The modules sold by Maplin in the UK are made by a Chinese company Fine Offset. Models N96GY and WH1081. These are increasingly being used by Raspberry Pi users as the basis for a home made wind and rain detectors. All are cheaply made but are waterproof in that they use dry reed switches which are physically separated from the outside world. This means that insects and other wild life will find it more difficult to stop your readings. Please think carefully before oiling the bearings as this can make them "sticky" and seize up more quickly!

So now it's time to take a look inside the three most popular modules, the wind speed, wind direction and rain detector. I have included some basic notes about how spending just a few pounds their reliability and accuracy can be improved. (soon).

  Let's start with the simplest detector in the set, the rain gauge. This is simply a small balance "tipping bucket" detector where the rain runs in one side of the balance  which make it heavier causing the balance to topple over. In so doing a small magnet makes the reed switch change from off to on and then back to off sending a pulse to the control unit. As the balance has toppled over the other side starts to fill while the first empties itself ready for the next time!

There are apparently different volume cups made but we have only ever seen the one that is "calibrated"(!) to topple at each 0.011" of rain. let's say it's fairly accurate. Perhaps someone would like to stand out in the rain and count the clicks?

The main disadvantage of this sort of detector is that no rain is detected until sufficient falls to make the unit topple over the first time and at the end of the rain it may stop just before filling for the last time.  There is also the problem of morning frost and heavy rain splashing out of the collector before being recorded. Meaning that the accuracy is good but not excellent.

  Here we can see the reed switch removed from the side of the detector unit - bucket unit on the right. The black blob is a rubber spacer used to secure the circuit board a little more accurately.

It's all a bit push together and hope and it rattles about a bit!

Incidentally the top cover (not shown here) just clicks on with two plastic tabs (no screws) and we have seen these tabs broken off when the plastic breaks down and gets brittle in bright sunlight. You might prefer to glue the cover on and not be able to service or repair this unit. They are not particularly expensive to replace so having a spare "in stock" might not be a bad idea.

  Now the wind direction unit or weather vane perhaps, the design of this is quite clever!

With the base removed here you can see 8 reed switches in a circle. As the magnet attached to the weather vane on top revolves it causes either one switch to close (if the magnet is directly above the switch) or two switches to close when the magnet is between the switches.

Apply a voltage across a resistor (say 10k) and you generate a voltage depending on the wind direction. Giving 16 points of the compass from just 8 reed switches - clever!

  This is the other side of the circuit board. There are 8 resistors here, each in series with a reed switch. The physical resistors are marked with an asterisk, other values come from Ohms law. Measured values are (from North going clockwise)...
33k       *
8.2k      *
1k         *
2.1k       *
3.9k       *
16k        *
120k      *
64.9k      *
  The magnet is the small circle right at the edge of the wind gauge but as you can see there are other places that it is possible to use. Not sure why but if your gauge is inaccurate a blob of glue to hold the magnet in a different place might be worth a try..

The bearing needs discussion as whilst it has some circular bearings inside it's pretty cheaply made. Grit can and does get in, especially if the unit has been oiled by an over enthusiastic owner as this can dry and get sticky over time.

We are looking for a supply of replacement better quality bearings, more here if we find any.

  Finally, it's wind speed time using an anemometer.

Again this module uses a dry reed switch which is not in physical contact with the atmosphere, with the associated magnet fitted near the centre of the bucket unit.

  The magnet is the small metal circle just above the centre point.
  This is the dry reed switch and its cable from the wind speed detector. This cable plugs into the direction unit which then conveys the pulses off to the control box.

I don't understand the mathematics or what determines the rotation speed, but apparently a wind speed of 2.4 Km / hr causes a rotation of one per second. Your computer will have to do the maths to work the equivalent speed for Miles Per Hour and other rotations!

The reed switch is an electrical switch operated by an applied magnetic field. It was invented at Bell Telephone Laboratories in 1936 by W. B. Ellwood. (Wikipedia). This means a glass tube that detects a magnetic field and closes the contacts. They are used in lots of different things, from burglar alarms to telephone exchanges. If we know anything Mr. Elwood probably got little reward but his name on a Company patent document. As you may guess your author once invented something which is in daily use and got little reward as well!

I intend to extend this article - please look back soon.

ŠJuly 2013
Terms of use of this website
Dedicated to bringing common sense to the mobile mast debate