No Longer Hiding in the Attic: Repairing an alarm module in a Bentley

I haven’t written for a while because I’ve been busy with my four part-time jobs! Once of them has me working at Robison Service, listing items for sale on eBay and occasionally, doing other odd jobs.

John has written in his car blog several times, lately, about the problem of Bentley and Rolls Royce electronics modules and the batteries on them that start to leak over time. This alarm module came out of a 1996 Bentley Continental. Bentley only made a few hundred of these cars and is no longer making components for them. The controls are easily damaged when the batteries inevitably begin to leak and Bentley’s stock of new replacement units has been depleted. Since these control units are no longer available from Bentley, there is no choice but to repair them when they go bad, although some preventative maintenance of replacing the batteries is preferable to trying to fix one after it has gone bad. This one has gone about as bad as they get.

You can clearly see that the dark blue battery in the middle has leaked badly, causing a great deal of damage to the top of the circuit board and destroying copper paths. There is a second battery, teal in color, in the lower right corner. This one had also leaked and damaged traces, although not in such an obvious manner as the dark blue battery. I’m pretty good with electronics and I have the patience to debug such disasters, so John asked me if I could try to repair it.

The first thing I did was unsolder and remove the batteries carefully, wearing nitrile gloves to keep the leaking electrolyte from burning my skin. Leaky batteries and their detritus can be very nasty.

With the batteries removed, the damage is more apparent. I used a Sharpie to mark the battery specifications on the board, in case I lost the piece of paper that I had used to record their information. I like to put vital information in more than one location because I lose paper.

You can see that there are even traces missing from the bottom side of the circuit board, where the electrolyte leaked through some of the component holes in the board.

I wanted to remove the blue connector on the bottom side of the board, to make sure there weren’t any damaged traces under it, but I didn’t have the proper unsoldering tool. I looked carefully with a flashlight and my magnifying glass and decided that I might not need to remove it, so I left my final decision about it for later.

Next, I needed to know what kind of electrolyte remained. Electrolyte can be either a strong acid or a strong base. It depends on the type of battery. The dark blue battery was Nickel-Cadmium (NiCad), while the teal one was nickel metal hydride (NIMH). I went to Wikipedia and looked up both types of battery. I discovered that they both utilize a basic electrolyte, so my first plan to neutralize the remnants with baking soda wouldn’t work; I needed a weak acid. I chose vinegar because it is safe and I had a gallon jug of the stuff in my pantry. I raided the bathroom closet for a big box of Q-tips.

I poured a bit of vinegar into a small cup and dipped in a Q-tip, then lightly scrubbed at the stain. It bubbled vigorously. I discarded the Q-tip and repeated the process until I had cleaned both sides of the board, a process that took several hours. The stains only bubbled in a few places and only the first time I wiped them but, eventually, the tarnish also began to come off. As I scrubbed, I repeatedly moistened the edge of the blue connector (on the bottom side), encouraging a bit of the vinegar to slip under it, to neutralize any possible electrolyte remaining under the connector. I used more vinegar swabs to clear off the dusty bits from components on the adjoining circuit board. I was patient because I wanted every speck of electrolyte cleaned off of the board before I tried to initiate repairs.

Now that everything was nice and clean, I was ready to start tracing the paths of the damaged copper foil. I scratched at some traces, to gauge how deeply they had been etched by the electrolyte. Some came off entirely. I started by trying to replace the traces, but gave this up when too many traces turned out to be only a few molecules thick - that's when I started to systematically scratch at them with an exacto knife, to test their solidity.

I used 30 AGW wire to replace most of the traces. For the power traces, I used a heavier gauge wire. The small blue rectangles are capacitors; the larger black components at the bottom are diodes and the bigger black rectangles are relays. There wasn’t any logic, to speak of, on this board, so I was fairly sure that if I just replaced the broken connections on the board with wires, the board would function properly - individual components are inherently sturdier than logic components, which contain many interior components. Some traces moved from one side of the board to the other via little plated holes; I decided that merely connecting their ultimate ends was a more efficient means of repair.

For some of the top traces, it was easier to just connect the components on the bottom of the board.

I couldn’t find exact voltage replacements for either of the batteries at Radio Shack. I needed a 6 volt Ni-Cad and a 4.8 volt NIMH. I did find 3.6 and 2.4 volt Ni-Cads and 2.4 volt NIMH batteries. I connected two batteries (of the same technology) in series to add up to the correct voltages (the amp-hours were a little higher than in the originals but that shouldn’t create any problems). All four replacement batteries were meant for cordless telephones. I taped the pairs together, connected one red to the other's black to add the voltages in series, cut off the remaining connectors, and soldered the packets in place.