It had seemed like a routine speed sensor replacement at first, but for me, it became a trip to the invisible forest surrounded by trees…

The Repair Order

1997 Escort

1.9L Engine

First Stop: Willy’s Bay

            Willy drew the repair order and started out with a test drive.  The speedometer was inoperative and the transmission wouldn’t shift out of first gear.  While the speedometer wasn’t mentioned on the repair order, Willy tied the two problems together, knowing that the 4EAT transmission (and just about every other automatic transmission in operation these days) needs a valid vehicle speed signal for shift timing.  Generally, if the Powertrain Control Module doesn’t get a speed signal, the transmission just hangs there in first gear and won’t shift because it doesn’t know the vehicle is moving.  If the speed signal is lost while driving down the highway, the speedometer needle can drop to zero, the transmission can drop to first gear, and if the cruise is operating, it will suddenly become dysfunctional.

Willy plugged Ford’s New Generation Star tester into the Data Link and pulled a code P0500 out of the PCM’s continuous memory, indicating the loss of vehicle speed sensor signal.  He installed a nice new Vehicle Speed Sensor, took another test drive, and everything was normal.  The speedometer was reading nicely and the transmission shifts were crisp and on time.  The Escort was parked, the repair order was turned in. Another job well done.

“Oh, By the Way…”

            Well, the lady came to pick up her car and the porter brought it around for her.  She drove out of the writeup area and noticed the speedometer wasn’t working, so she looped around the front of the dealership and pulled back into the writeup area.

            “My speedometer hasn’t worked for awhile,” she told the service writer, “could you take a look at that before I leave?”  The service advisor called me to the writeup and handed me a new repair order with instructions to check the speedometer.  He failed to mention Willy’s part in this little drama.

Read, Research, Verify

            I started out from ground zero with no idea that Willy had just replaced the Vehicle Speed Sensor.  My test drive produced an inoperative speedometer, and it didn’t take long to make my way back to the shop and wheel into the service bay.  My first course of action was to find out whether or not there were any visible problems at the sensor, which is mounted in the transaxle’s final drive housing like most front wheel drive speed sensors.

(Note: In another article, I wrote about a speedometer problem on a 1997 F150 where I had disconnected the speedometer, checked the wires and connections for voltage and continuity, and then reconnected the cluster to find that “rebooting” the cluster computer had brought the speedometer back to life. That couldn’t happen on the 1997 Escort, particularly because the relationship between the speed sensor and the speedometer is different.  While the Escort speedometer is electronic, it doesn’t operate through a built in instrument cluster microprocessor like the F150. It’s a stand-alone unit that receives power, ground and VSS signal directly through the printed circuit.)

On the Escort, there weren’t any problems visible to the naked eye, but the sensor looked suspiciously new. Next I needed to know whether or not the PCM was receiving a speed signal. 

The speed signal is generated by the sensor and makes its way along the circuit to a splice, then branches out from the splice to the PCM, the instrument cluster, and the speed control amplifier.  If I could determine that the speed signal was being delivered to the PCM, then I’d know there was a problem somewhere on the circuit feeding the speedometer head.  If the Data Link indicated that the PCM was receiving a speed signal, then I could check the wires for continuity from the instrument cluster to the sensor and thus further narrow down the source of the problem.  It didn’t hurt to make quick check of the fuses, but no fuses were blown…

No Communication…

When I connected the Worldwide Diagnostic System cable to the Data Communications Link under the dash, the WDS popped a message on the screen indicating that it couldn’t link up to the PCM.  I knew the WDS would be looking for power and ground at the connector first, so I checked pin number 16 in the Data Link connector to find nice healthy 12 volts there.  I also found a good ground at pin 4. These two wires provide power and ground for the scan tool. The datastream from which information is retrieved travels from PCM pins 15 and 16 to the Data Link connector pins 10 and 2 respectively. Think of these two wires as a telephone connection from the car’s computer to your scan tool. Like telephone wires, if they’re cut or shorted, no communication will occur.  The computer language by which they communicate is called “Standard Corporate Protocol” or SCP. 

