Just a slight bit of track damage...!!!

[1024MAK]

Just a slight bit of track damage to this PCB...

These are pictures of the damaged board after stages one and two of cleaning it.





It's nothing to do with retro computing.
So who would like a guess at (1) what it is, and (2) what caused the damage?

Mark

[Dave]

1. A PCB

2. Leak from a battery?

[1024MAK]

1. A PCB

Yes, it is a PCB. But what item is the PCB used in?

2. Leak from a battery?

The was no battery or cell in the unit.

Mark

[stephen_usher]

Looks like long term water damage, with the copper going into solution via electrolysis.

Is it some sort of keypad for use outside?

[1024MAK]

Looks like long term water damage, with the copper going into solution via electrolysis.

Yes, aided by the water being rich in limestone and other minerals...

Is it some sort of keypad for use outside?

No, not a keypad. It is designed to IP67.

Mark

[1024MAK]

I will give a hint at 21:00 and the answers at 23:00.

Mark

[1024MAK]

Hints: see what the other comments are on these other forums:
https://www.worldofspectrum.org/forums/ ... ack-damage
http://stardot.org.uk/forums/viewtopic.php?f=11&t=12422

Mark

[1024MAK]

Okay, for those of you who are still puzzled, the pictures are of the PCB from a 16 year old MARL LED bulkhead light fitting. It was used to illuminate and mark the location of an emergency telephone in a tunnel. The tunnel where this unit was used is not open to the public (it's on private property). Because of it's application, it is powered continuously, and because of the supply arrangements, it is supplied by a 110V AC mains supply. This 110V AC mains supply is derived from a system that includes an automatic starting back-up generator. Hence no batteries are needed in the unit.

The bulkhead light fitting is rated IP67 which means water should not get in when properly installed and mounted. But over the course of approx 15 years in service, some water got in through the screw mounting hole. Because it is in a tunnel, the water has tricked through limestone rock, so is rich in minerals. Once the water was in the fitting, it could not escape. As the fitting was mounted on the arched part of the tunnel wall above the emergency telephone, the water gradually filled the fitting.

The circuitry on the board has two separate circuits after the bridge rectifier, and because the fitting was not fully level, one half of the board remained just clear of the water. So on that side, the LEDs continued to work. Hence we did not get a report of a failed lamp.

Meanwhile, part of the other half of the circuit board was wet. As it was wet in mineral rich limescale water for a long time, it appears that the copper tracks went into solution via electrolysis. As well as that, there was a build up of limescale rock deposits on part of the board where the water was dripping on it.

So at some stage, this action "etched" through part of the copper PCB track, causing that half of the LED circuit to become broken and extinguishing the 22 LEDs on that side of the unit.

Because it was a while before the fault was reported, and then a further delay while we arranged to replace it, further PCB track and component damaged occurred.

Mark

[1024MAK]

I don't have a picture of it in service, but here is a picture by the manufacturer:-

MARL LED bulkhead light fitting

IMG_3601.JPG (535.64 KiB) Viewed 15833 times

Mark

[1024MAK]

And here is what the other side of the PCB looks like, complete with the LEDs (apart from two that fell off during the cleaning)...

PCB topside

IMG_3588.JPG (953.36 KiB) Viewed 15832 times

The brown marks are because the resistors run hot.

Mark