Well, I don't think any of us plan to make an electronic rain gauge as part
of any life support system, so a little problem once in a long while is not
a concern.  If bird droppings affect the gauge, one might have to live with
it.

But on the other side, I think the wheel idea might fly.

Suppose you set up some sort of paddle wheel at the bottom of the collector
funnel?  It could trip the optical path of an optical switch.  If you use a
wide enough funnel and small/light enough wheel, you could get some pretty
fine readings.  Of course you need to protect the wheel from wind, etc.

Also, how about running a small hose from the bottom of a funnel into a
protected area (attic, etc) that is free from the effects of wind?  This
will make it harder to measure the more insignificant amounts of rainfall.

Also, there are some pretty cheap "strain" gauge semiconductors out there.
If you get one sensitive enough, you could measure the weight of collected
water, but I'm not sure how you would empty it after :)

Keep the ideas flowing.  I'd be interested in hearing what turns out to be
the best solution.

John Mullan


-----Original Message-----
From: pic microcontroller discussion list
[mailto:PICLIST@MITVMA.MIT.EDU]On Behalf Of Wagner Lipnharski
Sent: Thursday, February 17, 2000 7:03 PM
To: PICLIST@MITVMA.MIT.EDU
Subject: Re: How to measure rainfall (digitally)[OT](?)


... what is the real response for the tipping bucket?

I mean, in a torrential flash rain as we have here in Florida, that can
get one inch of water in less than 2 minutes, the 0.01" tipping bucket
would tip how many times per second?  (100 times for an inch, in 120
seconds... so one tip per 1.2 seconds? It will be precise this way? How
long it takes to tilt - empty - return to filling position? It will not
lose accuracy or linearity?)

Of course it was designed to embrace the average rain values, it should
tip some times for a very little rain so you could measure it (should
have a tinny bucket and large funnel), and it also should be able to
measure litters of water per minute... so not so tinny bucket...  To
response well for both worlds, it should sacrifice both worlds a bit,
isn't it?

If rains only 0.01", the water will be just at the funnel, not dripping
into the bucket, will evaporate and measurements would not be made.  If
rains 1 inch in 1 minute, the unit will be just overflowing.

I am not sure if calibration can solve this overflow situation
precisely.

Any technical specs from it?


Suggestions to avoid overflow:

1) Using 3 or 4 little buckets in a rotating star formation, when one
gets full tilt and pour, another enter in position to fill, so no water
is lost. A floater could be used to lock the bucket while not full,
avoiding free inertia rotation in heavy rain.

2) Using an angular mechanic valve that close the funnel output when the
bucket tilts. Like if a wheel is installed at the funnel output, a hole
at the wheel would allow water to get out of the funnel and enter the
bucket only when it is in filling position.

Other problems:
How the tipping bucket solution solves for bird droppings and other
sticking material that enters the funnel and stick into the bucket
reducing its relative internal volume? Later on when rains, the bucket
will tip with less water then the expected.  All sort of flying objects
can stuck the bucket

.