Order Tray | Contact Us | Home | SIG Lists

[wxsig] snow depth sensor update

John Bennett jabennett at insightbb.com
Sun Mar 8 22:04:14 UTC 2009


Considering it is the 8th of March, it is seriously doubtful that there
will be any more measurable snowfall, so I am bringing the sensor back
indoors until next winter.

The one (and only) significant winter precipitation event was one of the
worst ice storms on record followed by snow and very high winds. Total
accumulation of ice and snow was about six inches on the 28th of January
with most of that being ice. There were areas down in western Kentucky
without power for literally weeks. The wooded area in and around our
transmitter site (WEVV-DT) has been thinned out by about 50% due to
fallen branches and trees. This is just across the Ohio River in
Henderson, KY. In the last two weeks while doing field strength
measurements it almost looks like in places that someone took a giant
pair of hedge clippers to the trees. One road we traveled had every
power pole replaced for a distance of about 3-4 miles.

But how did the sensor do?

Before the storm, I observed that the output was fairly stable until the
temperature fell below freezing. At that point the distance started
increasing (i.e., the ground was falling away from the sensor - HI).
There is some jitter in the output, but believe that can be dealt with
in the software.

At the start of the storm it appeared that the sensor was going to
correctly report the accumulation. However, as the freezing rain and
then snow blew in on sharp NW winds, the sensor began experiencing
problems. The temperature never fell much below 32 during the weather
event, but the output was all over the place. I believe that snow and
freezing rain were blowing up into the transducer cavity partially
blocking the plastic screen on the output. I plan on finding a large
plastic funnel and using that as a shield. If the funnel is large enough
it should not interfere with the sonar pattern and yet prevent most of
the snow and ice from blowing up into the cavity (hopefully).

Another problem encountered was the stability of the snowboard. The
support structure is a problem. The structure might work if the PVC pipe
is driven down below the frost line. This would prevent the pipes from
being pushed back out of the ground. I also observed that the tile is
curling a bit around the edges, although not seriously.

Two other alternatives under consideration is a concrete slab or large
ceramic tile set with grout or concrete on top of a sand base. I think
the former might be the better way to go. If in a grassy area, such as I
where I have placed mine, it would allow easy mowing during the summer
months.

MaxBotix is coming out with a transducer that is better suited to
measuring snow depth. Using the new transducer along with the other
changes I have enumerated here will hopefully lead to a successful
implementation.

Before bringing the sensor back inside for the warmer months, I decided
to run the test program to see if the sensor was still working properly
after going through the winter. Here's the output:

------------------------------------------------------

SndCal, ver 0.1, December 2008, John Bennett, N4XI
Utility for calbrating the 1-Wire Snow Depth Sensor.

1-Wire Device Serial Numbers:
    Data port serial number:    1200000037517C26
    Control port serial number: 0F0000002245FB05

    Current: 459, Ref: 459, Diff: 0, Distance: 57", Depth: 0"
    Current: 460, Ref: 459, Diff: 1, Distance: 57", Depth: 0"
    Current: 460, Ref: 459, Diff: 1, Distance: 57", Depth: 0"
    Current: 460, Ref: 459, Diff: 1, Distance: 57", Depth: 0"

------------------------------------------------------

The correct distance is still being reported and the jitter is about
what it was before. So, the MaxBotix and sensor circuitry works fine.
Hopefully I just need to find the right transducer and protect the
sensor.

However, for now, I will have to be content to wait until next winter
before doing anything more.

73,

John
N4XI





More information about the wxsig mailing list