Designing a low-voltage Moisture Sensor Grid or Pad?

[Firstly, my apologies to alt.engineering.electrical if this post is out of place there.]

Any ideas (and/or recommended references) as to where/how to get started? What I'm aiming to create is a floor pad, the size of a prototype to be about four square feet, equipped with densely spaced, evenly distributed moisture-detection points. Obviously the use here of many moisture sensitive microchips would get too expensive. So I'm thinking more along the lines of a grid pattern of wires: wire anodes criss-crossing wire cathodes, each kept vertically separated by some non-compressible material possessing excellent water absorbtion characteristics; such that a low voltage current will flow between anode and cathode (to trigger an ultrasonic-alarm device) whenever "Kitten decides to do her thing" and the material thereby becomes wet... Now, I don't happen to know of any specific materials that would fit this bill. Nor do I have the specific technical experience or knowledge to guesstimate whether this idea is even plausible or not.

What I do know I have this problem with 'people and their pets' inside the house. I have poured over the internet searching for possible technical solutions to apply to this problem. Some I've purchased and tried (e.g. cat facial pheromone or Feliway®), but none have worked. I have researched electronic pet repellants and found a couple of such products that do have a proven track record. But these simply deter animals from entering or remaining in a certain area by emitting an ultrasonic noise (silent to humans) upon the detection of motion --any motion, by any pet, person, or thing, for any reason. Such a device would definitely scare the cat out of the house, but wouldn't meet with the cat's owner's approval since it is not selective for specific problem behavior, and she wants the cat living indoors. Thus I need to develop a customized (yet affordable) technical solution, such as would briefly emit an ultrasonic "noise" to startle the cat, but *only in response to wetting behavior* by the cat.

Therefore, helpful feedback/references/possible plans/schematics/suggestions appreciated?

TIA, Ken

I made something surprisingly similar to this when I was about 11 years old for mice. A sheet metal square about 12" sq. attached to a piece of plywood with another smaller (2" sq.) block of wood in the center. Atop the small block was another smaller metal plate. Each plate was connected to one side of an ac line cord. This was (in the mind of an

11 year old) the current version of the better mousetrap. Bait was placed on top the center electrode and I set it up in the barn to watch the fun. I did have to install a small sacrificial fuse in order to avoid frying them like bratwurst.

However, to keep it on topic, I modified the same idea a few years ago to prevent my cat from digging in my rubber plant pot and peeing in the resulting dirt on the carpet. I placed a wire grid (screen) on the carpet under the pot, extending about 8" all around it. I put another grid just below the surface of the dirt in the pot, and hooked each grid to the poles of a small "electric fencer" circuit I had built. Not enough to fry the kitty, but definitely enough to keep her from doing it again. The next time the cat stood on the grid beside the pot with her hind feet and started digging in the pot with her front paws, she got the message quickly, and I removed the device.

Nels

You don't need anything that complicated. Just get a suitable non conductive substrate and use contact adhesive to cover it with a single sheet of aluminium foil. Then remove small strips of foil to leave an interlocking pair of spaced fingers. Or form the same sort of thing using conductive ink. Or use a soft pencil and then electroplate.

Now energise it with a low current alternating voltage between each set of fingers and sense the impedance. When there is a step change, stick a couple of thousand volts, very limited current across it - or sound your ultrasonic "noise". Use ac for sensing to minimise corrosion.

With different settings, it can educate, terminally educate if you wish, rats,etc. They need quite a large current before they catch fire, btw and it ruins the fingers.

With yet more different settings, it can play a reward "happy tune", when fitted to a kiddie's pottie and activated by No 1s or 2s.

Mixing the rat and the child application could be unfortunate..

--

Sue

relay switch | | E1--0 sensor 0-----------(+) 0---/ ----0 | Op Amp (out) ------relay coil--- ---------|------------(-) | | | Gnd R1 R2 | | gnd gnd

A simple Op Amp comparator circuit driving a relay would do it for you. There needs to be only very little current to get the op amp to go high at the output. I like the aluminum foil fingers idea that was posted. It is cheap and easily replaces after the critter soils it.

I don't know what the resistance of cat piss is but it will help you determine what value resistors are needed.. R1 and R1 should be equal. I am guess a 12V source and 100Meg resistors to start off with.. Look for a relay with a high coil resistance, They do make some "reed" relays that take very little current to fire..

Years ago Popular Electronics had a project for a "Rain Detector". Aluminum foil attached to a block of wood with rubber cement. Exacto knife cut through the foil to create two terminals. Drive it through a simple transistor to whatever signal device you want.

For a big sensor pad like you want, interlace several "fingers" of foil contact.

Lump

LOL, you are a troo loon!

