Hi,
I like to build a water detector with four states (no water, okay, lots of water, to much water). The principle I decided to use is simple. I have a pole with a down-wire and an up-wire. on the down-wire there are three resistors. The (non-destilled) water bypasses at its level resulting in a overall-resistans of 0, 1, 2 or 3 Resistors. In each case a LED should be lit.
Is there a way to do this without a µC? I know there should be and I have drawn a lot of resistors already, but it should not be that difficult....
My suggestion would be to use 4 water resistance thermistors, which would change their values when they would be covered with water. Upon studying the value change of each thermistor when covered with water, one should calculate the resulting voltage drop, whereupon assuring that only when the voltage is a certain value, should the LEDs turn on.
Both your examples I had already in mind. The issue is, that you have four seperate cycles. If one cycle is broken you do not get a warning.
When you find a solution to measure the resistance of four distinct steps and difine a minimum as well as a maximum you are safe.
Maybe its not clear enough what I am trying to do. I have done a bit of drawing. These were my first attempts, but calculating the resistors and finding transistors got me frustrated...
To break it down to the point I like to have some "blackbox" wich triggers four transistors, if the resistor reaches given values. Isn`t there a way to do it? Or even a componen, which does something like that by itself?
I'm having a similar challenge, where I want to prevent a water pump from activating below a certain water level. Like you, I want something reliable and dead simple to make sure the pump doesn't run dry (and burns as a result).
I have been toying with the idea of using either a magnetic reed switch of hall effect sensor attached on the outside of the tank and than have a magnet attached to a cork float up/down guided vertically by a wire or something anchored to the bottom of the tank.
Maybe you could use something similar, with four sensors or magnetic reed switches?
I have been toying with the idea of using either a magnetic reed switch of hall effect sensor attached on the outside of the tank and than have a magnet attached to a cork float up/down guided vertically by a wire or something anchored to the bottom of the tank.
Interesting making such a device failsafe.
What if hard water formed a scale on the wire below the cork, preventing it from dropping?
I agree with that. My current take on it is to give the the loop a lot of room around the wire/guide so that it cannot really get stuck, but it's not optimal.
I'm thinking there must be a "standard solution" to this problem which is really easy to build, but I haven't found one yet, but I'm just starting to think about it and looking around.
One could use something that floats on the water, connected to an arm which is on the other side is connected to a potentiometer. That's the way most fuel meters in gas tanks work. It probably is quite ugly in a fish-tank, but one could easily calculate the amount of water in the tank at every level and it's quite simple/cheap to build.
Reed switches with magnets and floats is the easy way:
Another option but spendy if you're doing multiple levels:
However:
One can measure the level of a tank by having a pressure transducer hooked up to a tube/pipe measuring the vacuum/pressure. Basically you have the tube submerged, half full of water when you hook up the pressure sensor and when the tank rises/falls there will be relative pressure changes.
It's like when you dip your straw into a glass then put your finger on the top of the straw to lift the liquid, you've created a vacuum. The sensor will measure the tank exterting pressure on that vacuum or adding to the vacuum when the level is low.
there are a lot of ideas around here I have already came across. Most of them have moving parts and have some trouble in being 100% dependable. There can be situations when they give the wrong answer. But what is the problem with my solution? Why doesn't anyone answer to that? Is there really no way to trigger a multiple-switch by the amount of voltage it gets? I really thought this should be the easiest way...
I just need a wisely chosen transistor-array. If the resistans of the probe exceeds a certain limit a transistor switches. I just hoped someone could help me with finding the exact transistors.
@Leon:
I often read about it, but I don't really believe it could measure the depth of a fish tank. there would be greater influence by the weather, than by some cm of water...
