The most difficult problems I have ever experienced in all the years I have been in this industry are the intermittent ones. My most favorite one of all was a mid 90’s Ford Thunderbird with an AODE.
The customer brought it in, saying it would neutralize from time to time going down the road. I took it on a road test for about twenty minutes, worked like a new one the whole time. Took it to lunch, then drove it some more later in the day.
We got the customer’s permission to drive it home and on some errands, still nothing. At this point, I asked the customer to come and drive it with me in the car to see if he was doing something different that might contribute to the problem. During the drive, we could not duplicate it, so I asked, how often does this happen? About once every six months to a year was the answer I got.
Thankfully, not all intermittent problems are that infrequent, and for several years now, we have had graphing scan tools and multimeters. Even problems that might happen once a day can still be difficult to diagnose, these tools like most, make life a little easier.
I recently had a 62TE in a Promaster giving my friend Mike some problems. The owner stated they mostly drove it around town on short trips. It would make three to four shifts and then neutralize until they slowed down enough for the transmission to go into failsafe mode. Then it’d set an overdrive solenoid circuit code. This might happen a few times a day, or only a couple of times in a week. Each time it happened, they would cycle the key, and drive several more miles.
The customer had another shop replace the solenoid pack and check the wiring harness. They drove it a couple days, and the problem came back. After taking it back, the shop rechecked everything and came up empty-handed. The customer then brought the vehicle to Mike’s shop to have him look at it.
Without knowing the extent of what the last technician did to the vehicle, I felt we should start from the beginning. I recommended we start with the battery to check it’s condition and clean the terminals.
For most solenoid circuit codes, the possible causes are the solenoid, the wiring harness, connectors, or the computer. In the connectors, we want to check for any kind of terminal damage such as bent pins or female terminals that are loose or corrosion. The wires need to be inspected for any damage to the insulation or the wiring. Next, we disconnected both ends of the harness from the computer and the transmission. We checked for continuity from one end to the other, then to ground, then to other wires. Everything appeared to be good, so next up was a load test on the circuit.
Load testing the circuit is one of the best methods to verify circuit integrity. Resistance testing only puts a small amount of voltage through the circuit to test it, whereas a load test will check how much amperage can be carried on a circuit. With the battery disconnected and the computer connector removed, we bypassed the solenoid power feed wire and installed a fused jumper wire from the battery positive to feed the solenoids. Next, we connected a graphing meter between the negative terminal of the battery to the pin on the computer connector for the overdrive solenoid command wire. We then set the meter on amperage and in graph mode.
There are two ways to find the amperage that is supposed to be on the wire. One way is to check resistance and check voltage, then do a little math, volts divided by ohms equals the amps we should see. Or you can also check the other circuits that are not setting codes and compare them to the ones that are.
While doing our test, we also wiggled the wires from one end to the other, and gently pulled on connectors to see if we could get the amperage to drop out even for a second. The amperage did not fluctuate even a small amount, this confirmed the circuit from the totally integrated power module to the solenoid, then back to the computer was in good condition. All of this lead to my worst fear, a potential computer problem.
Computer problems can be tedious since any kind of ground or power supply issues can lead to what seems like a computer problem. Power related issues can usually be spotted in the data stream of the scan tool. Ground problems need to be monitored by doing a voltage drop test at the time when the problem occurs. So, I asked Mike to connect his scan tool in graph mode to monitor computer command, and his back probed into the ground wires for the computer and the solenoid command wire to monitor voltage.
It took a considerable amount of driving to get the problem to occur, once it did, he hit the brake pedal twice to provide a marker on the scan tool movie and stopped the vantage so he could review both. He found the computer was commanding the overdrive solenoid to come on for a fraction of a second, and the transmission would neutralize until he slowed down enough for it to go into failsafe mode and stay in 2nd gear. On the vantage, the voltage did not drop at all, and the grounds stayed at three millivolts.
Since what we see on the scan tool is only what the computer wants to do, and what we see on the vantage is what is happening on the wire, along with the grounds properly functioning, this led us to believe we had an internal computer failure.
Like most people, Mike did not like the idea of replacing a computer, so he decided it would be best to run the test a couple more times to verify the data he found. With consistent results, he ordered the computer, programmed it, and went for an extended test drive. He could not get the code to reset, or the problems to reoccur, so he called the customer to get the vehicle. After a few weeks of driving, the customer stated they have not had any problems out of it since.
Proper diagnostics can be time-consuming. With an intermittent problem, they can become nightmares. Electrical problems can often be misdiagnosed, and replacing computers won’t fix the problem if the cause is in the circuit. Most customers won’t mind paying for the time it takes as long as the problem is gone after the work is done.