Picture of Make an acoustic rain gauge (disdrometer)
For a project at Delft University of Technology (The Netherlands) we are building cheap, durable acoustic rain gauges, known as disdrometers. These devices "listen" to the rain and calculate the rain rate from the acoustic signal. But you can also listen to the signal: that way, you can hear the rain, even when inside a cubicle. A nice way to bring the outside environment into the office, without getting wet.

The disdrometer presented here is better for "listening", since this instructable competes in the "Art of Sound" contest. In a future instructable, a version that is optimized for measuring rain will be presented.

Acknowledgements: This instructable is based on the work of Coen Degen, student at Delft University of Technology, whom I had the privilege to supervise.

Example file: the example file is a recording of a rainstorm in Tanzania, made by Coen Degen.
Remove these adsRemove these ads by Signing Up

Step 1: The components

Picture of the components
For this disdrometer you need:
-a piezo electric element. These can be bought at radioshack. They should cost you next to nothing. The one used in this setup is also used in alarm buzzers. When deformed, a small voltage difference forms between the poles of the element. This will be used to turn the rain into en electric signal.
-a vertical piece of glass (ceiling windows are great) or
-a smell piece of glass mounted on a picture frame (see first picture)
-an audio cable with a mini-jack connector. (old earphones form old i-pods will do the trick)

-soldering gear (carefull: HOT, only use when you know how to!)
-superglue (carefull: superglue is irritating to the skin. Be very carefull!)
-a computer with a microphone input, or a sound system with a mini jack input (or any input, but you'll have to use a different connector above, of course)
mattinflow7 months ago

Interesting project. I note that you say...

a version that is optimized for measuring rain will be presented.

Have you got around to doing that yet?

Cool. We've been getting a lot of rain here lately, and more is forecasted, so I'll try this. And when you opened the superglue with your mouth, did you actually superglue your mouth shut?
nusnel (author)  LuminousObject5 years ago
Yes, right corner. Can not recommend the experience... had to pull my lips from each other in a quite painful way. Rolf
Superglue is very strong against tension forces, but weaker against shear forces.
So for example if you glue your fingers together, instead of pulling the fingertips apart, you can break the bond much easier by sliding your fingertips back and forward.
Part of the reason why it's a good idea to use glue AND nails to hold two pieces of wood together.
pdub77 nusnel5 years ago
Another reason for warning labels.
Purevulcan3 years ago
I think this is a cool experiment and I was going to make. I went to Radio Shack and bought the piezo element, but its in a back plastic case unlike the one shown here, should i leave inside the case or take it apart and just use the element?
drieskazoo4 years ago
Hallo Nusnel, it is now januari 2011 and I realy wonder how far You are now with the final model disdrometer and the software. I've build me Your
disdrometer and the one discribed at the Vortex Electrica site.
I like the vortex model beter because it tells me when it's raining and how strong the raifall is !
nusnel (author)  drieskazoo4 years ago
Hi Drieskazoo,

we are still not in the stage where we can publish about our final disdrometer, working on the details of how to reliably calculate rain rate from the signal, taking into account that we developed this disdrometer mainly for the african market, limiting the amount of power our product may use. The solution on the Vortex website is nice: I'm very happy people are watching this instructable and improving on it. Using their type of diode would not work well for very small drops cause of the high threshold, but in general it is a very nice approach!

The disdrometer is part of the tahmo project, more on this project in this youtube:


beehard444 years ago
i can't believe noone has ever made an 'ible on a tipping bucket rain gauge or a regular rain gauge.
ibnutty5 years ago
Interesting little device ...

This has me thinking about inserting a mike into a bottle and placing in flowing (shallow) river or rain runoff. Curious if it would pick up enough of 'babbling brook' sound.

