Rain affecting signals and sensing rain.

[richardghole@yahoo.com]

Hi

I am thinking of making a rain sensor to detect light rain as the BOM
radar is often not accurate at measuring light rain in my local area. I
have an Internet weather station linked to http://weather.org.au .

I heard some wireless transmissions can be affected greatly by rain so
if I set up a transmitter and receiver, the signal would get weaker in
the rain and the strength of it could be measured. The distance between
the transmitter and receiver could be experimented with. Perhaps I
could set up one for the very local area where the distance between the
transmitter and receiver is only about 50 or 100 meters. However,
another one could be set up so that it spans a distance of a kilometer
or more. The signal could extend between my house and someone else's
place in the direction where the prevailing weather comes from. This
would alert us when rain is about to arrive.

It would be good if the sensor detected even light drizzle. Ideally it
would be nice to see or measure the signal reduce as the drizzle or
rain became heavier between the transmitter and receiver.

Is there a simple way to do this, such as by using an infa red
transmitter and receiver? The receiver could be housed in a pipe to
block out as much background interference as possible and so it only
received signals from the direction of the transmitter.

Another alternative could be a low cost radar if one is suitable.

What sort of transmitters and receivers could be used and would the
setup be affordable?
How accurate would the system be?

Your help is appreciated,
Regards Richard.

 

[Ken Taylor]

<richardghole@yahoo.com> wrote in message
news:1142368728.491174.263800@j33g2000cwa.googlegroups.com...

Hi

I am thinking of making a rain sensor to detect light rain as the BOM
radar is often not accurate at measuring light rain in my local area. I
have an Internet weather station linked to http://weather.org.au .

I heard some wireless transmissions can be affected greatly by rain so
if I set up a transmitter and receiver, the signal would get weaker in
the rain and the strength of it could be measured. The distance between
the transmitter and receiver could be experimented with. Perhaps I
could set up one for the very local area where the distance between the
transmitter and receiver is only about 50 or 100 meters. However,
another one could be set up so that it spans a distance of a kilometer
or more. The signal could extend between my house and someone else's
place in the direction where the prevailing weather comes from. This
would alert us when rain is about to arrive.

It would be good if the sensor detected even light drizzle. Ideally it
would be nice to see or measure the signal reduce as the drizzle or
rain became heavier between the transmitter and receiver.

Is there a simple way to do this, such as by using an infa red
transmitter and receiver? The receiver could be housed in a pipe to
block out as much background interference as possible and so it only
received signals from the direction of the transmitter.

Another alternative could be a low cost radar if one is suitable.

What sort of transmitters and receivers could be used and would the
setup be affordable?
How accurate would the system be?

Your help is appreciated,
Regards Richard.

You want sensitive, accurate, cheap and simple. Any other options?

Seriously, if you want a simple way to detect rain, Google rain detector
circuit. Build a sensor (basically just a mesh of tracks on a circuit board
which get a low-resistance path across them when they get wet. Stay away
from the RF stuff, it's cheap/accurate/reliable (pick two). Note I don't
include simple. The other sort of sensor though is easy - however it will
tell you it *is* raining, not that it *will*. Therefore if you have
neighbours who are interested also you could set up a network of these in
the district and maybe use wireless IP or some such to let each other know
what's happening. That'd be a neat community project.

Here's just one page plucked at random from the search result:
http://www.electronics-lab.com/projects/science/015/

Make the sensor really big with close tracks and it becomes more sensitive
to light rain. Note though that you'll probably detect fog too. Plenty of
room for experimentation!

Cheers.

Ken

 

[The Real Andy]

On 15 Mar 2006 02:22:49 -0800, "richardghole@yahoo.com"
<richardghole@yahoo.com> wrote:

Hi

Thanks for you suggestion. I have thought of the exact thing that you
mention and I was about to get someone to make it up. However, he said
that there would be a fair bit of time and electronics involved as the
resistance of rain water would be very low and difficult to measure and
sense. Also the acidity of rain can vary and this may cause major
inaccuracies. Another way I thought of doing it would be by putting two
bits of fine mesh very close together but not quiet touching. One would
be positive and the other negative. There would be a short when the
raindrop bridged the two pieces of mesh. Also a heater would have to be
going near the system so it would dry out quickly after the rain
stopped so that the system would reset itself for the next lot of rain.