Here's what he needs:

+6V>--+----------+---------+--------+-----------+ | | | | | | | | [10K] | | | | | | [SENSOR] | +---|--[1M]--+ | | | | | | | | [10K]------+--[600R]--B 2N4403 VIN>--+-------------------|-/ LM393 C | | | | +--------NO | | | +-----+ | [100K] | | |K | O--> |--+----------+---------+--------------+-----+

Ground the unused inputs and output of the unused comparator, then to set it up what you do is to run the pot all the way to ground, attach the sensor and pee on it, and then adjust the pot until the relay clicks in and then just a little bit more for good measure.

Then, you take the sensor out wash and dry it and put it back. The relay can be used to trigger any "disciplinary" device, remembering that Mark Twain said about cats, "Of all God's creatures, there is only one that cannot be made the slave of the lash. That one is the cat. If man could be crossed with a cat it would improve man but it would deteriorate the cat."

--
John Fields
Professional Circuit Designer

Lol, Can't argue with that..

But I did build this some years ago,(Not to monitor cat piss). 2 resitors, op amp and I belive the relay was 5V with a 500ohm coil. Worked fine. Light goes on, Light goes off.....

But I am sure you will tell my why it dosen't work...

--
If it did, you were just lucky.

--
And mine?

"...the cat that sits down on a hot stove-lid. She will never sit down on a hot stove-lid again - and that is well; but also she will never sit down on a cold one anymore."

-- Mark Twain

out

be

ted

w

d=20

o

e=20

cal

=2Eg.

cts

rom

se

re

=20

t

nic

*=20

=2E

h at

t is

=2E I

or a

at

ors,=20

Light=20

Personally, I wouldn't use either of them! Sorry about that.

Yours because your comparator is using ground as the=20 reference. Any "leakage" across the sensor is going to trip=20 it. The cat stepping on it, or even breathing on it,=20 probably would. For a non-perfect op amp, it is highly=20 likely that it will be permanently tripped due to offset=20 voltage. Because you don't control the back emf from the=20 relay when it switches off - which could easily zap the op=20 amp. Because you don't control the maximum voltage between=20 the inverting and non inverting inputs, which could easily=20 cause the op amp to permanently lock or self destruct.=20 Because you use dc across the sensor, which will lead to=20 corrosion and premature failure. Because you have no=20 sensitivity or trigger point or hysteresis adjustment. Those=20 sort of things. Depending on the sensor and the op amp, it=20 certainly could work - but the odds are that it wouldn't=20 work reliably in this application..

-- Sue

--
For this application, I tend to think of the sensor as something
like a couple of pieces of conductive window screen with a paper
towel between them and paper towels on their outsides, so I don\'t
think there\'d be a lot of time for much to happen, chemically, after
the cat peed on the affair.  Especially since, if it works, (that
is, if whatever happens to the cat as a deterrent to keep it from
peeing there again works) it\'ll probably only be used once or twice
and I suspect the paper towels would be replaced between uses.

I used it in an old Ikagami studio camera. We had a rash of Plumbicon tubes that has short lived filiments. I used the circuit to insert a light bulb in the filiment circuit when the camera was put into standby mode to cut the filiment current when the camera was not in use.

Could not have been luck because I don't have much of that...

Just a looney hack, I guess...

We always just rubbed there nose in it, a little swat and threw em in the litter box. Didn't take long for them to figure out where to do there business. They seem to prefer the litter box over the carpet. Hard to dig a hole in the carpet I guess.. Harder to train the humans to change the litter box regularly than it is to train the cat to use it.

Oh, every engineer has their own way of doing things, have their favourite chips and favourite building blocks.

I am quite possibly too hung up on always using ac with liquid sensors. I thought that your circuit, even with the current limited to microamps, would tend to encourage dendritic growth or electrochemical migration - particularly with the odd chemicals that might crystalise out of cat urine in place. It depends on the gap left between the fingers and the underlying substrate, of course. But you do have 6 volts across the gaps.

I may easily be wrong! I tend to be very lazy and stick to the building blocks I have used before..

--
Sue

Such a device would definitely scare the cat out of

So, finally, it comes to light that this is some babe, and she insists on bringing "her" cat into your house if you want to get laid?

It ain't worth it. She's addicted to cat, and can never love a man. Lose the bitch.

--
Flap!
The Pig Bladder from Uranus, still waiting for that
hot babe to ask what my favorite planet is. ;-j

Had you thought of buying or hiring the bell and pad system used for kids that wet their bed?

David

Ken Moiarty wrote:

I hadn't thought of using a layer of paper towel. Sounds a good idea as the liquid has to go somewhere...Better to have an instant mop. As you say, it will hopefully only be used a couple of times and is not the sort of thing that you would want to keep around afterwards as a conversation piece.

I assume that kittens do actually know when they are weeing? We may only be training it to avoid walking on paper towels/shiny metal surfaces. IIRC, several AI systems have failed due to the wrong choice of training sets..

--

Sue

That's life, if you want pussy, you have to have a cat...

--
Sue