Will test this hypothesis with experimentation.
scafool5 years ago
Superglue ( Cyanoacrylate) dissolves in acetone (Fingernail polish remover)
You should try a microphone`s Included in the volume control
moisture5 years ago
Cool idea, I'd love to see some data processing and a comparison to other measurement techniques. Post some software or filtering ideas and I'll be you (or Coen) would do quite well in the contest.
nusnel (author)  moisture5 years ago
I'd love to, but we're preparing a scientific publication on this, so I'll have to wait until that is submitted before I can post all that here. Off course, I'll do so the moment it is also published, but that will be (long) after the contest is finished. Rolf
AndyGadget5 years ago
I like the idea, but why are you using glass as the surface? I would have thought this would only transmit the higher frequencies. Have you tried other materials such as thin wood, plastic sheet etc to get a more overall sound?
(Hey . . . I'm getting an idea here - A bank of different membranes, each with a sensor, to give a disdromatic drum-kit! ;¬)
I'm with you. Initially I thought cool, it's a durable way to measure rain but then after listening to the wav, I'm not convinced glass is the best material. Sure it would be the most accurate and simulating listening to rain hit a windshield on a car but to actually listen to rain I would assume there are better materials. The roof of my old camaro was thin steel. It made a lot of noise when it rained. A drum (thin plastic stretched) seems pretty durable and more sensitive to lighter rain...
Glass is an excellent conductor of sound - one needs only use the volume to determine the size of the drops, and count to measure the total volume of water... It's not meant to be an accurate depiction of what the water sounds like...just something quantifiable.
nusnel (author)  frollard5 years ago
It off course all depends on your application: For measuring the rain rate, glass didn't work that well: the long ring (it does work like a drum, as you said) makes it harder to distinguish 2 raindrops falling close after each other. I will present a design optimized for rain measurement in a forthcoming Instructable (see other comment). For "listening" it all depends on what you want: I very much liked the comment about having multiple materials and hear the differences, glass, steel, wood etc. Thanks for all the positive feedback on this topic!! Rolf
bwalton5 years ago
So does this listen from inside? I understand you could stick it to a window but if you make it like you did can it go outside? And last of all is there a program to see how much rain is falling by using the sound file?
The glass acts like a drum - rain drops hit it and vibrate - the piezo picks these vibrations up and converts them to an electrical impulse.
DIY-Guy5 years ago
Nice! Many applications come to mind. (One idea is to use this signal to trigger a synthesizer or MIDI device to produce another sound.) Piezo element question: Is there a "best" side of the piezo element to glue to the timpanic surface? Will one side have less response than the other? If so, how do we identify the side which should be mounted to the flat surface? Thanks!
nusnel (author)  DIY-Guy5 years ago
Most piezo's have their contacts on 1 side, not both, so that leaves you with no choice. From a physical point of view there shouldn't be much difference between the two side. On the synthesizer idea: I am working on a disdrometer that has a very clear, dry sound: ideal for measuring rain. I want to make a small piece of software that calculates the rain intensity form the signal, and at the same time, plays "raindrop" sounds each time a drop is counted. Rolf
zoltzerino5 years ago
nice idea, I just listen to my conservatory though...
erosser5 years ago
Very nice, I should try some day. We definitely get enough rain here in Boston to make a drisdo-symphony....
Goodhart5 years ago
very cool
catherine785 years ago
This disdrometer really works! Excellent work guys!! Just what I was looking for... :-)
RadBear5 years ago
Could you make a surface out of several different types of materials ( glass, metal, plastic) to get a variety of sounds? Or maybe make multiple surfaces out of different materials, all with their own separate piezo sensors that would be hooked to a single input jack so you could get a collection of different sounds?
nusnel (author)  RadBear5 years ago
yeah, I think that would work in a way. You have to be careful how to connect the different piezo's together. I would recommend those "combine multiple mini-jack sources" connectors you can get at radioshack or any audio shop. Rolf
linhi5 years ago
I bought exactly the same piezo. Can you please tell me, how to solder the cables of the audio cable to the piezo element. The tin-solder doesn't hold on the piezo element! Thank you.
nusnel (author)  linhi5 years ago
Hi Linhi, strange: regular tin solder worked for us. Maybe roughing up the surface of the piezo a bit? Good luck!! Rolf
linhi nusnel5 years ago
I was too careful, roughing the surface works well. Thank you very much!
nusnel (author) 5 years ago
Hi all, thank you a lot for your feedback! The model presented here was one of the prototypes in our effort to make a cheap acoustic disdrometer. the focus of the research was to measure the rainfall, not listen to the sound of rain. The signal of this particular prototype had the typical "rain hitting windshield" sound, which we liked, but was unsuited for measuring rainfall: the "ring" of each drop was to long to be able to separate individual drops. So to answer your questions: Yes different materials would give a better sound. I think wood, sheet metal etc would all give very distinctly, different sounds, which would all be cool :-). No you can't measure the rain from the signal that THIS disdrometer gives, but: the final model of our project does allow you to do just that. I'm planning on submitting that one as an instructable as well, together with the software we use the calculate the rainfall from the incoming sound signal. Since we're still working on the last part of that project (including getting it published), this may take some time (I hope two months). Since in the final model the "sound" response is so "dry" that a human ear virtually doesn't hear anything anymore, that model would be unsuited for the "art of sound" competition. Thanks again for the feedback, Rolf PS as already mentioned in the instructable: This project is the work of Coen Degen, student at Delft University of Technology, whom I have the privilege to supervise.
graphak5 years ago
what about putting a microphone into a water proof container such as an aquarium or tupperware?