One way I thought of that may make the electronics simpler would be to
use very high voltage like on the bug zappers. This may make changes in
voltage or current larger and this may be more easily be able to be
measured. I am not sure what you think would be the best way to go if I
make up something like this.

However, even if I make the above up I would still like to be able to
try something like a lazar over a distance of about one kilometre to
measure rain that is approaching. I think this could be simpler than
the other suggestion of a network of sensors via the Internet. The
lazar would be pointed at a light dependent resister. The light
dependent resister would be housed in a tube to stop as much other
light entering as possible. When there is rain or drizzle around, it
would partially dim the light from the lazar so that the light
dependent resister would put out a lower voltage.

Another type of lazar or diode may be better than a light lazar. For
example an infa-red diode which could use resisters that respond to
infa-red instead. This may make it more accurate in the day time as
outside light may affect it less. I also thought other types of
radiation could be used such as micro waves etc.

What do you think would be the easiest way to go based on the above
suggestions?

How much electronics experience do you have?

I think the what the BOM use is probably the best solution. RADAR.

Laser doppler (as with any optical system) is no good in fog/polution.
PCB based sensors have poor response time, and when the rains stops
you need to wait for the water to evaporate. High volatage is
susceptable to moisture and high humidity.

Your help is appreciated
Regards Richard.

 

[richardghole@yahoo.com]

Hi

How much would radar cost? I think it would be very expensive?

We do not get a lot of pollution or fog in our area that obscures
visibility, so lazar may work ok. However, it would be a bit of trouble
to line everything up over a long distance.

I thought there may be some other signal that would also work which
would be easier to line up as they would have a wider beam and would be
less affected by outside radiation . However, I understand no matter
what I would use I would need line of sight

If anyone has any more ideas on this and what I stated in the previous
letter, please let me know.

I do not have a lot of electronics experience, but I know someone who
can help.

Your help is appreciated,
Regards Richard.

 

[Chris Jones]

richardghole@yahoo.com wrote:

Hi

I am thinking of making a rain sensor to detect light rain as the BOM
radar is often not accurate at measuring light rain in my local area. I
have an Internet weather station linked to http://weather.org.au .

I heard some wireless transmissions can be affected greatly by rain so
if I set up a transmitter and receiver, the signal would get weaker in
the rain and the strength of it could be measured. The distance between
the transmitter and receiver could be experimented with. Perhaps I
could set up one for the very local area where the distance between the
transmitter and receiver is only about 50 or 100 meters. However,
another one could be set up so that it spans a distance of a kilometer
or more. The signal could extend between my house and someone else's
place in the direction where the prevailing weather comes from. This
would alert us when rain is about to arrive.

It would be good if the sensor detected even light drizzle. Ideally it
would be nice to see or measure the signal reduce as the drizzle or
rain became heavier between the transmitter and receiver.

Is there a simple way to do this, such as by using an infa red
transmitter and receiver? The receiver could be housed in a pipe to
block out as much background interference as possible and so it only
received signals from the direction of the transmitter.

Another alternative could be a low cost radar if one is suitable.

What sort of transmitters and receivers could be used and would the
setup be affordable?
How accurate would the system be?

Your help is appreciated,
Regards Richard.

You know the noise that rain makes on a tin roof? Well my suggestion is to
get a tin can or plastic sheet and attach the element of a piezo buzzer to
the inside surface of the end of it or use a moving coil speaker with a
high gain amplifier, and then design some kind of filter (probably high
pass) that feeds a comparator that will trigger off the sound of a rain
drop hitting the diaphragm but not from ordinary noises.

If you can get the signal processing right then it should be very cheap and
the sensitivity could be improved by adding more of them and totalling the
number of pulses per minute.

Chris

 

[Pete]

Again, why not try out the simple sensor?

