Post by thread:"Plant out of water project"

I have an (old) idea for a new pic project. I would like to attatch this device to a plant and make it beep when the plant needs water. I was thinking about using : 1 3v litium batt 1 PIC12c508 1 Pizo-beeper 2 metal pins pluged into the plants pot (or what ever you call it) I would like to make some kind of A/D converter, and as far as i can recall it can be done with a cap and a res. And something with pulsing an output and reading an input.I was thinking that the water alarm should NOT beep during the night, messured either by an internal clock or by a photosenser. Dryness alarm should be dynamic. When this project is done it will ofcause be public. Regards Benjamin Petersen

Make sure you use AC to measure the resistance. The probes will corrode very quickly otherwise. Maybe a pushbutton to calibrate the device for wet and dry states? How about making the probes on flying leads so that you could connect several plant pots in series? Regards Mike Rigby-Jones > {Original Message removed}

I used such a thing, several years ago, but with some CD4000 glue : - 1st : generate a square wave (# kHz) - apply it through a capacitor (-> no continuous voltage applied so no electrolytic corrosion) to a "combed-shape" printed circuit - read the rectified AC voltage (through a diode doubler) - test (via a Schmitt trigger) if higher than the threshold - That's all It is installed on my rough-top for several years to close my roof-window if it begins to rain (very usefull in summer if a storm begins during our sleep :-) It should be easy to do it with a 12C508 although you could hardly set threshold (or perhaps via frequency variation ?) Hope it could help ... Tell us the result of your brainstorming ;-) Jean-Yves > >> {Original Message removed}

Problems I encountered when I built one of these: 1. Soil resistance changes over time. I would plug my two probes into the wet soil, then an hour later it would start beeping. I made a little circuit to plot this on my datalogger, and noticed a dramatic change in soil resistance. Of course I was measuring with an AC signal to prevent corrosion. 2. You can buy other people's attempts at doing this, that also don't work, at dimestores for $4.00. See Consumer reports, early this year. 3. I made a real nice cricket chirp, and made a resistance measurer using AN512 style RC timer measurement. Can't find the code anymore, so I must've erased it in disgust. {Original Message removed}

On Mon, 10 May 1999, Benjamin Petersen wrote: > I have an (old) idea for a new pic project. > I would like to attatch this device to a plant and make it beep when the > plant needs water. Check out Stephen Bolt's "Green Thumb", it's *very* similar to your suggestion, but much simpler: http://www.xs4all.nl/~sbolt/e-groeneVinger.html His site also describes other interesting projects: very low power BEAM robots, and a "spider" that uses an IR sensor/emitter as a form of radar! - j

There was a researcher in Canada in the late eighties who said that plants gave off ultrasonic sounds when they were running low on water. The Radio Canada interview about this topic made some humorous statements about the plants crying out because of their thirst, but the serious theory was that the sounds had to do with the mechanics of fluid flow in the plant's capillaries. The researcher at the time was using a Commodore C64 as the controller for the experiment and the sounds that it emitted when the plant was out of water were computer generated using the C64's sound chip. I heard all of this over 10 years ago so I have forgotten the details, but if there is anything to the ultrasonic theory, this might be a thought. I smile with sick glee as I think of the millions of weeds in our yard which will be howling their little selves to death in July and August during the dry time. Oh! the botany! Martin McCormick WB5AGZ Stillwater, OK OSU Center for Computing and Information Services Data Communications Group

Lawrence Lile wrote: > Problems I encountered when I built one of these: > 1. Soil resistance changes over time. I would plug my two probes > into the wet soil, then an hour later it would start beeping. I made > a little circuit to plot this on my datalogger, and noticed a dramatic > change in soil resistance. Of course I was measuring with an AC > signal to prevent corrosion. Do I take this to mean that the (continuous) measurement current (AC notwithstanding) is causing "polarization" at the electrodes, drying the soil locally? Surely the advantage of the 12C508 is the ability to sleep most of the time (and even then, execute the test cycle only every so many "sleeps",) and apply current and make a measurement only every five minutes or so? -- Cheers, Paul B.

> I would use a opamp configured as a comperator to measure the resistance > between two electrodes in the Plant pot. By using this approach you will > have a 0 or 5 v signal to the 12c508, and you can add a pot to adjust the > sencivity of the moist sensor. > > Regrads Henrik Holmgaard A LDR arangement could be fitted to this opamp configuration to meet your demand 'not to beep when it is dark' Cool, just switch the light off when it squeels for water........... Regards Cassie. > > > {Original Message removed}

Just made one of these for local Guides (Girl Scouts) project. No PIC though :-( Just one NPN transistor as a switch, a biasing resistor, LED, 3v batteries and two paper clip probes. No soldering - all screws into a 4 hole pice of terminal block. If I remember right a 6K8 resistor was about right to turn LED on if probe resistance was greater than 15K which seemed to be a reaonable wet/dry point for my house plants. [Completely OT] - I am looking for ideas for projects to make with youngsters. Rules are - must screw into terminal block (ie resistors, caps, diodes,etc and 3 legged devices such as transistors are OK, larger SIP and DIP devices are not, buzzers, speakers, batteries, mikes etc OK - I know I'll have to solder leads for them). Must be fairly simple (say 4 transistor or equivalent maximum) and must be fairly easy to explain. Since this is so [OT] replies directly to my reader please. Any ideas may possibly get included in a booklet for Scouts and Guides. Brian Replace the probes with a reed switch and the LED with a buzzer, put a magnet to hold reed switch on and hey -prestoi a burglar alarm. Brian E Jones Centre for Java Technology IBM HURSLEY

-----Original Message----- From: Paul B. Webster VK2BZC <spam_OUTpaulbTakeThisOuTmidcoast.com.au> To: Lawrence Lile <.....lilelKILLspam@spam@toastmaster.com>; PICLISTKILLspamMITVMA.MIT.EDU <.....PICLISTKILLspam.....MITVMA.MIT.EDU> Date: Monday, May 10, 1999 5:27 PM Subject: Re: "Plant out of water project" {Quote hidden}>Lawrence Lile wrote: > >> Problems I encountered when I built one of these: >> 1. Soil resistance changes over time. I would plug my two probes >> into the wet soil, then an hour later it would start beeping. I made >> a little circuit to plot this on my datalogger, and noticed a dramatic >> change in soil resistance. Of course I was measuring with an AC >> signal to prevent corrosion. > Paul; Replied > Do I take this to mean that the (continuous) measurement current (AC >notwithstanding) is causing "polarization" at the electrodes, drying the >soil locally? > I never did understand the phenomemon completely. I set up a 12C508 so it sleeps, takes a measurement, sleeps, etc. etc., and it would invariably alarm after an hour, even in obviously wet soil. A profile of the soil resistance over time looked suspiciously like an exponential RC charge curve (although many things in nature exhibit exponential behavious and this resemblence proves nothing) . It was wierd. > Surely the advantage of the 12C508 is the ability to sleep most of the >time (and even then, execute the test cycle only every so many >"sleeps",) and apply current and make a measurement only every five >minutes or so? >-- > Cheers, > Paul B. I wish this had worked.