Topic: Rainfall intensity

[SLOweather]

A few years ago, I developed an experimental rainfall intensity device using a microphonic Lexan plate transduder I built, the sound card input on a PC, and a strip  chart recorder program called Radio SkyPipe. It works pretty well and it was interesting to watch the bands of rain pass by on the chart.

The original thread is here: http://www.wxforum.net/index.php?topic=1428.0

I know of no PWS that has any sort of indication or measurement of "live" rainfall intensity. They all work "in arrears" determining intensity based on rain gauge tipper timing, and we all know how long it can take between tips, and how rapidly perceived ran rates can vary.

Yesterday afternoon, I got to thinking about the Hydreon RG-11 Mode 6- Drop Detection. From the manual:

The RG-11 may also provide drop detection. Use this if you want to
do your own, external data interpretation. The output will pulse
once with each detected drop. Normally, it will produce longer
pulses (in multiples of 200 mS) for larger drops. If set to multiple
pulses per drop (SW 3 ON), each detected drop will generate one
or more 100 mS pulses, depending on drop size.
In sensitive mode, the threshold for drop detection is lowered to
below the normal level. This makes the system more sensitive, but
raises the possibility of false detections. It is up to the system
designer to determine the proper tradeoff. Similarly, the Hi drop
threshold will provide an output only for large drops, making false
detections unlikely, for installations were a false detection is
especially objectionable.

It seems to me that the Dream Machine could easily monitor the relay output of an appropriately programmed RG-11 and integrate the drop detections into a qualitative,  real-time rain intensity.

[Bushman]

Such devices do actually exist, although having never heard of one either, I did have to search for it.

The Laser Precipitation Monitor (LPM) manufactured by Thies is a Present Weather
sensor based on an optical disdrometer. Thus rainfall amounts can be derived from the
disdrometric information. The factory calibration process for the volume measurement
is described and the resulting measurement uncertainty is given.

http://www.thiesclima.com/disdrometer.html

BTW, it will also do snow! 

[Weather Display]

I do know of one person using the
http://www.rainsensors.com/
and connected to a 1 wire counter, and having WD read that
then WD knows as soon as rain starts, and so can set the current weather conditions to rain

[Bushman]

I do know of one person using the
http://www.rainsensors.com/
and connected to a 1 wire counter, and having WD read that
then WD knows as soon as rain starts, and so can set the current weather conditions to rain

But that is not intensity.  For that you need a disdrometer.

[Weather Display]

intensity could be deduced by the frequency of the detected drops->tips->counts
by applying time

[Bushman]

intensity could be deduced by the frequency of the detected drops->tips->counts
by applying time

Perhaps roughly inferred, but studies on disdrometers show that even the high end units need careful calibration that your device won't support in any way shape or form.  Drop size will influence tips for example; how do you derive droplet size?

[SLOweather]

Well, the Hydreon unit uses infrared LEDs and sensors and an optical plastic dome to count drops, and somehow also relate size. the single drop counting mode can either send one 100 ms pulse per drop, a pulse in multiples of 100 ms for drop size, or multiple 100 ms pulses per drop.

That's what started this line of thought, and I have a couple of other threads running about it. One in the Hydreon area is about observations made during rain last weekend, and counting drops, to get an idea of how often it clicks in the rain. The other is in the Davis section, regarding using the input on a Davis temp station, the RG-11, and a 3v volt supply and a capacitor and a couple resistors to integrate the pulses into an analog voltage that the Davis console would then output as a temperature. That could then be converted by a script into a fairly instantaneous rainfall intensity.

[Bushman]

Well, the Hydreon unit uses infrared LEDs and sensors and an optical plastic dome to count drops, and somehow also relate size. the single drop counting mode can either send one 100 ms pulse per drop, a pulse in multiples of 100 ms for drop size, or multiple 100 ms pulses per drop.

That's what started this line of thought, and I have a couple of other threads running about it. One in the Hydreon area is about observations made during rain last weekend, and counting drops, to get an idea of how often it clicks in the rain. The other is in the Davis section, regarding using the input on a Davis temp station, the RG-11, and a 3v volt supply and a capacitor and a couple resistors to integrate the pulses into an analog voltage that the Davis console would then output as a temperature. That could then be converted by a script into a fairly instantaneous rainfall intensity.

Have you tried using the drop size mode?  I wonder how accurate it is.

[SLOweather]

I suppose I should say that, by intensity in the opening post, I wasn't necessarily looking for a definitive output in accurate inches per hour. Considering the variability over a short period of time, it would almost be meaningless.

Instead, it could be a range, or even a text value. (showers, light rain, moderate rain, heavy rain, build an ark).

When I was running tests with my flat plate microphonic transducer, a computer sound input, and Radio Skypipe for a strip chart display, one of the the coolest thing to watch was the qualitative display of intensity.

Here's a one hour trace from 2008 where you can actually see the rain bands intensify as they passed through.

[SLOweather]

No, I haven't. Mainly, I'm not sure just how to run an experiment in that mode. It might be interesting to run it in the multiple pulses per drop mode, just to hear the relay chatter.

Hmmm... It's supposed to rain this weekend....