A piece of Veroboard mounted on a steep angle (so most of the rain runs
straight off) might be an easy way to prototype a simple sensor. Just
connect all the odd tracks together, and then all the even tracks
together, and (depending on the requirements of the OP), check for
continuity or measure the resistance between the pairs of tracks.

I believe Jaycar and Altronics carry Veroboard. I doubt it'd be
suitable for long-term outdoor use, but it'd be hard to beat for a
"proof of concept".

Peter

 

[richardghole@yahoo.com]

Hi

Thanks for your letter.

I plan to get some Veroboard. However, stainless steel mesh would last
longer and may be easier to mount with a very small gap between the two
pieces of mesh. Or do you think fine mesh could have other
disadvantages? I am thinking of using some with holes of about 2mm
across.

The rain I want to detect would often be before there is any runoff so
the idea about the funnel is no good. I have a tipping bucket rain
gauge that measures the intensity of heavier rain and it is graphed out
at http://weather.org.au/tolga/

I would still like to try a signal as well as it would not matter if
other things interrupted it on occasions. It would still give an idea
and would be able to detect lighter rain than the radar. If you know of
any other radiation that would be better than light to use, please let
me know.

Your help is appreciated,
Regards Richard.

 

[Jasen Betts]

On 2006-03-16, richardghole@yahoo.com <richardghole@yahoo.com> wrote:

Hi

Thanks for the letters.

We get a lot of drizzle in our area which does not get detected by
radar or audiably by sound on a roof. Also the radar covers a large
area and does not detect small amounts of rain. The frames are also
only updated every 10 minutes. However, the drizzle we get is still
enough to wet things. I am on the Atherton Tableland at 737 meters
altitude.

So is there any other signals or radiation apart from lazar light that
would have their strength reduced by such precipitation so that their
strength could be easily measured?

just measure relative humidty,

if it's 100% there's excess moisture in the air. (sleet/hail/rain/drizzle/fog)


--

Bye.
Jasen

 

[Match]

"Chris Jones" <lugnut808@nospam.yahoo.com> wrote in message
news:121h57h38tikb6@corp.supernews.com...

richardghole@yahoo.com wrote:

Hi

I am thinking of making a rain sensor to detect light rain as the BOM
radar is often not accurate at measuring light rain in my local area. I
have an Internet weather station linked to http://weather.org.au .

I heard some wireless transmissions can be affected greatly by rain so
if I set up a transmitter and receiver, the signal would get weaker in
the rain and the strength of it could be measured. The distance between
the transmitter and receiver could be experimented with. Perhaps I
could set up one for the very local area where the distance between the
transmitter and receiver is only about 50 or 100 meters. However,
another one could be set up so that it spans a distance of a kilometer
or more. The signal could extend between my house and someone else's
place in the direction where the prevailing weather comes from. This
would alert us when rain is about to arrive.

It would be good if the sensor detected even light drizzle. Ideally it
would be nice to see or measure the signal reduce as the drizzle or
rain became heavier between the transmitter and receiver.

Is there a simple way to do this, such as by using an infa red
transmitter and receiver? The receiver could be housed in a pipe to
block out as much background interference as possible and so it only
received signals from the direction of the transmitter.

Another alternative could be a low cost radar if one is suitable.

What sort of transmitters and receivers could be used and would the
setup be affordable?
How accurate would the system be?

Your help is appreciated,
Regards Richard.

You know the noise that rain makes on a tin roof? Well my suggestion is
to
get a tin can or plastic sheet and attach the element of a piezo buzzer to
the inside surface of the end of it or use a moving coil speaker with a
high gain amplifier, and then design some kind of filter (probably high
pass) that feeds a comparator that will trigger off the sound of a rain
drop hitting the diaphragm but not from ordinary noises.

If you can get the signal processing right then it should be very cheap
and
the sensitivity could be improved by adding more of them and totalling the
number of pulses per minute.

Chris

I will post an image of my data, but I'm using a leaf wetness sensor with a
heater
and it works

Match

 

[Poxy]

richardghole@yahoo.com wrote:

I have written to them. However, I am not sure if the sensor is heated
so that it will evaporate the moisture quicker. I think it would need
to be heated to give a more accurate reading of when the rain stopped.
If it is not heated the water will take a long time to evaporate and
it will still be registering rain when there is none.

