Empty water bottle detector

[lapm]

Well how about low tech solution? Colorful plastic ball that floats in water, assuming you don't do daily fresh water change. Not sure where one might get such, toy stores maybe?

[GK]

This is probably the best solution:

http://www.taste.com.au/recipes/29437/tuscan+rabbit+with+pancetta+and+rosemary

[Tris20]

Hmm, thanks for pointing out the poisoning!

Why not just use something like smart card technology. I'm not to informed on it but from what I gather it's just a coil with a tiny chip. Put a coil inside a plastic float, put a sensor at the bottom, when the water is empty the sensor will detect the coil is in range and Bob's your uncle!



If anyone could expand on this with potential values for the coil length and frequency etc it would be appreciated. I don't do Electromagnetics till next semester

[strangelovemd12]

So far your initial suggestion looks to be the best.  But would "breaking the vacuum" really cause the bottle to leak?  I am thinking of the standard inverted plastic bottle with a metal tube and a little ball in the end.  Doesn't the water pressure just push against the ball, sealing the opening, then the animal pushes the ball back opening the check valve to drink?  I do remember this type of bottle leaking, but I always assumed it was the low quality seal from the little ball in the spout.  I  cannot imagine the rabbits are sealing their mouths around the spout and sucking out the water.

I'm also confused on all this fresh water business.   Unless these fools are ordering cylinders of hydrogen and burning it in a nonreactive chamber with a perfectly stoichiometric O2 mix, I think they are full of it.  As long as the bottle is regularly cleaned/disinfected the water isn't going to go stale, or any other thing detrimental to the health of your rabbits.   For all we know changing the water bottle more frequently than is necessary burdens poor bunny with undo mental stress as the source of nourishment known to revitalize it's precious bodily fluids is ripped from it's comfortingly static world.  Shame on all you more-frequently-than-is-necessary-water-changers.

[ddavidebor]

Policeman, that's a international forum, not british language fan club...

You did see the signature right below the post in question, right?

I'm not an English native speaker, please correct me if I write something wrong

Aah, no, my fault.

Please don't arrest me for justice interference. :-)

[AndyC_772]

Lol No problem... just don't let the Welsh hear you use the phrase "British language", you'll have a war on your hands!!

[Alana]

Went to pet store to look at bunny watering stuff and got and idea.
Laser pointer LED as light source takes care of most of the optics and it should be cheap to get.
It should give strong signal on photodetector and make it immune to ambient light change.
The only problem would be how to prevent laser from shining into wabbits eyes - for obvious reasons, but i think it could be done with proper mechanical assembly.
Cork... if its from wine bottle it is non-toxic to humans and should be ok for wabbits as well - provided that you wash away wine residue with hot water.
 

[IanB]

should be "on a daily basis" as in doing it on a schedule, or doing it on time.

And just to confuse things even more, usage can also vary by location. As a British speaker I would favour "doing it to a schedule" over "doing it on a schedule".

And now back to your regularly scheduled programming... 

[rexxar]

Two metal rods vertically inside the bottle. Both full insulated except one is exposed at the top(in air), the other exposed at the bottom( submerged). When resistance of bottom exposed rod is equal to the resistance of top exposed rod then the bottom is now exposed to air, therefore empty. Pretty sure this is along the same lines as other suggestions but this is how I would do it. But the device would also go off if you filled the bottle too much and submerged the top exposed rod.

Also, +1 for the timer. You don't want this going off at midnight

You will form a electrolysis cell, and likely strip metal ions from the electrodes, potentially poisoning your bunnies.  If you really must use electrodes, then only apply a small current on an infrequent basis (perhaps 0.1 seconds every hour, etc...) to make your measurement.

Depending on your level of expertise and/or motivation, and the assumption that you are looking at this as a challenge, here are some other ideas:
1. Optical, using the change in refractive index between air/water to alter the light path.
2. TDR - time domain reflectometry, wind a tight coil of wire on a hollow-former.  Send down a fast pulses and measure the time they take to return (easier than it sounds, and you can average it over perhaps 1E6 pulses).
3. Wind a coil round the outside of the bottle, use this in a colpitts oscillator and measure the frequency change.
4. Put a colored ball in the bottle, use a webcam with something like the roborealm/opencv machine vision.

