Wacko Wiring

By Richard McCuistian

Evaporating intermittents can be especially annoying…

1999 Ford Ranger

103,056 miles

3.0L Vulcan Engine

4R44E Transaxle

Starter engages with ignition in RUN.

Come-And-Go Problems

            There isn’t an experienced technician anywhere who hasn’t had to deal with those pesky wiring problems that come and go.  In 1977 I was working at an independent shop where the lead technician went head-to-head with a ’75 Monte Carlo that would intermittently lose its ignition power feed while driving down the highway.  He told me before he left on his test drive that he was headed out of town on a certain road and that if he wasn’t back within half an hour to come looking for him.  Well, forty five minutes later I remembered, and he was more than a little disgruntled when I finally got there.  As it was, he hadn’t thought to bring anything with which to check for spark, and so he fished a 4 inch piece of 5.8 heater hose out of the back of the shop truck and pushed the handle of his pocketknife up into the hose thinking the rubber would prevent a shock while he checked for spark.  Well, the fiberglass strands in the heater hose made a dandy high voltage path, and Mike was measurably impressed by the jolt he got from that HEI system.  The ignition feed to the distributor and the spark that went with it had quietly returned as the under hood temperatures stabilized.  The end of that story was that the Monte Carlo had a pushed back connector terminal in the big bulkhead shell.

            Come-and-go wiring problems can range from mildly annoying to downright deadly, depending on which circuit is at fault.

Even when a short or an open is consistently present, it isn’t always easy to pinpoint the exact location in some of the thicker and more circuitously routed harnesses, and in many cases, the smartest thing to do is simply get some loom and do a quality permanent overlay, but that can be a bit complicated if the circuit in question makes its way through a splice and then spiderwebs out to feed a lot of stuff.

An Engaging Starter

     The 99 Ranger was one of those leftover jobs that was simmering on the back burner; Spanky had tinkered with it off and on before going on vacation, and since I was putting in some tech time to satisfy NATEF requirements, the shop foreman acquainted me with the Ranger’s symptoms. While the problem had evaporated during the troubleshooting process and hadn’t reoccurred, everybody who was involved knew it wasn’t fixed.

Donnie (Spanky’s mentor) had installed a new ignition switch some eighteen thousand miles earlier for the same concern, and as far as I was able to tell (most service writers don’t always get the whole story the first time around) the problem hadn’t resurfaced until now.  Talk about an intermittent! 

As Spanky and the shop foreman attempted their diagnosis, they gathered some interesting data that I was able to sort through in my search for the source of the concern.

When the Ranger came in on the hook, simply switching the key to the run position would engage the starter and fire up the engine, after which the starter drive overrunning clutch would be working overtime until the key was returned to the off position or the gear selector was moved to any position other than Neutral or Park, at which moment the starter would be disengaged. That pointed to a short circuit somewhere between the TR sensor and the ignition switch.  Removing the starter relay would also disengage the starter.

This is a simplified wiring layout of the starting system on the 1999 Ranger. Notice how intimatelyassociated circuits 1000 and 1002 are in the harness. Both run from the ignition switch through the fuse panel to the Radio and the CTM. 

So far, so good.  With the relay removed and a test light probe carefully inserted into the coil cavity of the relay socket (pin 85), it was evident that a short to power was available at the relay coil any time the key was on.

Notice that when the ignition switch is in the start position, the START terminal on the ignition switch carries power from the battery through fuse 24, through the Neutral Safety switch all the way to the starter relay, which energizes and spins the starter with power from Battery Junction Box (formerly called the Power Distribution Center) fuse 5.

Further experimenting while the test light was glowing revealed the fact that removing fuse 20 (random fuse-snatching was the chosen method in this case) would cause the problem to go away.  Fuse 20 (see diagram) carries power from the ignition switch RUN terminal down to the radio and the Central Timer Module (CTM) via circuit 1002, but has nothing whatsoever to do with the starter (at least under normal circumstances). 

What Do We Know Now?

            At times like this, it’s wise to stop and get an overview of what we can deduce from the data we’ve gathered, and with the wiring schematic spread across several pages in more than one section of the book as this one was (I custom drew the schematic in this article for clarity), finding the exact source of a problem like this can turn into something of a nightmare. 

            If removing fuse 20 (which is hot in RUN and feeds circuit 1002) kills the short to power, how can we determine what path the current is taking to make its way back to the starter relay coil?

            Fuse 28 wasn’t checked during this fuse-snatching process, but as I did more research, I found that that circuit feeds START current from the ignition switch to the Radio and the CTM via circuit 1000, which runs in the same harness all the way from the ignition switch to the Radio and the CTM. 