True - I think it's designed to model a leaf more than rainfall as such,
although you could retrofit a heater underneath it easily enough.

So do you think these sensors will be able to tell you when there is
light drizzle and when it stops?

Again, with a heater possibly, although there may be an error when it stops,
as it might take time for the water to evaporate. That said, the unit's
output will vary coninuously with the quantity of water on the surface, so
as soon as it stops I expect you'd see some change.

Many years ago I also heard of a sensor that had some sort of salt or
mineral coating so that when it was wet it would conduct electricity
better. This could overcome the problem that water has a very high
resistance and also make the sensor more accurate.

This design is *very* sensitive - it responds to clean water.

 

[Chris Jones]

richardghole@yahoo.com wrote:

Hi

Thanks for your letter.

I have written to them. However, I am not sure if the sensor is heated
so that it will evaporate the moisture quicker. I think it would need
to be heated to give a more accurate reading of when the rain stopped.
If it is not heated the water will take a long time to evaporate and it
will still be registering rain when there is none.

Chris Jones mentioned that he has one with a heater. I would be
interested in a diagram of it when he can post it. I am not sure where
he got it or if he made it up.

Wasn't me... is this another Chris Jones?

 

[Poxy]

richardghole@yahoo.com wrote:

Hi

Thanks for your letter.

So is the sensor just a circuit board or does it have some chemical on
the surface to increase the conductivity of water?

They do seem to have a painted version, but I don't think this is to improve
conductivity. As I mentioned, I've built the circuit for another application
(moisture sensing) and it is very sensitive even to clean rainwater.

 

[richardghole@yahoo.com]

Hi

Sorry Chris. Actually it was Match who wrote the letter. It is just
that your name was near the top and quoted so I thought it was yours.
Do you know the contact details of Match, or could he email me the
information?

Your help is appreciated,
Regards Richard.

 

[Simone Merrett]

Richard,
If I were you, I would firstly get some contacts at the BOM, and ask them
about this problem. They have probably thought about it before.
As a hobbyist, cost is an issue - so whatever you try will be partly driven
by your'e expertise with the technology... (the more basic the components,
the more work required to develop interfaces.)

I like youre IR tx-rx idea the best, and I doubt you'll find many people
around the world who specifically tackled the problem.
An IR laser diode focussed to a narrow beam (or even just a laser pointer).
, and a di-chroic filter at the same wavelength for the RX (again, focussed)
so you dont get drowned out by ambient levels.

Talk to some optical people for this one.... research the technology
yourself, and try to copy a similar system (eg..lookup "LIDAR", as used for
ocean depth measurements), then perhaps contact physics dept.s or optics /
electronics suppliers for the parts. You'll need an amplifier of course, and
some basic electronics understanding, or someone willing to volunteer the
time to design it for you.
David merrett

<richardghole@yahoo.com> wrote in message
news:1143858350.216310.46400@j33g2000cwa.googlegroups.com...

Hi

Thanks for your letter.

Sorry not to reply sooner.

We had a cyclone which you should have heard about on the news.

How difficult or expensive could this be to set up? Could one of the
options you said be transmitted over a couple of miles to determine if
there is any drizzle over that distance?

Your help is appreciated,
Regards Richard.

Simone Merrett wrote:
There is another kind of senor no one has mentioned, that may be worth
considering...
VIBRATION / SOUND.
If you are only considering droplets moving largely under gravity, then
a
tallish un-resonant column with a sound-absorbtive inner surface (to
reduce
wind noise even more) may nicely complement ( a resistive sensor) , or
even
suffice as a 'drizzle' detector. It could/would work by having a
large-surface area thin 'plate' connected directly to either a piezo or
speaker-type tranducer. Fed straight into a low-noise high gain audio
amp,
the output would be super-amplifyed 'pings' of dirzzle landing on the
'rain-microphone'. The problem then of course is decoding this into
amount
of drizzle. I suggest putting output straight into sound card of PC,
simple
bit of SW would be able to separate out wind noise from droplet pings,
provided some calibration work is done before hand.
An active bandpass filter after LNA would be wise too.

