I have built an automated lawn spraying system for my backyard. A PC is used
to turn ON and OFF the water valve periodically according to a preset
schedule. The problem now is, it may sometimes be raining during, or shortly
before the spray time, which would simply waste water. Is there any
commerically available rainfall detector that could be interface to my PC? I
want to use it to stop unnecessary spraying. Or, I could build my own
circuit for doing this, does anyone get such a circuit?
Thanks in advance
John
______________________________________________________
Get Your Private, Free Email at http://www.hotmail.com
Sprague used to make a moisture detector chip, might still do. And....at
some point in the past year or so, someone had a PIC project for a plant
watering thing. Anyone remember this?
> Hi All,
>
> I have built an automated lawn spraying system for my backyard. A PC is used
> to turn ON and OFF the water valve periodically according to a preset
> schedule. The problem now is, it may sometimes be raining during, or shortly
> before the spray time, which would simply waste water. Is there any
> commerically available rainfall detector that could be interface to my PC? I
> want to use it to stop unnecessary spraying. Or, I could build my own
> circuit for doing this, does anyone get such a circuit?
Rain-Bird makes a rain detector that I use on my sprinkler system. This is
a made up of a small open top container (1.5"x1.5"x .5" deep) with two
probes that stick down into the container. It has some built-in
electronics. The way it works is simple. All the valves on the system have
a common lead to which they are all tied together. The rain-detector goes
between the controller and the valves common lead. When there has been
sufficient rain fall to cause there to be enough water in the container to
touch both probes, the circuit between the controller and the valves is
opened. The controller will go ahead and try to turn on the valves at the
appointed time, but with the common line open, there is no return path and
the valve stays off. The water in the container will eventually evaporate
enough to not be touching both probes, allowing the valves to work again.
Actually, an even better desing would be to sense the moisture in the soil. If
it rained a lot last night, you shouldn't need to rain all day, and maybe even
the next. It will also tell your system when you need to water more than usual,
and less than usual
I know you can get a pretty accurate reading by simply sticking two probes in
the soil, and measuring the resistance between them.
>
> Hi All,
>
> I have built an automated lawn spraying system for my backyard. A PC is used
> to turn ON and OFF the water valve periodically according to a preset
> schedule. The problem now is, it may sometimes be raining during, or shortly
> before the spray time, which would simply waste water. Is there any
> commerically available rainfall detector that could be interface to my PC? I
> want to use it to stop unnecessary spraying. Or, I could build my own
> circuit for doing this, does anyone get such a circuit?
>
> Thanks in advance
>
> John
>
> ______________________________________________________
> Get Your Private, Free Email at http://www.hotmail.com
Easiest solution would be to build some extra circuitry to detect moisture
and turn off power to the sprinklers independant of the PC, unless you really
REALLY need to tell the PC for other reasons.
You can make a cylinder with a "drumhead" of stretched thin
plastic (a real drumhead would work), and put a mic under it.
Amplify the mic and look for transitions.
The sound of water drops hitting the drumhead and making
it go "thump" will be much louder than any outside noise.
I'd divide time into 1 sec packets, and flag each packet that
has a transition.
If a sliding window average of the last minute has > 45 set packets,
it's raining.
For even more noise resistance, put a second mic under a piece of
thick styrofoam insulation board, exposed on the sides. If this second
mic hears something, that sample is omitted from the window. So it won't
go off on the 4th of July when neighborhood kids shoot off firecrackers.
Other way is a funnel leading to a container with a small hole in it.
Put a water level sensor (zillions of circuits for this out there) in it.
the water comes in at a rate proportional to the rain, and out at a rate
proportional to the level.
Don't forget to mount your sensor where the sprinkler won't affect it!
Now, the wierdest way -
find a weather site on the net and write some software that figures out
if the forecast includes rain for your area. If the forecast includes rain,
don't water.
Making your own page with one of those "weather underground" banners might be
an easy, controllable way to do it.
One rain measuring method we studied in a Physical Geography
class I took once is called the tipping bucket gauge. A rod is
connected to a shaft that can turn a counter and probably also
provides some detent to position it correctly. The rod is normally
vertical with a small bucket on each end. At any given time, one
bucket is pointed toward the sky to catch water and the other one is
upside down and empty. When the top bucket fills, the weight flips
the rod which increments the counter, spills the full bucket and
brings the other bucket up so it can fill. The process continues and
the counter counts the number of buckets filled. Such an arrangement
could be easily adapted to use a PIC as the counter.
