key-off position, I was sure that at least one voltage source was “back-feeding” through these fuses. The PCM wasn’t communicating with the scan tool, so the diagnostic arrow pointed to the PCM ground itself.
Going back to basics, the wiring schematic indicates that the ignition switch acts as a mode switch that commands the starter to crank the engine. When the PCM receives the command from the ignition switch, it grounds the starter relay located in the Tahoe’s underhood fuse and relay box. The relay engages the starter solenoid which, in turn, activates the starter motor and engages the starter drive gear with the flywheel.
The wiring schematic confirmed my diagnosis of a faulty PCM ground, but locating the missing ground proved to be problematic. First, the Tahoe’s engine cylinder head is grounded to the body by a woven strap located on the driver’s side, which is misleading because the woven strap does not ground the PCM to the engine cylinder head. A wiggle test on the main harness confirmed that we had an intermittent ground circuit. After much fishing around behind the engine plenum, my client finally located the missing PCM ground wire behind the passenger side of the plenum.
The frayed wire explained why the starter failed to remain engaged with the flywheel. Since the dangling wire acted like a pendulum, the simple act of opening the door and sitting in the seat was enough to allow a single strand of the wire to complete the ground to the cylinder head. While a single strand is enough to bring the PCM to life, it won’t carry enough amperage to complete the ground circuit for the starter relay. And, due to the shock of the starter engaging, the PCM ground wire would then disconnect. A few minutes later, we removed the wire end loop from the cylinder head only to discover that a medium-sized rodent had gnawed the ground wire completely away from the wire loop connection. Oddly enough, the PCM ground wire was the only wire that the rodent selected to whet his appetite for plastic electrical insulation.
Initially, it would have been easy to conclude that either the Tahoe’s starter or its PCM was defective, but because voltage was seeking ground through the ignition-activated fuses, I knew that the solution to The Case of the Missing Starter would be found somewhere in the ground circuits. The most critical clue was the erratic operation of the starter. Although we knew the PCM was grounded to the back of the passenger-side cylinder head, it took some time to discover that the single ground wire to the PCM had been cut in two by a large hungry rodent living in the wooded area surrounding the owner’s garage. We soldered an extra length of wire to the PCM ground lead and attached it to the front of the cylinder head where it could be more easily inspected and repaired in case the rodent returned.
The Big Picture
While the PCM-controlled starter circuit isn’t new by any stretch of the imagination, it’s an indicator of how vehicle architecture is changing.
Going back to the 1950s, a few vehicles were still equipped with foot-operated starter engagement pedals that protruded from the floorboard. Pushing down on the starter pedal simultaneously connected the starter motor to the battery and mechanically engaged the starter drive gear with the flywheel. Many vehicles were later equipped with starter buttons that would activate a starter relay or solenoid. Later, the starter button function was integrated into a spring-loaded position on the ignition switch. The starter relay would engage the starter motor and, on many early applications, inertia alone would engage the starter gear with the flywheel. In more modern applications, a combination relay and solenoid would mechanically engage the starter gear with the flywheel.
All we have to do is look at the large number of relays in an underhood fuse box to understand just how many vehicle functions are now controlled by modules like the PCM. In addition, we must understand exactly how information is shared among these various modules. The ABS module, for example, supplies wheel speed data to the PCM and from the PCM to the speedometer. Another example is the fuel level sensor reporting to the PCM because fuel level is part of the evaporative emissions monitor. The PCM then reports fuel level to the instrument cluster via a bus communication wire.
Vehicle architecture is also dictating how we bill for diagnostics. A fuel level gauge on the 2006 Tahoe can easily be diagnosed with a scan tool by monitoring the fuel level PID and by checking for related diagnostic trouble codes. In contrast, the fuel gauge on a 1995 Chevy K1500 is most commonly tested by grounding the wire from the fuel level gauge to a clean spot on the frame. Here we have two different architectures, two different diagnostic methods, and two different labor times for diagnosing a faulty fuel level gauge. In any case, keeping up with today’s vehicle architecture is the key to solving tomorrow’s Diagnostic Dilemma.