I've gone with #1. I discovered that if I point the LED and receiver at each other, put them so the curve of the bottle just blocks them, the water refracts the light out to the side, but the air lets it go through.

I got some opamps in the mail today, so I decided to learn how those work, and I think it's a much better solution than amplifying the signal from the receiver with two NPNs..

I've attached a schematic of what I've built. I'm sure there's plenty of errors, but this is the first time I've used an opamp, so don't make fun of me! It works well, gives about a 1 second buzz every 15 minutes or so. Is there a good place to stick a photoresistor in this circuit to make it not go off when it's dark?

[rexxar]

So far your initial suggestion looks to be the best.  But would "breaking the vacuum" really cause the bottle to leak?  I am thinking of the standard inverted plastic bottle with a metal tube and a little ball in the end.  Doesn't the water pressure just push against the ball, sealing the opening, then the animal pushes the ball back opening the check valve to drink?  I do remember this type of bottle leaking, but I always assumed it was the low quality seal from the little ball in the spout.  I  cannot imagine the rabbits are sealing their mouths around the spout and sucking out the water.

The ball is held down with water pressure, but it doesn't seal the nozzle at all. The vacuum inside the bottle is what holds the water in. When the ball is held up against the opening, it makes the opening so small that surface tension holds the water in, but when you move the ball, there's enough room for a droplet to form.

Trust me, if you don't seal it well enough (or it just has a crappy seal), these bottles can make a mess 

[daveatol]

not playing the grammar police, but from your sig line i'm guessing your trying to learn. should be "on a daily basis" as in doing it on a schedule, or doing it on time.

While maybe not everyone can agree if you want your project to be on schedule or to a schedule (both sound fine to me), I think everyone will agree that "you're" should be used to mean "you are", rather than "your" as you've written.

I've gone with #1. I discovered that if I point the LED and receiver at each other, put them so the curve of the bottle just blocks them, the water refracts the light out to the side, but the air lets it go through.

I got some opamps in the mail today, so I decided to learn how those work, and I think it's a much better solution than amplifying the signal from the receiver with two NPNs..

I've attached a schematic of what I've built. I'm sure there's plenty of errors, but this is the first time I've used an opamp, so don't make fun of me! It works well, gives about a 1 second buzz every 15 minutes or so. Is there a good place to stick a photoresistor in this circuit to make it not go off when it's dark?

Well done for making it work. There are some errors in your schematic. The resistor connecting to pin 7 on each 555 is in the wrong position. You're missing a connection to pin 2 of the lower 555.

You could put the photoresistor between the opamp and the 2n2222; when it's dark, there won't be enough current to power the 555. It's not a good solution, but should "work". A perhaps better solution would be to connect it into the timing portion of the first 555 so that it beeps more quickly when it's light, and not at all when it's dark.

Personally, I'd probably try your original solution of using a magnetic float and reed switch or have the whole bottle hanging on a sprung microswitch,  for the sake of zero power usage and simplicity.

[GeoffS]

This is probably the best solution:

http://www.taste.com.au/recipes/29437/tuscan+rabbit+with+pancetta+and+rosemary

 
As an Australian, I'd have to agree that the only good rabbit is a dead one.   

[SeanB]

I feel like some as well. Always was threatening to eat Wodger Wabbit, but never did so and he died from old age.

[deth502]

While maybe not everyone can agree if you want your project to be on schedule or to a schedule (both sound fine to me), I think everyone will agree that "you're" should be used to mean "you are", rather than "your" as you've written.

argh!! you caught me!! thats a pet peeve of mine, i cant believe i did that. the ' key on my keyboard is not my friend, so i usually just skip it rather than go through the frustration of getting it to work unless i really have to, so i mostly leave it out of contractions. but yeah, i should have at least typed youre in that instance, i screwed the pooch there. 

[rexxar]

Well done for making it work. There are some errors in your schematic. The resistor connecting to pin 7 on each 555 is in the wrong position. You're missing a connection to pin 2 of the lower 555.

D'oh!  It's right on the breadboard, I guess I just can't read.

You could put the photoresistor between the opamp and the 2n2222; when it's dark, there won't be enough current to power the 555. It's not a good solution, but should "work". A perhaps better solution would be to connect it into the timing portion of the first 555 so that it beeps more quickly when it's light, and not at all when it's dark.

