JIMSUN WMR-918/WMR-968 Rain Gage Calibration [back] [home]

Copyright, License, and Disclaimer

Checking and Adjusting WMR918/WMR968 Rain Gage Calibration
Copyright (C) 2001-2004 James S. Seymour 

This information is free; you can redistribute it and/or modify it
under the terms of the GNU General Public License as published by
the Free Software Foundation; either version 2 of the License, or
(at your option) any later version.

This work is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
GNU General Public License for more details.

You may have received a copy of the GNU General Public License
along with this work; if not, write to the Free Software
Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.

An on-line copy of the GNU General Public License can be found
here.

Lest these pages be construed as an endorsement for these products, let me make one thing abundantly clear: I definitely do NOT recommend you buy the Oregon Scientific WMR-968. In my opinion the build quality is poor. I found the anemometer (wind speed gauge) to be worthless. Both my indoor and outdoor hygrometers (humidity gauges) were inaccurate and very slow to respond. I say "were" because, after only seven (7) years, the outdoor temperature & humidity sensors up & died. Replacing that sensor/transmitter group is too expensive to justify--particularly since, from all I've read, the manufacturer hasn't bothered to address a single one of its deficiencies in those seven years. The rain gauge transmitter was marginal at well under the claimed maximum transmitter distance. That, too, died about a year after the outdoor thermometer & hygrometer. If you're at all interested in a quality, dependable system that will give you something resembling useful data, I suggest you look elsewhere. But if all you want is a neat toy that'll last a few years, maybe this is for you.

	Checking and Adjusting WMR918/WMR968 Rain Gage Calibration
			    by Jim Seymour


A number of folks have reported inaccurate readings (sometimes wildly
inaccurate!) yielded by their Oregon Scientific WMR-918 or WMR-968 rain
gages.  This FAQ puts together information, gleaned from several sources,
on how to check and, if necessary, adjust your rain gage.

How It Works

    The way the OS WMR-918/-968 rain gage works is by funneling collected
    rain to land on the pivot-point of a two-sided "bucket."  When one
    side fills, it pivots down--causing the "bucket-tip" to be recorded,
    emptying the bucket and bringing the empty side under the collection
    point.  Each bucket-tip is *supposed* to occur when 0.04 inches (1 mm)
    of rain has been collected.

Problem #1: Unequal "bucket-tips."

    One problem is that, as shipped from Oregon Scientific, some rain
    gages have been found not to tip equally on each bucket.  The problem
    with this may not be apparent when there's heavy rainfall--the unequal
    tips may average each other out.  But light or slow rainfall may
    result in inaccurate rate and total readings.

Solution #1: Equalize the "bucket-tips"

    Posted to CAIS' WMR-918 Wireless Weather Station Bulletin Board
    (http://www.weatherwatchers.org/bbs/wmr-918board/) by Robert Nicholas
    on June 07, 2000:

	When I measured the water required to tip the buckets, one took
	quite a few more drops than the other, even though the gauge was
	level. By adjusting the screw under the buckets you can calibrate
	the gauge.  raise the screws to make the gauge read higher: lower
	them to reduce the readings. Note that the bucket that is up is
	adjusted by the screw under the one that is down. It's also
	important that it takes the same amount of water to tip each
	bucket. I used a turkey baster and counted the drops for this. A
	dab of silicone rubber is a good way to fix the screws and easily
	removable if future adjustment is required.

Problem #2: Inaccurate Measurement

    Even if the bucket-tips are equal: if they're tipping too soon or too
    late, the gage will measure too high or too low (respectively).

Solution #2: Calibrate the gage

    Posted to CAIS' WMR-918 Wireless Weather Station Bulletin Board
    (http://www.weatherwatchers.org/bbs/wmr-918board/) by "Jeff" in North
    Central Ohio on June 08, 2000:

	I calibrated my unit by computing the area of the collection
	funnel (78.54 sq. cm.); therefore 1cm. of rainfall would equal a
	total volume of water collected of 78.54 cc.

	Since each tip of the bucket is equal to 1mm of rainfall, each
	bucketful should be 7.854 cc.

Conversions:

    Not many of us will have the tools on-hand to measure-out 7.854cc of
    water :-).  U.S. Americans are likely to have measurement devices in
    cups and multiples-of-a-cup handy.  Europeans and others: liters and
    milliliters.  So...

