I've an underground cement tank about this size: 3m x 1.5m x 1.2m and I need to monitor the water level of it.
I'm not sure about what kind of sensor to use, I was thinking about an ultrasonic sensor, for example JSN-SR04T. However it doesn't feel robust (looking at the pictures) and I don't know if it's the right choice, with this topic I hope to collect some advices, here some requirements:
The lid of the tank must be kept closed, so the sensor, if inside, must be waterproof or however able to bear high humidity.
Yes I can drill a hole in the lid for the sensor, but then the hole should be closed well in order to keep the tank isolated.
The solution must work for years, I don't want a system that breaks every other month
The Arduino needs to be around 6 meters away from the tank (in another room)
There is no need for real-time monitoring, sampling time of one minute is already enough
There will be a display or some other device to allow me to check easily the water level
Accuracy is not too important, even an error of 5% is acceptable, I don't need to know how many litres of water are in the tank but rather what it the water level in percentage.
Hardware shouldn't cost more than 100€ (Arduino excluded)
The water in the tank is used for "home uses": cooking, showers, washing dishes but not for drinking.
Resolution of 30cm is enough, in this way, since the tank is 3m high I can monitor these percentages: 10%-20%-30%-40%-50%-60%-70%-80%-90%-100%
Water won't be exposed to freezing temperates, it's probably gonna operate in the range 10°C - 25°C
I don't need help with the code, I need suggestions on the hardware side, in particular in this phase I want to get some ideas on which sensor could satisfy my task.
Any powered electronics kept in a high humidity environment will corrode in weeks, unless completely protected by either "potting" in silicon compound, or painted with conformal coating (similar to nail polish).
An approach that does not require electronics inside is a bubbler -- the air pressure required to force air out of a tube open at the bottom of the tank is proportional to the height of the water above it, relative to atmospheric pressure.
A float sensor with resistive feedback is another possibility. Power it only when taking a measurement.
Your tank, obviously, hast to have a screened air vent to the outside, so that is one way to connect to a sensor.
For years I have used two float switches for my irrigation storage tanks. I use two float switches to indicate time to turn on the water to fill the tank and one float switch near the top to indicate the tank is full.
Both switches are PVC plastic with a reed switch inside. The float has a magnet to close/open the switch. Both switches are screwed into PVC pipe fittings with the wires going into the pipe and exiting from the pipe at the top. The top is fastened to through a hole in the top of the plastic storage tank.
You can put as many float switches as you need in that one PVC pipe and you will need to drill a hole to fasten the pipe.
That is the way I would go about it. I have used the bubbler method for years on small tanks to large tanks. Nothing exposed to corrosion or in the humid atmosphere of the tank.
I do this with my koi pond outside. The pump is inside and the pond outside. I taped the airline going outside and placed a differential pressure sensor on the line. I built a small analog circuit and add a digital meter. It has been operational for about 8 years now, no problems. It is still within 1/2" without any calibration. It has alarms so if it gets to low we know. Note: I left the second port open to atmosphere.
For those who are using the bubbler systems, I have a question. Can one detect the variation from the bubble releases, as discussed in the linked article above, and if so, could they reasonably be used as a self-diagnostic to warn the user if the bubbler is passing too much, or too little gas? I think the answer would be yes, within limits, but it would depend on the system, and I don't have one to test.
Yeah, umm, that was the article I referred to. But I'm asking, has anyone done anything useful with the signal that would give you, to self-detect impending failure?
That’s a pretty tall order I have worked on countless level measuring systems. None would work for “years” without some amount of servicing. And none were installed for anything close to 100 euros.
Of course that is an industrial installation. A bubble tube is the simplest way to go without a doubt. But that type of system will only work with an open vessel, one that is open to the atmosphere. We don’t know if that is the case.
A bubble system can still be used on a sealed tank if a reference to the empty top of the tank can be obtained. Also the tank will have to be able to tolerate the slow increase in pressure that the air bubbles will create.
There is not enough information to start designing a measurement system yet.
Is the tank vented or sealed?
Is a source of compressed air available?
Will the system be exposed to freezing temperatures?
If the water in the tank is intended for human consumption, a whole other list of questions need to be answered.
I use the DYP-A02YY and finds It to be precise and pretty resistant for my rain water tank (using both the UART and RS485 versions in various settings )
Thank you all for the replies, I'll try to answer everybody:
Thank you, this sensor seems to be better than the ultrasound since the latter somehow depends on temperature, will take it in consideration, I just have doubts if it would have long term problems being underwater all the time
Thank you, do you think the sensor suggested by @paulpaulson is likely to corrode and fail?
I like the bubbler principle, however it seems more complicated to deploy:
I would need an air pressure supply, a pressure sensor and some pipes. Also that air would need to go out of the tank and I'm not sure it would create troubles to the pump. I would prefer a solution that requires less hardware but won't discard the bubbler just yet.
About the float sensor, from what I understood it helps to understand if water has reached a certain level but I would need more than one of it in order to be able to monitor the level.
You're right, should have be a requirement, I think being able to measure a change of 10% of the whole tank would be sufficient, so I would say 30cm
Understood, in order to be able to measure these water levels: 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, 100% I would need 10 float switches which would make it "too much work"
As said above, bubbler method would require more hardware, if possible would consider a simpler solution first
I'm not sure I understand your solution, a sketch would help
Read above, I would need 10 of it which would make it complicated
Not bad, only problem is that it use batteries, I wonder how long would they last before having to reopen the tank and change it. For the moment let's move towards Arduino kind of solution, I might want to do more than displaying the level in the future (relè etc etc)
Thank you, the tank is sealed, however there is somewhere a motor/pump that takes the water from the bottom of it and brings the water to another tank placed on the roof, however I've no access to it. No source of compressed air available atm, I would need to get an air pressure supply if I decided to move with the bubbler solution. No, the system won't be exposed to freezing temperature. Water is used to wash dishes, cooking and showers, but not for drinking, I'll add this on the initial post, thank you.
Thanks you, do you get false measurement sometimes? Do you have less accuracy when the water level drops below 80%?
To summarize, these seem to be the possible solutions we have got so far:
Water level sensor HGY TL-136
Bubbler system
Float sensor
Ultrasound sensor DYP-A02YY
Stratco system
Let's see if you guys have more to say, is there any reason to not use the HGY TL-136? Thank you all.
What ever you do , you will need a high quality industrial sensor , which will cost you more than $100.
Think carefully about what you are doing , and the accuracy .
You might find a stick with a ball on the end , with the stick pointing up through a hole will do the job
Have a look at sensors for cess pit levels , widely available and good for a sh** environment .
It’s pretty stable - you can average a few readings and remove outliers. My tank is a flat one 7500 litters but less than 1m high so I don’t have huge distance to measure
