Test, don't guess

Several years ago I worked for a chain outfit with a few younger techs. These youngsters were fast and performed the routine services that we offered quickly and efficiently. One day, a customer brought in a car that was built well before the days of electronic ignitions and control modules with a simple request for a tune-up.

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

The young tech assigned the task dutifully replaced the plugs, ignition cap and rotor and plug wires. He also installed new ignition points and condenser. With the ignition side of the tune-up complete, he reached into the car to turn the key. The engine cranked over smoothly but now would not start.

The youngster began his search for the cause of this no start by checking for spark. After all, he just reconditioned the entire ignition system. Of course, there was no spark and he began backtracking his work, looking for something left disconnected. Perhaps one

of the supplied parts was defective. He swapped out the new cap and rotor, he replaced the ignition coil and still the car would not produce a spark. He rechecked the point gap and static timing of the point opening. Still no spark.

By now, every tech (except me) was huddled around the car watching and offering their humble opinions. Most of the tips involved the words, “Have you tried replacing…?” I don’t recall hearing anyone say, “Have you tested…?”

This is a very simple circuit to diagnose. Key on power enters the ignition coil primary winding (did I mention there is only one ignition coil to deal with?) and the ground lead from the coil leads to the ignition breaker points. This mechanical switch is open and closed by the distributor’s cam lobes. The points themselves are attached to a contact breaker plate in the distributor assembly, with one side of the contacts fixed and the other free to move.

I walked over to the young tech and asked, “Have you tested for voltage at the coil?” The answer was, “No.” So I suggested he start there. Isn’t that the first step in a voltage drop circuit test, checking for power at the load?

“Now, let’s check the ground on the ignition coil. But let’s not crank the engine over with the key, let’s do it manually so we can watch the circuit voltage as the points open and close.” The youngster connected his voltmeter to the ground terminal at the coil, and with the points closed the meter read 0.30 volt.

“That’s good,” I told him. “That means there is no unwanted resistance on the ground side and the circuit is intact electrically. Now let’s try it with the points open.”

Rotating the engine slightly, we watched the meter as the points opened. The meter reading remained the same. As a side note, we could have opened the points manually and gotten the same results. “What do you think this is trying to tell you?” I asked the tech.

“Looks like the circuit is still grounded,” he answered.

“And why is that? The points, the control device in this circuit, are open.”

I could see the lights coming on in his eyes. He was straining, looking hard at the ignition points assembly. Finally, a little brown ceramic caught his eye and all the lights came on at once.

The wire from the coil has to be isolated from the frame of the points assembly. The assembly is attached directly to ground. There is an insulating washer used to separate the wire contact from the breaker body, and the youngster had the wire on the wrong side of the washer, resulting in a direct short to ground. He corrected this minor mistake and the car fired up on the first tickle of the key.

In his defense, he was not familiar with this antiquated ignition system. But had he applied a little thought and performed a few minor electrical tests, he could have saved himself a lot of time and

frustration. Instead, he assumed that he had done everything right and that it had to be a defective part. Rather than test anything to verify his diagnosis, he simply started throwing parts at the problem.

The Lessons Continue
My youngest son, now in his first year of college, is your typical Southern boy. In a corner of my property rests a 1989 Jeep Comanche Pioneer that is his project truck. He started with a two-wheel drive version of this neat little pick-up, and has since converted it to four-wheel drive with a moderate four inches of lift. Some of you may recall earlier adventures I’ve had with this little Jeep, including a resurrection from the dead after he drove it through a local mud puddle that turned out to be a bit deeper than he originally anticipated.

Now there is a new (read, “used”) engine in it. The replacement engine has been dealing with a yet unexplained rear main seal leak, and he and his friends have removed the oil pan on more than one occasion in an attempt to fix it. He insists on working on his project on his own, so I stand quietly off to the side, offering advice only when asked.

After the last pan removal, he found himself dealing with a no start of his own. Unlike the first I shared with you today, this problem was in the starter circuit. The engine would not turn over and no “click” could be heard when the key was turned to the start position. Suspecting that the starter may have failed, my son removed it and took it to the local parts house for testing. They found no problem with the starter, and he reinstalled the part.

