On Fri, 21 Jan 2005 12:35:55 -0000, "Alan Gardiner"
wrote:

With the Davis equipment I was led to believe that the stik reed switch was
a particular problem with early vantage Pro stations. From what I have seen
recently it seems to have been a problem with other models. More recent
stations may be better but I don't know if this is the case.


Just as a general comment on the reed switches in Davis anemometers
and stressing that this is my personal take on the issue and in no way
based on direct official information from Davis:

The first point is that of all the outside sensors, the anemometer
(and specifically the wind cups) is the component that in a sense
takes the most battering from the weather. While reed switch life (in
static - eg bench-mounted - devices) is indeed measured in the
millions of operations, if you do the calculation then the wind cups
also revolve and trigger the reed switch many millions of times per
year. (1 rev per second - a very low wind speed - gives 30 million
operations per year by my calculation). Factor in the vibration and
buffeting that many anemometers will experience and it's not difficult
to see why the reed switch may not have an indefinite life and may
typically need replacing after eg 5 years although the exact period is
very variable, depending on location, exposure, mounting vibrations
etc.

That said, I've come across WMII anemometers that are a number of
years old (13y is I think the oldest) and that are still providing
sound wind speed readings. But it is true that there seem to have been
a few batches of anemometers that have been more prone to premature
wind speed failure. My guess (and it is just a guess) as to the cause
is as follows: Reed switches can be quite delicate devices, eg where
the connection wires pass through their glass envelope. The switch
needs to be mounted in such a way that there is no stress on the
immediate connections, otherwise what can happen over time is that
minute cracks can start to appear in the glass envelope, which allows
oxygen and especially moisture into the sealed, dry, inert, interior
of the switch. Then, again over time, the switch contacts become
impaired and ultimately results in intermittent wind speed operation.

Now how much stress is on the switch wires will depend on the skill
and care of the operator on the assembly line. It might conceivably be
that the occasional operator over the years has not been quite as
adept at this aspect of the assembly operation as the design demanded,
(which would account for why the problem seems to have appeared in
batches). When new, the anemometer worked fine and therefore passed
all its factory tests, but ended up having a shorter life in the
field.

In the last couple of years (roughly since mid-2003 but depending on
station model), the switch design has been changed a little. AIUI the
reed switch is now mounted on a tiny circuit board, this operation
being done under carefully controlled conditions with the result that
there should now be little or no stress on the switch wires. The
little circuit board is then mounted into the anemometer in a separate
non-critical operation. While it's fairly early days yet, the
indications thus far are that the incidence of premature switch
failure is now very low.

John Dann
www.weatherstations.co.uk