In your kind of problem, it may not be a simple matter of "drizzle or
not?".
One might consider there to be a 2 dimensions of atmospheric water :
droplet
size, and droplets / volume (density) , where drizzle free, high
humidity
represents very small droplets, but a high density, and a shower in a
dry
southern area might be large drops, but low density.
I only say this, because it is likely that for good accuracy, it is
often
better to combine detections of 2 or more sensors (eg radar and
resistance)
to increase probability of correct output. (EG With resistance detector,
How
do know the difference between high humidity and drizzle? ..(without
also
knowing humidity) .)

Another thing that has not been discussed is droplet resonance, where at
a
particular radio or sound frequency, the absorbtion of energy by the
droplet
rapidly increases.
There should be science literature (esp at BOM) available that discusses
droplet size for drizzle, and any decent sound or radar book will have a
small section (or refs at least) on radio / sound properties wrt
drop-size.

If you are lucky, drizzle droplets will all be roughyly the same size,
meaning that a simple ultrasonic path-loss measurment at a narrow band
centred on the resonant frequency will reveal drizzle.

Obviously a good topic, with so many responses!
Good luck.
David Merrett.

richardghole@yahoo.com> wrote in message
news:1142368728.491174.263800@j33g2000cwa.googlegroups.com...
Hi

I am thinking of making a rain sensor to detect light rain as the BOM
radar is often not accurate at measuring light rain in my local area.
I
have an Internet weather station linked to http://weather.org.au .

I heard some wireless transmissions can be affected greatly by rain so
if I set up a transmitter and receiver, the signal would get weaker in
the rain and the strength of it could be measured. The distance
between
the transmitter and receiver could be experimented with. Perhaps I
could set up one for the very local area where the distance between
the
transmitter and receiver is only about 50 or 100 meters. However,
another one could be set up so that it spans a distance of a kilometer
or more. The signal could extend between my house and someone else's
place in the direction where the prevailing weather comes from. This
would alert us when rain is about to arrive.

It would be good if the sensor detected even light drizzle. Ideally it
would be nice to see or measure the signal reduce as the drizzle or
rain became heavier between the transmitter and receiver.

Is there a simple way to do this, such as by using an infa red
transmitter and receiver? The receiver could be housed in a pipe to
block out as much background interference as possible and so it only
received signals from the direction of the transmitter.

Another alternative could be a low cost radar if one is suitable.

What sort of transmitters and receivers could be used and would the
setup be affordable?
How accurate would the system be?

Your help is appreciated,
Regards Richard.

 

[richardghole@yahoo.com]

Hi

Thanks for you suggestion. I have thought of the exact thing that you
mention and I was about to get someone to make it up. However, he said
that there would be a fair bit of time and electronics involved as the
resistance of rain water would be very low and difficult to measure and
sense. Also the acidity of rain can vary and this may cause major
inaccuracies. Another way I thought of doing it would be by putting two
bits of fine mesh very close together but not quiet touching. One would
be positive and the other negative. There would be a short when the
raindrop bridged the two pieces of mesh. Also a heater would have to be
going near the system so it would dry out quickly after the rain
stopped so that the system would reset itself for the next lot of rain.


One way I thought of that may make the electronics simpler would be to
use very high voltage like on the bug zappers. This may make changes in
voltage or current larger and this may be more easily be able to be
measured. I am not sure what you think would be the best way to go if I
make up something like this.

However, even if I make the above up I would still like to be able to
try something like a lazar over a distance of about one kilometre to
measure rain that is approaching. I think this could be simpler than
the other suggestion of a network of sensors via the Internet. The
lazar would be pointed at a light dependent resister. The light
dependent resister would be housed in a tube to stop as much other
light entering as possible. When there is rain or drizzle around, it
would partially dim the light from the lazar so that the light
dependent resister would put out a lower voltage.

Another type of lazar or diode may be better than a light lazar. For
example an infa-red diode which could use resisters that respond to
infa-red instead. This may make it more accurate in the day time as
outside light may affect it less. I also thought other types of
radiation could be used such as micro waves etc.