I’ve become accustomed to working on later model cars, and it has literally been years since I’ve touched a car this old. Should that matter? Can’t I apply basic knowledge to this problem as I would any other? Or would the differences in the technologies of the day throw you off?

If this came into your bay, would you look at the components in wonder or ask yourself how they did what any starter system has to do?

Gathering Information
Any diagnostic process begins with gathering information. As I’ve preached before, relying on generalities is a guaranteed way of getting yourself into trouble. You need to spend a few minutes reading up on the system you intend to troubleshoot and learn its nuances so you can plan your diagnostic attack.

In the Jeep’s case, the first step was to pull a schematic of the system. There are two relays in the circuit. The first relay, when activated, sends power to the coil winding of the second. The second (the starter solenoid) completes the path between the battery starter cable and the starter armature. The other side of the armature is grounded through the case to the engine itself and from there, back to the battery via the battery ground cable.

We’ve been doing a lot on voltage drop testing in the magazine lately, with a print feature appearing in our June 2012 issue, and all sorts of video examples posted on our community. That’s because it is such a great technique for electrical troubleshooting. Let’s use it again here.

The first thing I want to know is whether the battery has sufficient energy to turn the starter through. Measuring 12.42 volts with my voltmeter provided a reading that was a tad low, but enough for now. Time to move on to the second step of the test. I left my negative meter lead at the battery’s negative post and moved the positive lead all the way down to the dark green wire bringing power to the starter solenoid. This was the first load I wanted to test and the current path (I love that term and owe Joe Glassford a thanks) leading to it arrived on that wire. I had my son rotate the key to start. This yielded a reading of 0.0 volt, indicating an open circuit between the battery positive cable and the end of the dark green wire.

Tracing back on the current path shown in the schematic, I could see that the relay had to close in order for current to flow. It also was possible that power wasn’t making it to the relay in the first place, or that the wiring itself had a break somewhere. What would you do next?

Following basic voltage drop procedures, I moved my lead back towards the battery’s positive post. My first stop was somewhere accessible, the battery side of the starter relay located on the right front shock tower. The color changed here to black, according to the schematic, and just before the relay connection an 18 gauge fusible link was installed as a circuit protection device. But measuring nearly the same voltage as I measured originally at the battery proved that wire (and fusible link) was intact. The problem had to be a relay contact that wasn’t closing, right?

Would you swap the relay next? That might be hard to do! I doubt you have a drawer full of this old-style component! No, we don’t swap. We test.

A relay is both a control and a load, depending on the task you’re diagnosing. I’ve already determined it wasn’t functioning as a control device. The switch wasn’t working. But the relay needs to be activated, doesn’t it? In that case, I need to diagnose it as a load in the circuit. Checking the schematic, I isolated the current paths for that side of the relay.

A green wire brought power to the relay when the ignition switch was turned to the start position. The relay ground was a black wire with a white tracer stripe, and it passed through the automatic transmission’s gear selector switch when the tranny was in Park. With the two current paths (power and ground) identified, the next step would be easy.

I still have the negative meter lead on the ground terminal of the battery. This is to ensure that I test the entire current path. First, I measure for voltage on the power side and I’m rewarded with a reading of 12.02 volts. Some drop but not enough to be concerned with yet. On to the ground side, where I measure 0.54 volt with the key turned and current flowing.

Wait a minute. That’s a good reading, isn’t it? The load side, the coil of wire that is energized to produce the magnetic field that closes the switch contacts, is intact and working.

I think that confirms the failed relay contacts, don’t you?

Not quite. There was one test I failed to make and had I done so I would have been done much sooner. Remember when I moved my lead to the battery side of the starter relay switch? I measured near battery voltage there, indicating the problem lay between that test point and the starter solenoid end of the dark green wire. I assumed

it was in the mechanical contacts of the relay. All it really told me was there was still an open circuit between the test points. Correct procedure would have been to start moving back to the first test point until I isolated precisely where the open circuit was located. Had I done that, I would have noticed that there was an open terminal on the starter relay, with no wire attached.

Guess what terminal it was?

The problem was simply a disconnected wire, inadvertently pulled off by my son’s efforts under the hood. No parts were replaced, either for “best guess” testing or for repair! And even with the mistake I made, testing time was minimal using this valuable technique.

Have articles like this sent to you weekly by signing up for our enews blasts here.