Hi All
I need a simple sensor, that can test if water level has dropped below a certain limit in a container.
I do not need any measurements for this application, simply if water has reduced below a certain limit.
Could a simple pulse on a GPIO and then using another GPIO to read it be adequate? It would not be the pin directly, but via cable/wire.
https://www.allelectronics.com/c...
Whether or not your method would work, would depend on the purity of the water!
Lots of ways to measure water level, conduction, level from top via radar/lidar or ultrasound, one of the oldest ways is a float switch.
Jim
You can do it electronically, use AC (or alternating DC pins) to prevent plating of the "sensor". Life is simpler with a switch float, though that won't give a reading of different levels, like a gas gauge. You can get a float arm with a pot instead of a switch.
For just measuring yes/no level, it is hard to beat a float switch (hall efffect & magnet)
you can emulate some of this using the AVR instead of the LM1830
https://datasheetspdf.com/pdf-file/64122/NationalSemiconductor/LM1830/1
here is one that might be subject to plating issues (are you constantly exposed to liquid?)
http://www.learningaboutelectronics.com/Articles/AVR-liquid-level-sensor-circuit.php
The LM1830 has been obsolete for a while I think, but I have a few in stock from when I used to make water sensors for coffee machines (in case anyone wants one or 2 
)
I bought one of these:
https://www.hydrocheckproducts.c...
I opened it after calling the vendor and got the approval to disable the buzzer >very nice people< and found that they use a TINY441 in it.
It works great. No issues with it
Jim
I have a few in stock from when I used to make water sensors for coffee machines (in case anyone wants one or 2
OK the board is on the way, you only need to put a bit of metalwork around it, a tank, heater etc......
Another way to measure water is to weigh it - though the mechanics might mitigate against it. That has the advantage that you can both read an absolute mass of water (or of water + hardware, depending where you put the load cell) and can also measure usage with high precision... but disadvantages include that it requires individual calibration; that the mechanical fitting might be tricky; and that the sensor requires specialised electronics.
Neil
Hmm. Other ways to measure water level... 
These have the same problems as barnacle's solution, but here's a few thoughts:
Turn on a known power heater for a known amount of time and measure the temperature increase.
Test conductivity, throw in a known amount of salt, mix, measure conductivity again.
Tilt the bucket to a known angle, see if anything slops over. (Rain gauges do this.)
Akin to the above - add water until it overflows, the level was how much you had to add.
Akin to another above, turn on a heater and a microphone, listen to how it takes to boil... (you might want to correct for atmospheric pressure here).
S.
Given that the OP is only looking to sense if the water is above or below a certain level, not to measure how much water is present, a simple float switch would do it.
As an alternative, depending on what the container is made from, I wonder if you could do it optically? Either by transmission or reflection.
To be a lot more helpful, yes. If you're using typical tap water all you need is two copper strips about 1cm apart, at least one of which ends at the desired level. Wire the strips up to your GPIOs. Pick one as an input, and turn on the pullup resistor. Drive the other low. If there's water, the input's low. If there isn't, it's high.
Doesn't work with pure distilled water, though (as per ki0bk).
Done? S.
(Incidentally, one thing you can do for fun is pour yourself a glass of your finest local tap-water, then put your multimeter probes into it and measure the resistance... ) S.
Other methods - less precise but possibly acceptable:
Neil (going off to dry his hands)
Refraction.
But I hafta admit the idea of a liquid-cored inductor is fascinating on many levels. 
S.
I once prototyped a capacitive sensor -- worked OK, but I couldn't use it because the actual application ( auto coolant level in the overflow tank)as the tank was too deeply buried to mount the sensors.
IIRC I tested it with a milk jug and a couple of metal strips mounted on the outside. Just used a 555. I either hooked it up to change the frequency or just to allow the signal through when the water was present -- it's been a while :). I remember being surprised at how well it worked.
I've seen this type of thing being used in RVs to check water levels.
hj
You can use PTC in a hotwire, like so : https://youtu.be/jTqHXeWDT50
Then, it is independent ( mostly ) of electrical conductivity.
