Check the site below. He made a digital rain gauge and circuit that measures rainfall by counting the number of raindrops dripping from a funnel for each 0.01 inch of rainfall.

I am trying to figure out how to make a rain drop counting gauge and not use an electro-mechanical counter as shown in the website
Dave

Homemade DIY Weather Instruments

 

Welcome to AAC

Interesting idea, but needs updating with a better sensing arrangement. The electronic counter side of things is easy to implement and there are many options..

Here's a simple one from Amazon:


It will work off a 6v or higher supply, and needs a normally closed 'switch' to ground and counts each time the 'switch' is opened. Replace 'switch' with some form of drop sensor and you've got rain gauge of sorts.

Of course a better gauge would display in actual inches, millimeters, liters/sq m, etc. and there are other ways to capture rainfall amounts, eg by volume or by weight. All of which could be relatively easily implemented.

 

Counting drops is a fun challenge BUT also not very accurate. Rainfalldensity certainly looks variable if you just watch for a while.
Irving is corect!Also, what matters is the total per unit area. The more accurate annd stable schemes use a variation of the tipping bucket device.Those can be very repeatable and quite accurate. And counting is simple.
The most durable sensor uses a coil and magnet to generate a pulse with every tip. No contact and no friction, and only an occasional cleaning.

 

The most durable sensor uses a coil and magnet to generate a pulse with every tip. No contact and no friction, and only an occasional cleaning.

Or, easier to interface, a magnet and hall sensor.

 

Typical 'tipping bucket' continuous measurement design. There are multiple sense options: reed switch/magnet, magnet/hall sensor, opto-interruptor/vane (my preference as its much easier to implement physically and has no dependence on distances/position).

 

Igot a smaller version of one like that. The only pain is that it is RF linked and uses a weird battery. So what I need to do is convert it into a wired version. The fallacy of wireless links is that they constantly require, and use, power, which always seems to be batteries.
AND, is that "200" dimension in mm, or cubits, or inches?? or possibly furlongs??

 

AND, is that "200" dimension in mm, or cubits, or inches?? or possibly furlongs??

Its 200mm,/8", a common size for rain gauges. I borrowed the illustration from here

 

OK, all I need to do is determine a way to sense the tips and add a counter. And mount it up on the roof and run wires inside. No problem. BUT first I need to measure how much rainfall per pulse.

 

I have made a tipping bucket rain gauge in the past.







There are many ways to detect when the bucket tips.

 

BUT first I need to measure how much rainfall per pulse.

The commercial versions use a plastic or thin ally sheet 'cups' which come in various sizes, typically 0.1mm, 0.2mm or 0.5mm based on a 200mm dia funnel.

By my reckoning a 200mm dia collector is 0.2^2 *π/4 = 0.0314m2, is 31.4ml. That is, a rainfall of 1mm/hr = 1l/m2/hr is 31.4ml/hr. So construct a double sided bucket in 1mm sheet as shown by @MrChips above 15cm long, 2cm high in the middle and 2.5cm wide, that's internally 15 * 2.3 * 1 = 34.5ml and it'll tip twice an hour for rainfall around 1l/m2/hr. Calibrate by filling each side with 15.7ml from a syringe and adjusting the starting point for the tip with the screws under each side as per original diagram.

 

I used thin sheet brass to make the see-saw bucket. The centre divider is rectangular piece that is soldered at the middle. A piece of brass wire is soldered on the underside to create the pivot.

 

I am fortunate to have been given a tipping bucket sensor that suffers from a failed electronics portion. So now I need to add a counter with a display, and determine the rainfall amount per pulse. So this will require some accurate water volume delivery experimenting.

 

How about trickling water into it, count 100 tips and measure the total water volume?

 

Theoretically, the smaller the cup, the faster the tip rate and therefore better accuracy/resolution. There must be some limit to tip rate above which overflow occurs. That will depend on mass of cup, friction of hinge, input & output flow rate and probably a few other parameters.

 

Certainly, at some point in the reduction of the size of the volume of water per tip-cycle, there is a point at which fluid kinematics and friction effects become an accuracy reduction issue. The question becomes one about accuracy versus resolution. AND, certainly DYNAMIC RANGE will enter the discussion as well. The easy answer is multiple rain gauges of different ranges and resolutions, and possibly of different technologies.

 

Counting drops is a fun challenge BUT also not very accurate. Rainfalldensity certainly looks variable if you just watch for a while.

Counting drops in free air is worthless, but counting drops dripping out of a collector is very accurate. One of the very few things I remember from high school chemistry is that there are ten drops per cc. My guess is that the NWS has a more refined opinion. Anyway ...

The tipping bucket approach has always seemed gimmicky to me; I'm surprised to see it is still used. Years ago I played around with an optical detector to count the drops dripping out of the collection funnel, a variation of my 9th grade science fair project. Today, a kinetic variation:

I have no idea if this is an original idea, but here is an alternate thought. Take a bare piezo disk, like one removed from a beeper, put it under the funnel, and let the drops hit it. You get one small voltage spike per drop. The greater the distance between the bottom of the funnel and the disk, the larger the voltage spike (thanks, gravity).

Disk / high impedance amplifier circuit / filter / pulse former / counter / conversion factor / display

Now that I've written it out, it is clear that everything after the amplifier circuit is identical to the tipping bucket method.

ak

Fun variation: the pulse former output drives an integrator. Scale factor the analog output voltage to drive a simple voltmeter. No digital conversion factor calc.

 

Drop counting could work until a rainstorm produced a steady stream.

 

Drop counting could work until a rainstorm produced a steady stream.

That's one advantage of the tipping bucket. For a drop counter, you could have a group of two or three with different size funnel apertures.

ak

 

Of course, it would certainly be possible to have two tipping bucket sensors in sequence, the first one with smaller buckets, the second one with buckets 20 to 50 times greater volume. Not only would that greatly expand the dynamic range, it could also provide a continuing performance verification.