That didn't work out too well, so I decided to make the photoresistor part of a voltage divider and feed that into another opamp. This probably isn't the 'correct' way to do it, but it works for me.

Personally, I'd probably try your original solution of using a magnetic float and reed switch or have the whole bottle hanging on a sprung microswitch,  for the sake of zero power usage and simplicity.

Unfortunately, the hobby budget has been rescinded until we can get the car fixed, so I wasn't allowed to order any new parts  . I used what I had to hand, and I think it was a fun project. I got to learn about opamps, charge pumps, and 555s.

Here's my final schematic, I don't think there's any errors this time.  I'm still trying to figure out DIP Trace, but anything is better than my hand-drawn ones...

[daveatol]

Well, it's great that you're getting it to work, and sorry for the bad lead. If it works reliably, then it's the right way to do it

Your schematic is very strange to me, and is certainly not standard. The opamp part with the variable resistor appears to be wired incorrectly with with the parallel 4k7 resistors going between 12V and an assumed/omitted ground. The charge pump is incorrect with the series diode and capacitor.

All the best.

[Psi]

Sparkfun just released a water alarm that maybe useful.
It uses a MCU with arduino code so it would be simple to change it so the alarm sounds when there's no water instead of water present.

https://www.sparkfun.com/products/12069

[SeanB]

That will corrode the electrodes as DC is present. Mutant bunnies anybody? If you are going to do contact probing just use the old fashioned method used in all level controllers of low current AC voltage and detecting the AC voltage across the probe. those work for water and last decades in service, probes and all.

[deth502]

since we keep talking about electrodes and such inside the bottle, if we are going to violate the stated rule of not puncturing the bottle, why not just a single stainless steel tube down to the bottom and detect the level based on the pressure in the tube.

[rexxar]

For maximum ridiculousness we could stick a stainless steel bolt and a thermocouple in the bottle attached to a peltier cooler, and estimate the volume of water based on how long it takes to reach a set temperature.

Bonus: nice cool water for the bunnies on hot summer days.

[strangelovemd12]

That's a bit impractical, Rexxar.  Much better off using separate O2 and H2 reservoirs running into a fuel cell, which powers a small XYZ table.  Set a standard keyboard on the table and it can send a tweet whenever the gas supply is low.  The bunnies simply drink the clean H2O exhaust from the fuel cell.  Don't beat yourself up about it, these simple solutions always seem obvious in retrospect.

[Ryan]

I like the magnet that floats.  Add a reed switch, battery & buzzer, and you're golden.  However,  I have to agree Bunnies need fresh water daily.  How about an automatic watering system that gets them water directly from the faucet.  Those are readily available, get the bunnies fresh water (as long as the hose isn't very long), and a much lazier approach.  However involving electricity, water and bunnies sounds fun.

[Mr Smiley]

Bunny’s are highly intelligent creatures 

Just teach them to press a button next to the water bottle, when it's empty, that latches on a light   

Bunny bonding

 

[rexxar]

Bunny’s are highly intelligent creatures 

Just teach them to press a button next to the water bottle, when it's empty, that latches on a light   

Bunny bonding

 

They already tell me when it's empty. They're expressive enough to communicate with humans, if you've been with them long enough to recognize it.

I like the magnet that floats.  Add a reed switch, battery & buzzer, and you're golden.  However,  I have to agree Bunnies need fresh water daily.  How about an automatic watering system that gets them water directly from the faucet.  Those are readily available, get the bunnies fresh water (as long as the hose isn't very long), and a much lazier approach.  However involving electricity, water and bunnies sounds fun.

I didn't have any magnets or reed switches handy, and I can't buy anything new. So, optical detection built from my limited inventory it is.

[rexxar]

That's a bit impractical, Rexxar.  Much better off using separate O2 and H2 reservoirs running into a fuel cell, which powers a small XYZ table.  Set a standard keyboard on the table and it can send a tweet whenever the gas supply is low.  The bunnies simply drink the clean H2O exhaust from the fuel cell.  Don't beat yourself up about it, these simple solutions always seem obvious in retrospect.

That's a bit inefficient, though. I think a better way would be to have the fuel cell power an arduino which interfaces with an Rpi to do the tweeting.

I wonder if there's an arduino sensor shield to detect water freshness....