We studied this briefly in class and the instructor said it was
used in areas that get lots of mist. I am not sure how well it would
work in high wind.
> Actually, an even better desing would be to sense the moisture in the soil. I
f
> it rained a lot last night, you shouldn't need to rain all day, and maybe even
> the next. It will also tell your system when you need to water more than usua
l,
> and less than usual
>
> I know you can get a pretty accurate reading by simply sticking two probes in
> the soil, and measuring the resistance between them.
>
> -Adam
[Removes Computer Science Hat, dons Botanist :-) ]
You actually need to measure *available* soil moisture. This is done using a
tensiometer. There are many commercial devices - they are used by irrigation
persons. I don't have any URLs - the last time I worked with this stuff was a l
ong
time ago and involved an LSI-11/03!
/Kevin
--
-----------
Kevin J. Maciunas Net: kevinKILLspamcs.adelaide.edu.au
Dept. of Computer Science Ph : +61 8 8303 5845
University of Adelaide Fax: +61 8 8303 4366
Adelaide 5005
SOUTH AUSTRALIA
>
> Now, the wierdest way -
> find a weather site on the net and write some software that figures out
> if the forecast includes rain for your area. If the forecast includes
rain,
> don't water.
>
Generally, given my experience with weather forecasts, I think it would be
better to water when rain is forecasted and not water when _no_ rain is
forecasted <bg>.
On Thu, 26 Aug 1999 16:04:50 -0500 Martin McCormick
<.....martinKILLspam.....DC.CIS.OKSTATE.EDU> writes:
> One rain measuring method we studied in a Physical Geography
>class I took once is called the tipping bucket gauge. A rod is
>connected to a shaft that can turn a counter...
This is a standard device that I think is still the state of the art in
rain measurement. To keep wind, etc. from affecting it the bucket
mechanism is enclosed and placed under a large funnel that forms the top
of the gauge. The one I saw had the two "buckets" as rectangular
compartments on a lever that tipped back and forth rather than turn all
the way around. Whichever bucket was on the top side of the lever would
fill with rain from the funnel, while the one on the bottom would pour
out. The lever also has a magent that operates a reed switch to produce
a pulse each time it tips. SInce the buckets were rather small and the
funnel collects a lot of rain it has good resolution. I think each pulse
was 0.01" of rain.
It seems like a relatively easy device to make or you could buy one
ready-made in various levels of quality vs price. It would be good to be
able to measure the quantity of rainfall possibly along with other
weather factors then have the computer use some model to decide how much
additional water is required.
___________________________________________________________________
Get the Internet just the way you want it.
Free software, free e-mail, and free Internet access for a month!
Try Juno Web: dl.http://www.juno.com/dynoget/tagj.
Dallas Semiconductor has a promotional kit for a weather station, with
a rain-gauge add-on. It's interfaced with a 1-wire bus, and you get
an RS232 adapter as well.
The rain gauge would be usable on its own as well. It requires assembly and
calibration, and has a .01" resolution. I have one sitting on my roof right
now, and just have to get around to doing some more software.
This method is on the right track - measure the amount of rain falling and
adjust your watering rate accordingly (your program already accounts for
seasonal variations like hot, drying winds *doesn't* it ? ;) )
Check out sites that describe rain gauges that interface to PCs, PICs, etc
and select an appropriate one. Mount it higher than your sprays for
monitoring *just* the amount of rain, or mount it under your sprays for
measuring the *total* amount of watering done (positive feedback !), or do
both !
Constantly measuring soil moisture is missleading, you may accidentally pick
a wet spot (sub soil clay sump?) or a dry spot, or your probes *may* degrade
over time. Better to use a hand held moisture detector to make lots of spot
measurements and develop a rain-to-moisture profile for various climatic
conditions throughout the year (just like golf course gardeners do).
Keep on this, it could make a good, simple, cheap product that will sell
godzillions of units to every home gardener in the world (not to mention all
the horticulturalists who *need* this information).
> One rain measuring method we studied in a Physical Geography
> class I took once is called the tipping bucket gauge.
We've got a couple of these here even simpler.