The “telephone” circuit leading from PCM 15 to DCL 10 is Bus (-) and PCM 16 to DCL 2 is Bus (+).  When a voltmeter is connected to those two wires on a healthy system, the meter will read almost exactly 5 volts if no data is being transmitted.  If you backprobe these wires at the DCL while PID data readings are being transmitted, you’ll read about 2.5 volts here.  During Diagnostic Trouble Code retrieval, you’ll see 5 volts, but with occasional downward spikes as the data is sent. When I connected my meter to the Bus, I read 5.55 volts. That in itself indicated a problem, but I didn’t yet know what the problem was.  When I connected an oscilloscope to the bus, I got a dead flatline across the screen. 

While I wasn’t completely certain this communication problem might have something to do with the inoperative speedometer, I should have figured it did and followed that line of reasoning further. But since the speed sensor was hard-wired to the PCM and the cluster, I couldn’t see any connection between the Data Link and the problem I was having. I decided to check the circuits leading to the PCM and the cluster and worry about the Data Link communication concern afterwards.  It was an error in judgment that would cost me some extra work…

Sensor Ohms and Signal Voltage

            Removing the console is a simple matter on the Escort and the PCM is mounted right in front of the gear shifter, with the connector readily exposed and easy to access after the console is out of the way.  Carefully probing pins 58 and 33 at the PCM, I clicked my meter over to ohms and found just over 200 ohms coming through the Vehicle Speed Sensor windings. When the speed sensor was disconnected, the reading changed to infinite resistance. 

Reconnecting the PCM and backprobing the same two pins, and with the Escort on jackstands, I clicked my meter to AC volts, started the engine, and spun the wheels up.  I saw a nice AC voltage appear and increase with wheel speed.  The signal was definitely making it to the PCM.  I decided to move my test leads to the Instrument Cluster connector and check the speed sensor output there.  It takes a little time to get the cluster out, because the steering column has to be lowered, but a few minutes later I was checking pins 9 and 7 at the instrument cluster to find the same reading I’d had at the PCM, but the Speedometer was still as dead as a hammer. 

This electronic speedometer head has four wires feeding it through the printed circuit. Two wires are 12-volt feeds, but one is hot all the time and the other is hot only with the key on.  Both of these 12-volt wires have to be hot for the speedometer to work.  The third wire delivers the Vehicle Speed Sensor signal, and the fourth wire comes from the ground side of the Vehicle Speed Sensor. That’s where I found my problem.  I knew for a fact that both VSS wires were making it to the cluster, but what I didn’t know and hadn’t checked until now (duh!) was that the ground side of the VSS circuit wasn’t grounded to the negative side of the vehicle electrical system!  As I tracked that ground, I found that it went to the little cluster of connectors that Ford had plugged into the negative ground cable at the battery.  While the negative cable was a bit rusty, it wasn’t corroded, but a glance at the ground bus where the wires plugged in told me where the problem was.  The part of the bus that was visible was green and chalky in appearance. This ground feed was obviously creating an intermittent problem with the speed signal received by the Speedometer and the PCM.  Since there were several ground leads originating here, it was conceivable that the PCM could have received a signal at times while the Speedo head was inoperative.  That would explain how the Speedometer could fail to work while the Transmission still shifted normally.

Fixing it Right

I cut the old battery terminal off and clipped the ground wires leading to the chalky bus.  The nasty green chalk had used capillary attraction to wick itself about six inches into the negative battery cable, and I had to keep on clipping wire until I found clean copper.  While I could have replaced the whole cable,  I decided to handle this a different way.  I acquired a piece of battery cable a little longer than I needed, along with a length of 18 gauge black wire.  I soldered the piece of cable and the two wires into a solder-on battery terminal, using my handy little butane torch for a heat source and good rosin-core solder.  With the application of heat shrink, I had a battery terminal that looked almost factory and was a whole lot more dependable.  Then I soldered this home made outfit onto the negative cable and applied more heat shrink.  Thirdly, I connected the ground wires from the vehicle harness to the two 18 gauge black wires I had soldered into the new negative terminal.  The clamp-on universal terminals have never appealed to me except as a quick fix on a lonely highway in the rain.

Living and Learning

            This experience taught me a lot that I should have already known.  For instance, I should have spent more time studying the schematic, and I would have realized that the Data Link connector has a ground feeding pin 5 in addition to the ground I had checked earlier at pin 4.  One thing I did know but had failed to factor in was that one of the wires from the Vehicle Speed sensor is supposed to have continuity to battery ground. Had I factored all that in, I would have figured this one out a lot sooner.  It was a classic example of not seeing the forest for the trees, and we’ve all been there time and again.

R.W.M.