h a l f b a k e r yThe embarrassing drunkard uncle of invention.
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
Inspired by [Maeljin]'s "Piezo rain plate" [link].
The proposed instrument will have all the standard features of a home weather station, and will use a piezo-electric diaphragm to record rainfall.
This will work on the principle that the force of impact of the rain drop can be recorded
as an electrical impulse created by the piezo-electric sensor (slightly angled such as to allow run-off of the drop).
Since the raindrop will most generally fall from a height sufficient to allow the rain drop to reach it's terminal velocity, the instrument will use that value (calculated internally by using the current temperature and barometric pressure sensed by the instrument, and altitude *settable by user) to calculate accumulated rainfall.
Also, the piezo-electric diaphragm is acting as an analogue sensor, so partial rain drops which land about the perimeter of the sensor are recorded accurately as well.
There can be a remote resettable display positioned indoors, which will display current values along with accumulated rainfall since the instrument was last reset.
I'm probably overlooking something here, as I put this together rather hastily. Your comments are appreciated.
[Maeljin]'s rain plate
http://www.halfbake...iezo_20rain_20plate
[X2Entendre, Oct 04 2004, last modified Oct 05 2004]
[link]
|
| |
I don't see how this could work. Rain drops vary greatly in size and rate of fall. |
|
| |
I did not see any reference to using this as a rain gauge, [bliss]. Nor did a quick Google produce anything closely or directly related.
[Waugs], as the sensor is analogue,and records the actual force of impact, won't this help to allow for that? Granted, I am not certain how much variation there will be in TV, dependent upon the dimensions of said rain drop. I would be happy to receive discussion on the aerodynamic properties of rain drops relevant to this application. |
|
| |
Your impact sensor will record drops of so many different sizes traveling at so many different speeds, I don't see how it could extrapolate an estimated rainfall. Some will be accelerated by wind... some will be slowed by wind too. Plus, once the device is covered with water, new drops will be impacting on top of already-fallen drops, dampening their force somewhat. |
|
| |
There are self-emptying gauges in existence, if this is an attempt to solve that problem. |
|
| |
// Rain drops vary greatly in size and rate of fall // |
|
| |
Apply a bizare and esoteric technique known as "averaging" ? |
|
| |
How can you average when you don't know the values you have to average? This device will record a force impact - it cannot know if it's a small drop traveling very fast, a smaller drop traveling even faster, or a large drop traveling normal speed, or slow. It would have to know this in order to average. And if it did know, it wouldn't need to average. |
|
| |
You can't just presume and average for a whole storm, because the whole storm might be made up of slow moving big drops. |
|
| |
Shortly after posting this, I considered the effects of wind. Is there a practical solution (possibly using the average effect on the rain drop in a certain velocity of wind)? or should I delete this idea? I'm aware that certain barometric pressure sensors are piezo-electric, but to measure rainfall in a similar manner would require a bucket. |
|
| |
I can't think of a way you could get this to work, no. Don't delete it though - maybe someone else can think of something. |
|
| |
Could you use an ultrasound or microwave detector to measure the size and velocity of the drops as they pass through the guage? |
|
| |
Infra red transmissive detection might be better. |
|
| |
Oh my... I fear that soon this won't fit atop the fence in my back yard without some clever reinforcement.
Another thing occurs to me... what the hell am I doing talking about recording rainfall when I live in a desert? : ) |
|
| |
You could in theory use video cameras and 3D image sensing software to find the dimensions of each raindrop as they fall through a given volume (allowing for deformations due to gravity and rain, and perhaps position it in a place where you wouldn't get splashes off other surfaces. |
|
| |
I wonder if (depending upon the sensitivity of the piezo-electric sensor) we could add some sort of tube above the sensor, allowing the sensor to read the change in air pressure induced by the rain drop entering the chamber, and the time duration prior to impact. This would (maybe?) allow you to measure the volume and velocity of the rain drop somewhat accurately, no?
