Post by thread:How to measure rainfall (digitally)

Wesley Moore wrote: >I have a friend interested in making his own logging weather station. >I was wondering if anyone had any ideas on how to measure rainfall >for such a project. Tipping bucket rain guages are very common. These basically have a funnel of known cross-section which dumps water onto a little teeter-totter like thingy with two buckets. Whichever bucket is in the "up" position catches the water falling from the funnel. Once it fills with enough water to shift the center of gravity over the pivot point, the teeter-totter tips the other way and the bucket that was "up" drops down dumping out its water. A reed switch or optical interrupter is used to count each time the bucket tips. Each tip represents a known amount of rainfall. ... I did a quick search to try and find a diagram. The site I found doesn't have a diagram but it does have a picture (the lower picture shows the tipping bucket mechanism): http://www.greenspan.com.au/rain.htm I have code for a PIC16C73 that runs at 76.8kHz (so it can go about 6 months on a 9V battery), maintains a real-time clock, and records the time of each bucket tip to the nearest second in an I2C EEPROM. It communicates with PC software over an RS232 port to offload the data. Software on the PC can analyze the bucket tip times to provide measurements of rainfall intensity as well as total rainfall amounts for each rainfall event. We used a whole bunch of these at work (hence the cheap PIC-based design) to analyze rainfall patterns over various crops in a study area. Since I did this work as part of a government-funded research project, the code and design are in the public domain. The code is here if you are interested: http://kdsl32.dnvr.uswest.net/cgi-bin/tl2.exe/RAIN I could dig out the schematics and board design too if you like. Cheers, Ken

The usual method is by means of a "tipping bucket" device. It is essentially two chambers either side of a pivot which are a lot easier(for me) to sketch than describe. A gathering funnel is placed over the pivot point and as one chamber fills, it upsets the center of gravity and tips. This places the other chamber under the funnel and empties the first chamber. This repeats back and forth and the "tips" are counted electronically. Quantitation is derived from the volume of the chambers and the area of the gathering funnel. The larger the gathering area and smaller the chamber volume, the more sensitive the device in terms of tips per unit time. A search on "rain gauge" AND "tipping bucket" should yield both commercial versions and, perhaps, a home-brew unit. Email me directly and I'll scan my sketch if you don't have any luck. Steve Kerrin (a new lurker and learner) Escondido, CA Wesley Moore wrote: {Quote hidden}> I have a friend interested in making his own logging weather station. I > was wondering if anyone had any ideas on how to measure rainfall for such > a project. > > ___________________________________________ > Wesley Moore > RMIT - BEng/BApp.Sc. 2nd Year > > .....wmooreKILLspam@spam@cs.rmit.edu.au > http://wmoore.tsx.org/

You can use a bistable bucket. This has two compartments and looks a little like this: \ | / \ | / \ | / \ | / ----------- ^ pivot | The bucket will start in one state or the other. The rain input fills the upper bucket until the quantity of water causes the bucket to tip and empty all its water. The cycle then repeats for the other bucket. If you use a sensor (reed relay, optical etc) you can could the number of bucket transitions. All you need to do is calibrate to find out how much water (mm) it takes to change the state of the bucket. Steve Wright vk2kha > {Original Message removed}

Wesley, others have already mentioned the standard tipping bucket rain collector. I decided to buy a commercial unit from Davis Instruments for my PIC-based weather station. It's very accurate. You get a switch closure each time the bucket tips which is equal to 0.01" of rain. They also have a metric version. I installed mine around 4 years ago. Davis charges $75 but if you shop around you can probably find it for $60-$65. It's used in several weather station packages from Radio Shack and others. If you want to build one, the October 1993 issue of Electronics Now has detailed plans. There is also plans for building an anemometer and a wind vane. The same author, Fascinating Electronics, now sell kits. Their rain collector kit is around $35 I think. In both cases, you need to calibrate it. Another source is Dallas Semiconductor's weather station using their IButton 1-wire devices. The basic weather station is around $80 and the rain collector is around $50 but needs calibration. This is probably the best overall package for the money. Especially for interfacing to a PIC. For more info, take a look at: www.davisnet.com/weather/products/sensors.asp http://www.gernsback.com/ www.columbia-center.org/fascinating/obs/observer.html http://www.ibutton.com/weather/index.html - Tom At 02:48 PM 2/17/00 +1100, Wesley Moore wrote: >I have a friend interested in making his own logging weather station. I >was wondering if anyone had any ideas on how to measure rainfall for such >a project. > >___________________________________________ >Wesley Moore >RMIT - BEng/BApp.Sc. 2nd Year ------------------------------------------------------------------------ Tom Handley New Age Communications Since '75 before "New Age" and no one around here is waiting for UFOs ;-)