What do you think would be the easiest way to go based on the above
suggestions?

Your help is appreciated
Regards Richard.

 

[Ken Taylor]

richardghole@yahoo.com wrote:

Hi

How much would radar cost? I think it would be very expensive?

We do not get a lot of pollution or fog in our area that obscures
visibility, so lazar may work ok. However, it would be a bit of trouble
to line everything up over a long distance.

I thought there may be some other signal that would also work which
would be easier to line up as they would have a wider beam and would be
less affected by outside radiation . However, I understand no matter
what I would use I would need line of sight

If anyone has any more ideas on this and what I stated in the previous
letter, please let me know.

I do not have a lot of electronics experience, but I know someone who
can help.

Your help is appreciated,
Regards Richard.

Note: it helps to include at least some of the post you are replying to

so we have some context for the discussion.

Yes, the radar will cost a lot more than the other simple methods here.
Also, you are unlikely to get it licenced. Or calibrated. The BoM has
pretty strict procedures for cal'ing radar (and everything else) so that
they get meaningful results. I also would hesitate to suggest you use
one when they are actually dangerous when used incorrectly.

A laser or similar beam is susceptible to being interrupted by animals,
birds, people, etc. Can you reliably get around that problem? Also note
that a laser powerful enough to go the distances you are talking about
is likely to be a hazard to eyeballs.

I didn't bother to take you to task earlier on your statement that their
radar is unreliable in your area - what do you mean by this? Are you
outside the area(s) of coverage, or you just feel the radar is unreliable?

Have you considered checking out the satellite coverage? Also available
at BoM's site - not as real-time as the radar though. The reason you are
not seeing the very light drizzle you want is that it is essentially
insignificant so gets averaged out. Do you really care about
precipitation in such small quantities?

Not trying to tell you how to suck eggs or anything but the simple
detector with a large fine mesh should work well enough. A high voltage
won't be of significant use though it may kill the moths.

Cheers.

Ken

 

[Poxy]

Ken Taylor wrote:

I didn't bother to take you to task earlier on your statement that
their radar is unreliable in your area - what do you mean by this?
Are you outside the area(s) of coverage, or you just feel the radar
is unreliable?

Unreliable readings from the rain radars are quite common - there's an
explanation on the BOM site explaining this. The most common issue is the
curvature of the earth and the multiple beam angles used to try and achieve
a consistent sensing altitude across the scan radius, which I think is
something like 3,000 feet. If rain is forming below this it won't be seen.

 

[richardghole@yahoo.com]

Hi

Thanks for the letters.

We get a lot of drizzle in our area which does not get detected by
radar or audiably by sound on a roof. Also the radar covers a large
area and does not detect small amounts of rain. The frames are also
only updated every 10 minutes. However, the drizzle we get is still
enough to wet things. I am on the Atherton Tableland at 737 meters
altitude.

So is there any other signals or radiation apart from lazar light that
would have their strength reduced by such precipitation so that their
strength could be easily measured?

Your help is appreciated,
Regards Richard.

 

[Ken Taylor]

richardghole@yahoo.com wrote:

Hi

Thanks for the letters.

We get a lot of drizzle in our area which does not get detected by
radar or audiably by sound on a roof. Also the radar covers a large
area and does not detect small amounts of rain. The frames are also
only updated every 10 minutes. However, the drizzle we get is still
enough to wet things. I am on the Atherton Tableland at 737 meters
altitude.

So is there any other signals or radiation apart from lazar light that
would have their strength reduced by such precipitation so that their
strength could be easily measured?

Your help is appreciated,
Regards Richard.

The problem with using path loss to detect rain is that other things

will cause path loss as well. It won't be any more dependable than the
radar. Again, why not try out the simple sensor?

Cheers.

Ken

 

[Pete]

How about a large collector (think *big* funnel, or possibly even the
existing roof drainage system) and in the output pipe, either a flow
sensor (to give a range of outputs depending on actual rainfall), or a
simple continuity test to detect rain/no rain.

Rethinks: perhaps the "dynamic range" of the existing guttering and
downpipes might be just a *little* too large between "barely raining"
and "extreme storm"

Peter