Overvoltage adversity

Kevin was not shocked when a Transit came in with some interesting electrical issues, so he got to work

Published:  31 October, 2022

Recently we had a 2009 Ford Transit Connect 1.8 TD come through the doors with 140,000 miles on the clock. The customer explained that if he drove the vehicle above 2,500RPM, the dash display all fell to zero and the engine would cut out. Then, within a second, it all came back online and would drive again, until the next time he hit 2,500RPM. The van already had a new alternator fitted in an attempt to resolve the issue, but this had been unsuccessful.
Armed with that information, a test plan was drawn up as follows:
1. Road test and verification of the fault
2. Global scan
3. Result-driven approach
4. Look up fault code
5. Gain access to relevant information, know faults and fixes, wiring diagram.
6. Equipment required; multi meter, scope etc. What type of testing voltage and continuity using a scope to monitor signals while system is operational.
7.  Study results from testing and plan a fix
8.  Apply the repair and retest
9.  Prove repair successful including road test
10. Write up job card and return keys and job card to the office.

Step 1: Road test. This lets us experience and validate the customer’s concern. I didn't need to go far as a quick acceleration up the hill had it doing its thing. The dash suddenly all fell to zero and the engine cut out. Then, just as quickly the dash came back to life the engine picked up and away the van went, until the next acceleration. Upon returning to the garage, a test plan was put together.

Step 2: Attach the scanner and do a global scan. This is always my preferred method, as the way systems interact means the same fault code can appear in many other systems. If we only concentrate on a single control unit, such as the ECU or dash, without considering the vehicle as a whole, we can miss vital pieces of evidence that could identify the reason for the failure. While the scanner was running through the systems check, I opened the bonnet and carried out a visual inspection. There was nothing obvious to report there, so I just verified that the new alternator was installed as described. How many of you are saying dash fault right now?  We have all seen them. Next, I checked the scanner fault report, which was where I found the golden nugget; Overvoltage.

Step 3: Results-driven approach. Having read the scan report, I highlighted the recurring fault code with the same description in each ECU it appeared. In fact, overvoltage was present in several ECUs. This gave us a place to start testing. See Fig.1.

Fault-finding mission
With this information in hand, my multi meter was attached to the battery and 15.34 was the reading on the screen. In order to catch exactly what was going on, a scope was used. When the vehicle revved up, the following trace was captured. It spiked over 18 volts. Please refer to Fig.2.
It is a well-documented fact that these alternators suffer with harness faults. With this in mind, the harness was visually inspected from underneath. It all looked good up to the plug by the battery box. These are smart charge, utilising four connections. The large one goes to battery positive, then the three in the plug are as follows: One is used for battery voltage sensing, the next sends the ECU command signal to the alternator charge level request, and the last one transmits the acknowledgement signal from the alternator back to the ECU.
The command and the acknowledgement signals are square wave PWM so I prefer to test with a scope for more accuracy due to a multi meter just averaging the signal. With the scope back-probed in to all three of the black plug wires, I could see the command and acknowledgement signals, but no battery sensor voltage was present. The next step for testing was to check the fuse, which was intact and ok.

This corrosion
I then removed the air filter housing to gain access from the top. This would allow me to test at the plug between the harness to the alternator and the vehicle harness. Next, the fourth channel of the scope was back-probed into the top part of the plug on the battery voltage sensing wire. No voltage was found here either, so a wiggle test of the harness was employed which caused the scope trace to jump up to battery voltage. At this point I pulled on the red battery sensing wire just above the plug and the insulation parted. Here we found what we were looking for; Corrosion, the green crusties of death for wiring. Please refer to Fig.3.
The other wires were pulled to see if their fate was the same, but the scope trace stayed steady for the command and acknowledgement, so they were fine; No attention necessary.
Then, a repair was made by stripping the terminal out of the plug and replacing it with a new terminal with 15cm of new wire attached, of the correct gauge and colour. The harness was then opened up and the green wire disease was cut out. A new wire was soldered to the clean old one in the loom, then covered over with heat shrink tubing re-tape. The loom was then put back up, we joined the connector back together and we re-tested.

Related Articles


©DFA Media Group
Terms and Conditions