Water level detection

[Sajjadkhan]

Ok guys i know there are lots of methods out there to detect water level but i need a serious one because i am being paid for to detect water level and turning on/off the pump as required. The tanks are huge like 500000 Gal which are used to provide water to all the homes in that area. This is only one part of the project but its bugging me because they are asking me that sensors should work for no less than 5 years.

I have 3 methods of measuring water.

1. I am thinking of using Aluminum probes (as i think they are the best to use).
2. using IR method.
3. Using thermistor.

Problem with Aluminum probes:
1. When probes are in water and you apply voltage, electrolysis starts and its life decreases. i can increase its life by sending some very short voltage signals when required.
2. I have noticed that when tank temperature increases the water evaporates and if we use a weak pull up/pull down configuration it will false trigger due to that moister around the probes which helps in conduction. and if i use strong pull up/pull down resistors then it may happens that water conductivity (due to any reason) changes and it wont trigger.
3. I don't have any idea that what effects will be on that probes when they are submerged in the water for long time.

Problems with IR method:
1. It can be use to detect only top level since it floats on top.
2. I can use them in water but i don't know the water condition i.e. how much pure the water is (clarity point of view).

Thermistor is attracting me because i can fix it in some thin Aluminum pipe which will be sealed. there will be some current flowing through the thermistor which will heat it up. when that single Aluminum pipe probe touches the water thermistor current will reduce and i can use that as a detector.


So if you guys have any suggestions then do hit me. Remember i need long life span sensor and minimal maintenance or none.

 

[keith1200rs]

I would suggest stainless steel not aluminium for a conducive probe. You can avoid electrolysis by using an AC signal and a fairly high impedance (low current).

Have you thought of a float and switch?

Keith

 

[Sajjadkhan]

You can avoid electrolysis by using an AC signal and a fairly high impedance (low current).

Let be clear here. by AC signal you mean like conventional AC in our homes which also have negative amplitude (With low amplitude and isolation)? or you are talking about square wave signal which alternates between 0-5V?

Have you thought of a float and switch?

Are you talking about like float valve in flush tanks and connect two contact points with it? i wonder how can i make small version of it since everything should give a professional look.

 

[KerimF]

To sense the various water level in my tank (also to control the pumps), I used chrome probes available in the market for this purpose and a low level AC signal derived from the transformer auxilliary winding that supplies the circuit. But only 12 years have passed without problems

 

[keith1200rs]

By AC I mean no DC component. You could use a square wave but you need to capacitively couple it to ensure there is no DC component.

Keith

 

[Sajjadkhan]

To sense the various water level in my tank (also to control the pumps), I used chrome probes available in the market for this purpose and a low level AC signal derived from the transformer auxilliary winding that supplies the circuit. But only 12 years have passed without problems

12 years wow. ok if i dont find chrome probes then where i can get them i mean is there any commonly used instrument where i can take this material off?

---------- Post added at 19:30 ---------- Previous post was at 19:25 ----------

By AC I mean no DC component. You could use a square wave but you need to capacitively couple it to ensure there is no DC component.

Keith

yep i was thinking the same. but AC signal with negative amplitude prevent electrolysis i think.

So how to get rid off noise that is around the probes due to water vapors?

---------- Post added at 19:42 ---------- Previous post was at 19:30 ----------

I just found that bicycle spokes are made of chrome. Is it always true, how can i find out?

 

[FvM]

Bicycle spokes are regular steel with chrome plating. For electrodes, use bulk NiCr steel. NiCr steel welding wire or rods are e.g. fine.

 

[KerimF]

In my humble market, I just ask for water probes likely in stores which sell industrial electrical materials.

The AC current avoid polarization of the probe in the long run.

I found in my very old archive the circuit I made at that time. I used a small 220V/(9V+9V) transformer since my relays were 12Vdc 280R.
By a voltage divider (27K and 39K) connected to one 9Vac terminal, a weak 5Vac (about 0.2mA if shorted) drives the two probes; one is common and the other at a certain level. This 5Vac is monitored by a simple diode rectifier. If no water, the rectifier output is high otherwise it becomes low due to water small impedance. The rest of the circuit (using transistors) has the role to act as a low pass filter, schmith trigger and relay driver. If you like I will redraw the circuit for you (it was drawn on a program, EasyPC, running on DOS and old matrix printers only :sad: )

Kerim

 

[keith1200rs]

I did a similar circuit many years ago. I simply used a 2N7000 to switch the relay. Hysteresis was provided by the time delay from detecting the signal and the water shutting off, but these were small hot water tanks. With a large tank you would need some hysteresis. My probes were simply stainless steel rods. Despite scale buildup they worked for many years without problems and several thousand were made every year.

Keith

 

[Sajjadkhan]

If you like I will redraw the circuit for you (it was drawn on a program, EasyPC, running on DOS and old matrix printers only :sad: )

Kerim

Its ok i understood.

---------- Post added at 22:14 ---------- Previous post was at 22:11 ----------

Yes now scale build up is the only issue left. Thermistor method is the answer for that? Anyone's opinion?