	7.854cc (ml) == 1mm rain == .004in rain

	250ml  ==  31.83mm rain  == 1.25in rain
	500ml  ==  63.66mm rain  == 2.51in rain
	  1l   == 127.32mm rain  == 5.01in rain

	1 cup  == 236.588 ml
	1 qt   == 946.353 l

	1 cup  ==  1.19in rain ==  30.12mm rain
	2 cups ==  2.37in rain ==  60.25mm rain
	1 qt   ==  4.74in rain == 120.49mm rain

    Robert Nicholas submitted an easy-to-remember set of numbers:

	1 inch rainfall == 199.5 ml - darn close to 200 ml.

Procedure:

    Here's how *I* would go about it:

	1.  Find out if the gage is accurate in the aggregate--i.e.:
	    regardless of whether the bucket-tips are equal or not.  Do
	    this by feeding it enough "rain" so that any unequal bucket
	    tips average out.  At least the equivalent of 1 inch.  Two
	    would be better.

	    ----------------------------------------------------
	    Note: When you're doing this step, and step 3, don't
	    pour the water in too rapidly!  You're trying to
	    simulate rainfall, not a tidal wave :-).  If you
	    pour too quickly, you may loose accuracy due to
	    water over-flowing the bucket divider as it tips
	    back-and-forth.  (Don't know what this says about
	    the unit's accuracy in a for-real deluge.)
	    ----------------------------------------------------

	2a. If the aggregate is accurate, but the bucket-tips are
	    uneven: equalize the bucket-tips by adjusting one up and the
	    other down--in tandem.  Go to step 3.

	2b. If the aggregate is high or low: first see if one of the
	    buckets is tipping sooner or later, respectively, than the
	    other and first adjust *that* bucket to be even with the
	    other.  Go to step 3.

	2c. If the aggregate is inaccurate, but the bucket-tips are
	    equal, adjust the buckets to tip sooner or later--each by
	    the same amount.

	3.  After the bucket-tips are equal: check for accuracy (as in
	    step #1) and re-calibrate if necessary.  (As in step # 2c.)

    Perform these steps with the funnel in place.  You'll hear the
    bucket-tips.  And, as Robert notes, this will lead to a more
    accurate simulation of the unit in rainfall.

Tips, Tricks and Other Miscellania

    From Robert Nicholas:

	I would like point out a real accurate way to equalize the
	buckets. You can count drops of water from an eyedropper or a
	turkey baster, but this is very tedious as it takes many drops. I
	recommend you find a small container or hollow object (I used a
	very small test tube) who's volume will fill the buckets to the
	threshold of tipping, then counting the drops needed to complete
	the tipping. It's a real time saver.

	[For the "high-volume" accuracy tests] I measured the water out,
	then used a squeeze-type squirt bottle (like an empty gator-aid
	one) to slowly add it: Make sure you don't squirt directly into
	the collection hole, or the force of the stream may contribute to
	the bucket tip!

	I also would like to point out that I think the water drop
	leveling gauge built in the the gauge is cumbersome to use:  Just
	uses a small spirit level on the top of the cover. Much easier!!!

	I noticed that the rain total shown in the main display does not
	always increment in steps of .04. It probably uses metric units
	internally and converts to inches, hence the occasional step
	variation due to rounding. For example, my unit will read 1.02
	inches, which isn't even a multiple of .04.

	Real perfectionists may also want to remove the cover and make
	sure that [very little or no] amount of water remains in the
	bucket that is the last one up!!

	Also, when adjusting the screws I found it was easier to carefully
	remove the factory plastic thread lock completely (with a small
	screwdriver) than try to force them to turn.

	[Note from Jim Seymour: I removed the stuff they used to lock the
	screws on my rain gauge--which appeared to be clear RTV "caulking"
	material--by grabbing it with a pair of hemostats and pulling it
	off.]

Acknowledgments:

    This FAQ is little more than consolidating the work and ideas of
    others.  So thanks and a tip o' the hat to:

    Robert Nicholas for most of the material!

    "Jeff" in North Central Ohio for the calibration volume work.

    And, of course, Jessie Ferrell of CAIS, who so kindly provides the
    bulletin board system upon which I found posted the articles that
    comprise the core of this FAQ.

    I followed a procedure similar to that detailed under "Procedure",
    above.  It seemed to work out well.

    Jim Seymour
    Mar. 10, 2001


Comments or Questions?
Created: 13 Jan, 2001 / Last updated: 29 June, 2008 [100% MS Free Site]