Have you tried using the drop size mode?  I wonder how accurate it is.

[SLOweather]

BTW, the original thread for the microphonic disdrometer is here:

http://www.wxforum.net/index.php?topic=1428.0

The experiment is easy and cheap if anyone wants to try it. I bought the smallest pane of Lexan at the Depot, about the size of 8.5 x 11 paper, and made a 1x4  wooden frame for it. The plastic was mounted with double sided foam tape.



I got a piezo transducer from Radio Shack, and broke the transducer out of the plastic housing and epoxied it to the underside middle of the pane.

After that, it's just extending the cable long enough to pass through the window and plug into the mic input of the sound card.

After watching the RG-11 click, and watching the rain fall on the surface of the hot tub, I think I lucked out on panel size, anything much smaller would not intercept enough drops to be picked up by the mic, except in a downpour.

[Bushman]

No, I haven't. Mainly, I'm not sure just how to run an experiment in that mode. It might be interesting to run it in the multiple pulses per drop mode, just to hear the relay chatter.

Hmmm... It's supposed to rain this weekend....

Have you tried using the drop size mode?  I wonder how accurate it is.

I think your answer lies in titrimetry.  A few burettes with various openings would allow different drop sizes to be created (I think - it has been a while since I was in a chem lab!).  And a few burettes running simultaneously would give a rain shower.  Same or diff would give a rain shower.

[Weather Display]

I did try finding a sound component that would get the volume as a value...but never could find one
they would have volume indicators...but the actual volume was never available as a value (so that it could be plotted in WD etc)
I will have another look
as I do like what you did there SLOweather

[Weather Display]

I found a component to do the job
so I could now add support for the sound device you did into WD (eg to plot the volume like you had done SLOweather)
if anyone was interested

[wxtech]

I worked in an astronomy/physics department in the 80's.  They needed a raining warning sensor to alert the astronomer when to close the telescope dome .  We copied a rainfall detector that other astronomy departments used.  It was two tin cans; the larger was an inverted large empty tuna can open side down, over a smaller soup can, open side up.  The larger can was the one that the rain impacted.  Inside the top can was an electret microphone cemented to the can.  The top can was suspended on three posts.  The microphone was connected to an amplifier and speaker.  The astronomer would close the telescope dome when he/she heard the raindrops falling.  It was intended to be an alarm, not measure rainfall rate.
A rain intensity measuring mechanism that I saw in a met textbook used water falling through oil. 
It was a vessel containing the accumulated rainwater with an oil layer above the water.  A catch funnel formed the rain into drops which were carefully inserted into the oil layer.  A light/photocell looked at the water stream in the oil layer and showed the different transmissivity of the light as drops of water fell through the oil.  It was a rain-rate device, not rain accumulation measurement.

[SLOweather]

I found a component to do the job
so I could now add support for the sound device you did into WD (eg to plot the volume like you had done SLOweather)
if anyone was interested

Well, he!!, I'm obviously interested. I'd be pleased to test it with my flat plate transducer, and I'm a registered WD user already.

Rain's a-comin' this weekend!

[Weather Display]

cool,email me how we need to do this...i.e I guess its a channel input that needs to be monitored...i.e is it just the microphone input?
and could plot the volume on the temp/hum/rain chart, along side the rain total?

[SLOweather]

I contacted the company in Germany and got a quote on a basic unit. They are pretty proud of it, if you know what I mean:
Laser Precipitation Monitor                $3,931.47
Instrument Holder                            $ 595.44
PC-Program LNM View                       $ 794.24
Connection Cable                             $ 218.57
Fed Ex Import "Economy"                   $602.81
=====================================
Total                                             $6,142.52

(I  know, the numbers are off by a penny. That's what's on the quote...)

I like my way better.

Such devices do actually exist, although having never heard of one either, I did have to search for it.

The Laser Precipitation Monitor (LPM) manufactured by Thies is a Present Weather
sensor based on an optical disdrometer. Thus rainfall amounts can be derived from the
disdrometric information. The factory calibration process for the volume measurement
is described and the resulting measurement uncertainty is given.

http://www.thiesclima.com/disdrometer.html

BTW, it will also do snow! 

[iisfaq]

Earlier this year I puchased an Aurora Eurotech Cloud Sensor - http://www.auroraeurotech.com/CloudSensor.php

It has a rain sensor on it - I do not know what it is made from but it is very good at giving you an early warning to intensity. I use it on my site so that the icon changes and the current conditions says that percepitation is being detected if it is small and as the values go up I add that it is raining or bucketing down.



Rain sensor is the large middle item.

I put this graph together from 1 days data - the orange is the values I get back from the Cloud Sensor rain sensor. Not sure what they mean as numbers but 30 is very heavy rain and it is always sitting around 1.5 so I take it that there is rain or precipitation only when it gets to a value of 2.

The black dots at the bottom is when the davis did a bucket tip.



I am sure someone would know what the rain sensor is in the cloud sensor.

Chris

[Weather Display]

I had posted on another thread that I did add support for sound card input in WD for a rain sensor based on sound volume now
(for early rain detection (sets the rain icon/intensity, etc)