Just a small piece of tin pipe (100mm) cut in half (length wise). In the
center, is a piece of metal that divides the two halves. This all pivots
about it's center point on a pin. A small metal extension extends up
from the bucket pivot point where a small magnet is attached. Each time
one side of the bucket assembly fills it over balances, falls towards
the heavy end and empties that side. The process continues with the
opposite side filling. As the bucket assembly oscillates, the magnet
passes a nearby reed switch momentarily closing the contacts.
I had this idea simmering at the back of my head for a while now, and
was wondering if a piezo sensor could be used to detect rain drops
hitting it, or a surface attached to it. I thought it may make a rain
detector for an automatic windscreen wiper.
At 14:08 27/08/99 +1000, you wrote:
>> One rain measuring method we studied in a Physical
Geography
>> class I took once is called the tipping bucket gauge.
>
>We've got a couple of these here even simpler.
>
>Just a small piece of tin pipe (100mm) cut in half (length wise). In the
>center, is a piece of metal that divides the two halves. This all pivots
>about it's center point on a pin. A small metal extension extends up
>from the bucket pivot point where a small magnet is attached. Each time
>one side of the bucket assembly fills it over balances, falls towards
>the heavy end and empties that side. The process continues with the
>opposite side filling. As the bucket assembly oscillates, the magnet
>passes a nearby reed switch momentarily closing the contacts.
>
>
>I had this idea simmering at the back of my head for a while now, and
>was wondering if a piezo sensor could be used to detect rain drops
>hitting it, or a surface attached to it. I thought it may make a rain
>detector for an automatic windscreen wiper.
>
Just what you need your wipers to come on when you are in the car wash :)
Pegot have tried this with sensors in the windscreen, seems that they would
also need to add a dirt detector to wash the screen at the start of a cycle
and losta uder tings that they hd fuggottun
Mike Keitz writes:
>This is a standard device that I think is still the state of the art in
>rain measurement. To keep wind, etc. from affecting it the bucket
>mechanism is enclosed and placed under a large funnel that forms the top
>of the gauge.
I am glad somebody has actually seen one of these. It was long
enough ago that I forgot all the important details that would actually
make it work right.
>It seems like a relatively easy device to make or you could buy one
>ready-made in various levels of quality vs price. It would be good to be
>able to measure the quantity of rainfall possibly along with other
>weather factors then have the computer use some model to decide how much
>additional water is required.
>
>
>
> On Thu, 26 Aug 1999 16:04:50 -0500 Martin McCormick
> <RemoveMEmartinTakeThisOuTDC.CIS.OKSTATE.EDU> writes:
> > One rain measuring method we studied in a Physical Geography
> >class I took once is called the tipping bucket gauge. A rod is
> >connected to a shaft that can turn a counter...
>
> This is a standard device that I think is still the state of the art in
> rain measurement.
The Physics Department of the University of Auckland in New Zealand
designed and built a rain gauge that uses a cone and tube shaped
device that forms a drop of water of exactly the same size at its end
everytime, as the rainfall varies all that changes is the frequency
of drops forming and dropping.
I recall the drops were detected (and timed) with wires that the drop
makes contact with (this I cant remember but drop counting wouldn't be
too hard).
I know this mainly because the tech who programmed the
microcontroller for it helped me get started in micros.
I cant find the device on the Uni web site but check out..
Annie, thanks. Belive me, after a week like this I really needed this.
It took awhile to pull myself off the floor and dry the tears ;-)
I can only add; staring at the sky with an encoded wireless switch while
listening to the old band, Weather Report ;-)
- Tom
At 01:25 PM 8/26/99 -0700, Anne Ogborn wrote: {Quote hidden}
>You can make a cylinder with a "drumhead" of stretched thin
>plastic (a real drumhead would work), and put a mic under it.
>Amplify the mic and look for transitions.
>The sound of water drops hitting the drumhead and making
>it go "thump" will be much louder than any outside noise.
>
>I'd divide time into 1 sec packets, and flag each packet that
>has a transition.
>If a sliding window average of the last minute has > 45 set packets,
>it's raining.
>
>For even more noise resistance, put a second mic under a piece of
>thick styrofoam insulation board, exposed on the sides. If this second
>mic hears something, that sample is omitted from the window. So it won't
>go off on the 4th of July when neighborhood kids shoot off firecrackers.
>
>Other way is a funnel leading to a container with a small hole in it.
>Put a water level sensor (zillions of circuits for this out there) in it.
>the water comes in at a rate proportional to the rain, and out at a rate
>proportional to the level.
>
>Don't forget to mount your sensor where the sprinkler won't affect it!