Sorry, 1st mail didn't have a "Who" entry.... How would you like to count practically every raindrop? A piezoelectric disc is connected to a MCU's ADC (PIC of course) through an interface circuit (peak voltage limiting, variable gain amplifier and a digitally programmable filter). What it does is "listen" for raindrops. How will it know it's rain or birdseed? Heavy rain or a light drizzle? And more importantly how much rain is falling? By knowing the "signature" of falling raindrops and the amount. Use a piezoelectric disc of known area (say 1 inch dia) and calibrate a table of signatures for a known amount of rainfall. The PIC will set the interface amplifier to maximum gain and wait for any "hits" to the piezoelectric element. Once *any* hit is detected (birdseed, sand or rain), the PIC sets the amplifier gain so that the peak signal to the ADC is at say 3/4 of full scale. The PIC then sets up the programmable filter for whatever it's supposed to detect, in this case "RAIN". So by varying the gain and filters the PIC can be programmed to detect and count a specified object (sequencial hits) or to determine the approximate amount of a specified object (averaging hits). Slanting the piezoelectric disc at an angle is necessary to prevent the counted object or liquid from accumulating on the sensor. Thin aluminium, glass, ceramic, steel, teflon, etc can be attached to the piezoelectric element for wear resistance and/or electrical insulation. Benefits are, it will have no mechanical moving parts, no corrosion, adaptable to function as a rain detector, liquid flow detector or count small falling objects like screws, nuts, ball bearings, etc. A patent search i did in Dec '98 has a similiar but simple circuit to count seeds as they were discharged from a farm machine (seed sowing). I offered my improved idea for a local development grant and nothing came of it. So if this improved idea is "new", so to speak, and doesn't infringe on any existing rights. It's free to anybody that wants to use or develope it. I sure hope the piclist is used for a prior art patent search coz i'm not going to patent this and it'll be available to any and everybody that want's it. Cheers Terry At 02:48 PM 2/17/00 +1100, you wrote: {Quote hidden}>I have a friend interested in making his own logging weather station. I >was wondering if anyone had any ideas on how to measure rainfall for such >a project. > >___________________________________________ >Wesley Moore >RMIT - BEng/BApp.Sc. 2nd Year > >.....wmooreKILLspam.....cs.rmit.edu.au >http://wmoore.tsx.org/ >

How would you differentiate, say..... a frog. Exodus 8:2 -----Original Message----- From: pic microcontroller discussion list [EraseMEPICLISTspam_OUTTakeThisOuTMITVMA.MIT.EDU]On Behalf Of Terry Sent: Thursday, February 17, 2000 12:31 PM To: PICLISTspam_OUTMITVMA.MIT.EDU Subject: Re: How to measure rainfall (digitally) Importance: High Sorry, 1st mail didn't have a "Who" entry.... How would you like to count practically every raindrop? A piezoelectric disc is connected to a MCU's ADC (PIC of course) through an interface circuit (peak voltage limiting, variable gain amplifier and a digitally programmable filter). What it does is "listen" for raindrops. How will it know it's rain or birdseed? Heavy rain or a light drizzle? And more importantly how much rain is falling? By knowing the "signature" of falling raindrops and the amount. Use a piezoelectric disc of known area (say 1 inch dia) and calibrate a table of signatures for a known amount of rainfall. The PIC will set the interface amplifier to maximum gain and wait for any "hits" to the piezoelectric element. Once *any* hit is detected (birdseed, sand or rain), the PIC sets the amplifier gain so that the peak signal to the ADC is at say 3/4 of full scale. The PIC then sets up the programmable filter for whatever it's supposed to detect, in this case "RAIN". So by varying the gain and filters the PIC can be programmed to detect and count a specified object (sequencial hits) or to determine the approximate amount of a specified object (averaging hits). Slanting the piezoelectric disc at an angle is necessary to prevent the counted object or liquid from accumulating on the sensor. Thin aluminium, glass, ceramic, steel, teflon, etc can be attached to the piezoelectric element for wear resistance and/or electrical insulation. Benefits are, it will have no mechanical moving parts, no corrosion, adaptable to function as a rain detector, liquid flow detector or count small falling objects like screws, nuts, ball bearings, etc. A patent search i did in Dec '98 has a similiar but simple circuit to count seeds as they were discharged from a farm machine (seed sowing). I offered my improved idea for a local development grant and nothing came of it. So if this improved idea is "new", so to speak, and doesn't infringe on any existing rights. It's free to anybody that wants to use or develope it. I sure hope the piclist is used for a prior art patent search coz i'm not going to patent this and it'll be available to any and everybody that want's it. Cheers Terry At 02:48 PM 2/17/00 +1100, you wrote: {Quote hidden}>I have a friend interested in making his own logging weather station. I >was wondering if anyone had any ideas on how to measure rainfall for such >a project. > >___________________________________________ >Wesley Moore >RMIT - BEng/BApp.Sc. 2nd Year > >@spam@wmooreKILLspamcs.rmit.edu.au >http://wmoore.tsx.org/ >

