W-8681 wind direction indicator disassembly
As mentioned in the update to my review of my Watson W-8681 wireless weather station about 2 years after first setting the unit up the wind direction indicator stopped working. As I was soon to move house I didn’t investigate at the time.
When I came to set up the weather station at my new home I found that the wind gauge was quite stiff to turn. It was not obvious how to remove the top section which carries the vane even after removing the components from the lower section and I decided to risk breaking it by gently prising the vane up from the base using a couple of table knives.
It turned out this is a good way to remove it as it is just an interference fit on a small cage bearing which it turn is an interference fit on a plastic spigot on the lower section.
The magnet which operates reed switches in the lower section can be seen on the rim of the vane under the pointer.
The problem was that the small bearing cage had rusted. I squirted some WD40 into it several times while turning the cage until the oil running out looked reasonably clean then squirted some lithium grease in. I don’t know how long this will last.
I assumed at the time that this would fix the problem but it didn’t. Dismantling it again I found that in fact I’d damaged the cable while cutting off the zip ties I’d used to fix it to the mast and not noticed. Repairing the cable fixed the gauge. I took some pictures of the innards too.
Underside view revealing the socket for the wind speed cable and the three small fixing screws which hold on the lower cover.
After removing the lower cover a circuit board carrying eight reed switches is revealed. It is held in place with a further three screws. You should note or mark the orientation of this board before removing it as the N S E W markings on the cover will be incorrect if not correctly reassembled.
A top view of the circuit board showing the surface mount components (probably resistors).
As time was short I didn’t investigate the circuit board to discover what those surface mount components are or how the gauge manages to indicate 16 directions from only 8 reed switches – it may be that the magnet in the vane is strong enough to switch the two reeds nearest to it but that implies some pretty impressive tuning of the power of the magnet and I’m more inclined to think it works some other way. Maybe next time I have to take it apart I will have found the handy little socket I have that takes modular jacks and has wires coming out of the back which can easily be connected to a multimeter so I can investigate it without needing four hands.
After the mechanical repair I never gave this matter another thought but I’ve just received an email from Darren Worley who came across this page and took the effort to trace out the circuit board tracks in the picture and confirm that the surface mount components are resistors. I’ll just quote part of his email on his take on how this works which I’m sure is correct…
“FWIW, I looked at the picture you took, and determined that the devices are in fact resistors, and the cct layout is 8 reeds in parallel. Each resistor would need to be a different value, and a combination of two resistors in parallel gives the extra 8 values of position when the magnet enables two reed switches (magnets can trigger two ‘nearfield’ reeds). This gives you 16 different resistive values across a two wire interface.”