>
>Now, the wierdest way -
>find a weather site on the net and write some software that figures out
>if the forecast includes rain for your area. If the forecast includes rain,
>don't water.
>
>Making your own page with one of those "weather underground" banners might be
>an easy, controllable way to do it.
>
>--
>Anniepoo
>Need loco motors?
>http://www.idiom.com/~anniepoo/depot/motors.html
------------------------------------------------------------------------
Tom Handley
New Age Communications
Since '75 before "New Age" and no one around here is waiting for UFOs ;-)
Martin, tipping bucket rain gauges are fairly standard for measuring
rainfall. The basic design is a collector that funnels the water into a
small orifice that fills the `buckets'. Simple designs use a magnet
that closes a reed switch. I designed a PIC-based weather station around
four years ago and I use a Davis Instruments rain gauge where each `tip'
triggers an RB0 interrupt. Since I live in the USA, each `tip' equals
0.01". Davis sells these for around $60 and it has proven to be very
accurate here in Oregon where we get a lot of rain. There are also kits
and plans out there.
Back to the original message, a `rain detector' is simply two probes
or a PCB with traces in a serpentine pattern.
- Tom
At 04:04 PM 8/26/99 -0500, Martin McCormick wrote: {Quote hidden}
> One rain measuring method we studied in a Physical Geography
>class I took once is called the tipping bucket gauge. A rod is
>connected to a shaft that can turn a counter and probably also
>provides some detent to position it correctly. The rod is normally
>vertical with a small bucket on each end. At any given time, one
>bucket is pointed toward the sky to catch water and the other one is
>upside down and empty. When the top bucket fills, the weight flips
>the rod which increments the counter, spills the full bucket and
>brings the other bucket up so it can fill. The process continues and
>the counter counts the number of buckets filled. Such an arrangement
>could be easily adapted to use a PIC as the counter.
>
> We studied this briefly in class and the instructor said it was
>used in areas that get lots of mist. I am not sure how well it would
>work in high wind.
>
>Martin McCormick
------------------------------------------------------------------------
Tom Handley
New Age Communications
Since '75 before "New Age" and no one around here is waiting for UFOs ;-)
> Back to the original message, a `rain detector' is simply two probes
> or a PCB with traces in a serpentine pattern.
Yep: mine consisted of a piece of Veroboard with alternate tracks commoned
together. Detects a drop of rain just fine. You do have to wait for it to
dry out after the rain has stopped, though, and it also gets affected by
morning dew.
I rather like the funnel/microphone idea. More direct.
There was a previous subject like that months ago, plenty discussed with
several very good ideas about it. The problem was to close a door over
an electronic device, and it should be done as fast as possible when
rain starts.
One of the best ideas was to use a large funnel to collect rain drops
and use a pair of led (emmiter/receiver) to "read" the drops running
down the funnel small end. The funnel would "amplify" the rain drops
quantity and the operation is quickly.
I already used a simple solution but demand analog measurements. Two
resistors (around 330 Ohms 1%) connected to +5VCC, one is encapsulated
and protected from the weather, another is also encapsulated in thin
epoxy and exposed to the rain drops. Both resistors heat with the
current, the one exposed to the rain drops goes to a lower temperature
(water dissipation), creating a difference of current between both
resistors, what triggers the electronics "it is rainning". When the
rain stops, the actual resistor heatting dryes it in less than a
minute. Both resistors should be able to receive the same air and wind
temperature.
In message <spamBeGone19990826164116.35788.qmailspamBeGonehotmail.com>, John Waters
<TakeThisOuTjohn_fm_watersEraseMEspam_OUTHOTMAIL.COM> writes
>Hi All,
Hi
>
>I have built an automated lawn spraying system for my backyard. A PC is used
>to turn ON and OFF the water valve periodically according to a preset
>schedule. The problem now is, it may sometimes be raining during, or shortly
>before the spray time, which would simply waste water. Is there any
>commerically available rainfall detector that could be interface to my PC? I
>want to use it to stop unnecessary spraying. Or, I could build my own
>circuit for doing this, does anyone get such a circuit?
Bosch produce a rain detector for use in automotive applications, don't
have a model number anymore as the company I dealt with seem to have
dropped the range.
>
>Thanks in advance
>
>John
>
>
>______________________________________________________
>Get Your Private, Free Email at http://www.hotmail.com
autocyte.com <.....dfanslerKILLspam