> > How would you like to count practically every raindrop? > > > What it does is "listen" for raindrops. How will it know it's rain or > birdseed? Heavy rain or a light drizzle? And more importantly how much rain > is falling? >..... > Benefits are, it will have no mechanical moving parts, no corrosion, > adaptable to function as a rain detector, liquid flow detector or count > small falling objects like screws, nuts, ball bearings, etc. > There are devices that measure rain by shaping every drop into the same size and the Frequency of the drops coming off the bottom of this plumb-bob looking device is proportional to the actual rainfall. I know this because the chap who taught me to programme microcontrollers was working on the controller for such a device and I have found the units for sale on the web..cant remember where though. _____________________________ Lance Allen Technical Officer Uni of Auckland Psych Dept New Zealand http://www.psych.auckland.ac.nz _____________________________

Not only that, but you could pipe the sound into your bedroom, put it through a chorus effects processor, and listen to the rain outside... -Adam Terry wrote: {Quote hidden}> > Sorry, 1st mail didn't have a "Who" entry.... > > How would you like to count practically every raindrop? > > A piezoelectric disc is connected to a MCU's ADC (PIC of course) through an > interface circuit (peak voltage limiting, variable gain amplifier and a > digitally programmable filter). > > What it does is "listen" for raindrops. How will it know it's rain or > birdseed? Heavy rain or a light drizzle? And more importantly how much rain > is falling? > > By knowing the "signature" of falling raindrops and the amount. Use a > piezoelectric disc of known area (say 1 inch dia) and calibrate a table of > signatures for a known amount of rainfall. The PIC will set the interface > amplifier to maximum gain and wait for any "hits" to the piezoelectric > element. Once *any* hit is detected (birdseed, sand or rain), the PIC sets > the amplifier gain so that the peak signal to the ADC is at say 3/4 of full > scale. The PIC then sets up the programmable filter for whatever it's > supposed to detect, in this case "RAIN". > > So by varying the gain and filters the PIC can be programmed to detect and > count a specified object (sequencial hits) or to determine the approximate > amount of a specified object (averaging hits). > > Slanting the piezoelectric disc at an angle is necessary to prevent the > counted object or liquid from accumulating on the sensor. Thin aluminium, > glass, ceramic, steel, teflon, etc can be attached to the piezoelectric > element for wear resistance and/or electrical insulation. > > Benefits are, it will have no mechanical moving parts, no corrosion, > adaptable to function as a rain detector, liquid flow detector or count > small falling objects like screws, nuts, ball bearings, etc. > > A patent search i did in Dec '98 has a similiar but simple circuit to count > seeds as they were discharged from a farm machine (seed sowing). I offered > my improved idea for a local development grant and nothing came of it. So > if this improved idea is "new", so to speak, and doesn't infringe on any > existing rights. It's free to anybody that wants to use or develope it. I > sure hope the piclist is used for a prior art patent search coz i'm not > going to patent this and it'll be available to any and everybody that > want's it. > > Cheers > Terry > > At 02:48 PM 2/17/00 +1100, you wrote: > >I have a friend interested in making his own logging weather station. I > >was wondering if anyone had any ideas on how to measure rainfall for such > >a project. > > > >___________________________________________ > >Wesley Moore > >RMIT - BEng/BApp.Sc. 2nd Year > > > >spamBeGonewmoorespamBeGonecs.rmit.edu.au > >http://wmoore.tsx.org/ > >