But then again, if this were added, there would be no need to stick with the original system of measurement, as volume and velocity measurement would be sufficient to record accumulated rainfall. Any input on the feasibility of this? It's a bit out of my area of practice. |
|
| |
I was thinking of a cylindrical detector that measured the volume of the rain that fell through the centre. |
|
| |
How about <engage rube goldberg mode>a funnel that feeds the accumulated water into a sloped tube. The water collects in the tube and then drips, from a known height and in a wind-protected column, onto your detector.</goldberg> |
|
| |
Actually, I kind of guessed you might be living in a desert when your idea suggested collecting data on individual raindrops. Know what you mean... *name* them thar suckers, timestamp 'em, record their vital stats. |
|
| |
blaise, I would dispute your 'baked' assertion with that patent link. That device merely determines _if_ it's raining. It does not gauge the rainfall amount, which is what this idea intends. |
|
| |
Assign each raindrop an ipv6 address and a very small RFID tag that broadcasts the address. Upon impingement with, or near, the raingauge, the recovered IP address is used in a query to recover the properties stored in the raindrop database for that address, in this case, the water volume of the particular raindrop. |
|
| |
The volumes are aggregated over time and divided by the number of recorded droplet impacts to calculate the total volume of rainfall. |
|
| |
SSI's FP2000 uses IR to determine precipitation type by looking at the frequency spectra. Eg rain, snow, freezing rain, and even rain drop size. You might be able to add precipitation identification to the piezo system to make accurate measurements |
|
| |
The bucket method is nice and simple so I guess that the point of this idea is to do the measuring in an extremely clever and cool way? Well, I'm hip to that beat daddy-o. If you want an amazingly obtuse way to measure rainfall then don't collect or measure thae water in the droplets. Collect the little dust particles at their centre. I bet some clever person (i.e. not me) could calculate the rainfall by measuring the density of these particles that collects on your plate over a period of time, especially if you combine this info with other information collected by your weather station such as humidity, temperature etc. |
|
| |
I think it's starting to converge here. Somewhere up in the Pacific Northwest, [bristolz] is going to engage in ipv6 (ipv12?) cloud seeding. Then, the rest of us (at least the continental US) will sample, extract and return these RFID tags, much in the same way that one would return an ID band from a bird. Knowing which droplets fell through our sensor, we can extrapolate to find the entire water intake of our respective properties. The piezo sensor helps form an accurate picture of water collected by particular populations of tags. In the end, we receive a much more detailed picture of how water flows in our weather systems. And we also receive a bill from [bris]. |
|
| |
I'll bill in advance. [lurch], what's your mailing address? |
|
| |
[waugs]: Just by the by, there are statistical methods for 'averaging' a data set within a certain probability of error - although I admit that aren't appropriate here (probably, 95% sure). |
|
| |
The frequency of the sound produced by a drop striking the sensor would be directly proportional to its velocity, and inversely proportional to its diameter. This is because for a given velocity, there would be a longer period between the first contact of a larger drop and when its top surface reaches the sensor than for a smaller drop. One can therefore calculate the volume of water in the drop that made a sound from the frequency and the total energy of its impact.
This method might not be so simple though, for multiple drops striking at the same time.
One could alternatively have two duplicate sensors, one placed a known height (a meter or two) above the other (and off to the side to avoid the higher one interfering). The difference between the readings of cummulative impact energies would indicate the average nearness to terminal velocity of drops falling in the area. Hopefully then, statistics gathered in this way could correct for errors an individual sensor would make. |
|
| |
// I'm probably overlooking something here // |
|
| |
That this is baked, and called an impact disdrometer? |
|
| |
(Aside: I don't like the word 'disdrometer', because
apparently it has a graceless etymology: 'drop size
distribution' -> 'DSD' -> 'disd'? -> 'disdrometer'. I much prefer
the term 'spectropluviometer'.) |
|
| |