Since pulling fuse 20 killed the short, then it was a pretty good bet that circuits 1000 and 1002 were shorted together somewhere and that circuit 1000 was backfeeding ignition switch RUN power through fuse 28 to the starter relay coil.  If that was true (and I believe it was) the short circuit had to be located inside the Radio, inside the CTM, or somewhere in the harness where wiring circuits 1000 and 1002 run external to the solid state components.  

Disconnecting the harness connectors, I found nothing melted together or crossed, and the harness itself didn’t show any signs of having been disturbed.  How those two wires might get together inside a taped up harness when there had been no other known electrical concerns is pretty mysterious (although I had on occasion seen it before), but at this point it would be financially foolish to replace the $190 CTM/GEM and an even more expensive Radio on a hunch.  It would be labor intensive to rip the harness out of the dash and dissect it from stem to stern looking for a problem I might not find.

            It would, however, seem simple enough to pinpoint the concern at this juncture except for one thing; before I ever even saw the Ranger (it had been sitting there in the back burner for a week), the problem had evaporated.  Before I got there, the shop foreman had noticed that when he hit the dash panel above the instrument cluster the problem would go away.  He could bump the dash again (he has a sledgehammer fist like the Thing in Fantastic Four) and the problem would return, but after a few of those bumps, the problem simply wouldn’t re-occur.  It was an innocent mistake, but it caused me to rip half the dash out looking for some type of short circuit, and I never found anything.  Should I run a series of overlays?  Possible, but unsavorable.

Looking for a wiring problem when it’s present is kind of enjoyable, at least for me.  Looking for a fix for problem like this that isn’t there now but used to be and will be again is no fun at all.

What To Do?

            I always hate situations like this:  Everybody involved knows the vehicle is 98% likely to come back, because nothing at all was done to repair the concern, and in the case of wire harnesses, temperature and vibration generally work things back to where they were before and the problem returns in all its glory. 

Some customers tend to think the technician is incompetent in such cases, but wiring isn’t an exact science, particularly where space age electronics are concerned.  The tiniest whisper of current into or out of an electronic box can cause all sorts of problems. 

On some of the first ‘catfish-looking’ Tauruses (96-97), the neutral safety switch could develop an internal short that would fool the GEM module into thinking the ignition switch was in the start position, and the GEM would kill power to the radio and the wipers.  I once saw 8 volts leaking out of a GEM module on one of those Tauruses through a relay coil in the Battery Junction Box back inside the vehicle, where it made its way to the weaker Accessory Delay relay coil so that the radio was always on.

With these two wires running such a parallel path and into the same module, it certainly seemed plausible that these two wires were getting together somewhere, somehow. If that was the case, then pulling fuse 28 should solve the problem with no side effects whatsoever, especially since fuse 28 has no purpose on this particular truck.

This pinpoint test step in the Ford shop manual claims that a dead circuit 1000 can prevent radio function, but that simply isn’t the case on this truck. Besides, circuit 1000 is only hot with the key in START. Why would the radio need that in order to work?

As I examined the schematics I began to ponder why it was that the CTM module and the Radio needed to know the ignition switch was in START.  Going to the symptom chart pinpoint test in the Radio section, I found an error.  According to one particular test step (see above), the circuit in question (Circuit 1000, fed through fuse 28) has to be juiced up before the radio will work, but that simply wasn’t the case on this truck, because  I removed fuse 28 and the radio still worked.  Further research revealed that the only reason the CTM/GEM module needed that circuit was related to illuminated entry as it interfaces with Remote Keyless Entry.  The shop manual reads this way:

The GEM illuminates the interior lamps when an unlock signal is received from the remote anti-theft personality (RAP) module. The illuminated entry feature will be canceled if:

· 25 seconds have elapsed since the illuminated entry feature was activated and the courtesy lamp feature is not activated.

· If the GEM receives a request from the RAP module (remote transmitter lock button pressed) and the courtesy lamp feature is not activated.

· If the ignition is in the RUN or START position and the courtesy lamp feature is not activated.

Since fuse 28 was obviously there for no particular reason on this no-frills, no power accessories truck, why not remove it and see if the problem reoccurred?  It seemed to be the simplest solution, since the customer probably wouldn’t want to spend the money he would have to cough up for a wire harness overlay.

Conclusion

            In this particular case, reason presented a scientific solution that seemed plausible, especially since the concern had disappeared and wouldn’t re-occur.   Removing the unnecessary fuse may or may not have solved the problem, but we were honest with the customer about what we did and why we did it. Let me also say that what we decided to do on this particular vehicle is almost never possible; 99.9 percent of the circuits on today’s cars are necessary and can’t be shut down without some sort of consequence, so be extremely careful and do your homework. As exact a science as electrical wiring may seem to be, there are times when things just don’t work out according to the book or the labor guide.  This was one of those times.                                            R.W.M.