Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Liquid level sensing

Status
Not open for further replies.

Bizzarri

New Member
Hello all,
ive recently been asked to measure the liquid level of a horizontal pipe. To do so there is a 4mm hole at the top of the pipe in order for some sensor, which will then be welded/ screwed to a modified micrometer. The sensor then enters slowly into pipe.Eventually when the sensor makes contact with the water, some light/audio sensor will then go off to indicate to the operator, to stop lowering, and read off the micrometer.

Ive looked into modifying a conductivity meter to attatch to the micrometer , so when the sensor reachs the water, a reading is given, and then a reading from the micrometer can be taken. this way is looking quite ineffective as it takes several seconds for the conductivity meter to recognise the water and turn on:(. I was wondering if anyone knows of any other water sensing methods, that would fit in a 4mm diameter hole and which is cheaper than non intrusive methods like ultrasonic.

I am relatively knew to this so was hoping i have overlooked a more effective method. Thanks in advance:)

(the liquid being water/80 mm diamter pipe/level measurment accuracy 0.5mm).

ps. dont hesitate to ask about my poor explanation above as i will try to clarify.
 
There are several threads on this forum for sensing the presence of plain water.
Basically, with two probes 1 mm apart you can expect 10-100 KΩ resistance. With distilled water it's way higher.
I think a simple CMOS circuit could give you millisecond response times.
 
Since you're looking for just a contact/non-contact indication there are a lot of simple "water alarm", "leak alarm", "flood alarm"... circuits out there. These respond instantaneously. Google any of these. One contact to the pipe, and the other to a stainless steel rod attached to your electrically isolated (from the pipe) micrometer.

Ken
 
Since you're looking for just a contact/non-contact indication there are a lot of simple "water alarm", "leak alarm", "flood alarm"... circuits out there. These respond instantaneously. Google any of these. One contact to the pipe, and the other to a stainless steel rod attached to your electrically isolated (from the pipe) micrometer.

Ken

Firtsly, thankyou for the replies. Secondly i have googled the above and each unit seems to be the same. They consist of 1 main base unit (the alarm), with 1 power port, and 1 port for a wire with a large water sensor. If it is these you were refering to, i am a bit confused as to what to contact to the pipe and what the contact to the stainless steel rod.

I can understand how the above would work with a simple circuit, as when the stainless steel rod touches the water, the circuit completes, (and hopefully overcomes the resistance of the water, which i doubt will happen at higher water levels) to power the visual/audio output.:confused:
 
Sorry about that. Almost all the alarms, battery or line operated, have two contacts to sense the water. I noticed one was just set on the floor and both contacts were probably hidden underneath. Most have a cable that connects the alarm box to a pair of metal contacts that are placed where water would be a problem. Here is what I was visualizing:

https://www.defensedevices.com/flood-alarm.html

The cable coming out of it has two tires. You would cut off the little do-dad on the end and use these two wires for your sensor. One would go to the metal(?) pipe. The other wire would be attacked to a metal rod (insulated from the pipe) that is attached to your micrometer, and is moved into the pipe like you original sensor example.

Ken
 
Hi Bizarri,

this is how I did the circuit:

I used an NE555 as astable multivibrator supplied with +/-6 (to minimize electrolytic effects), connected circuit ground to the main contactor (probe), thus having 6V effectively across the sensor and a height adjustable stainless steel rod (1mm) for the second part of the sensor.

The return signal is very weak and has to be amplified using an OpAmp. (I use an amplification factor of 1,000:1.) After amplification I rectify the output to have positive going pulses which in turn charge a small electrolytic cap which then opens an NPN-transistor to activate the alarm circuit.

To deactivate the alarm circuit the cap has to be discharged pretty rapidly for fast reaction. If discharging too fast the voltage (half wave) won't suffice to control the transistor, if discharging too slowly there will be some delay in deactivating the alarm.

Boncuk
 
hopefully overcomes the resistance of the water, which i doubt will happen at higher water levels

I don't think your mental image of how this works is correct.
 
Sorry about that. Almost all the alarms, battery or line operated, have two contacts to sense the water. I noticed one was just set on the floor and both contacts were probably hidden underneath. Most have a cable that connects the alarm box to a pair of metal contacts that are placed where water would be a problem. Here is what I was visualizing:

https://www.defensedevices.com/flood-alarm.html

The cable coming out of it has two tires. You would cut off the little do-dad on the end and use these two wires for your sensor. One would go to the metal(?) pipe. The other wire would be attacked to a metal rod (insulated from the pipe) that is attached to your micrometer, and is moved into the pipe like you original sensor example.

Ken

i can see what you mean. i'm just a bit worried that when the 2 contacts are 80 mm apart (seperated by the water and pipe) that it will still alert :confused:
 
Should work...but...without trying it, I can't say it will for any particular commercial device. It sounds like a cop-out, I know, but it is something to try. As willbe mentioned about the variation in resistance of water, your water contant would be different than if I tried it in a lab with piped-in DI water, or waste water from a chemistry lab.

I have some flood alarm/shutoff-valves I made for one of our Biology labs. I'll try one today, with 80mm spacing.

Ken
 
i owe you 1. never too certain wether the charge/eletricity provided from my battery/mains would just dissipate completely when the probe hit the water. im hopig the distance between the 2 probes holds more bearing than the size of the body of water its testing. its an 80mm diam pipe so im guessing 50 mm distance between probs is acceptable.
 
OK, my circuit :
**broken link removed**
Since this is a commercial kit, I will not post their schematic.
My test consisted of a 600mL beaker (80mm ID) filled to a depth of 80mm. One electrode, a 1/8" rod, was submerged on one side of the beaker to a depth of 80mm. The second electrode, a 1/8" rod, was a sharpened to a point. The circuit has a sensitivity trim pot. I looked for the alarm to sound just when the point touched the water's surface, 80mm away from the first electrode. In tap water, the alarm triggered through a large portion of the sensitivity range. With DI water the alarm was consistent only at the high end of the sensitivity. With double distilled water the pointed electrode had to be submerged about 5mm to even start triggering the alarm. A little higher sensitivity setting, and the circuit started to "gurgle" by just my finger touching the water. Capillary action, pulling water up the tip of the electrode, and water movement may be a problem for 0.5mm accuracy. This is just one test with one circuit.

There are probably others (add "circuit" to the other terms I said to Google) that are more sensitive, with better noise immunity.


Ken
 
Thats circuit is looking very promising. Distilled water is 2 micro siemans a cm where as my water is 30-50. Hopefully the pipe and larger body of water wont sap all the charge. And 1 of the sensory probes can be converted into a copper rod while the other on the outside of the pipe. the shape of the 2 sensing probs is not too obvious on the website. Your help has been invaluable and greatly appreciated.
 
Distilled water is 2 micro siemans a cm where as my water is 30-50.

So 40 micromhos = 25 KΩ-cm = ρ (i.e., rho)
50 mm (=5 cm) apart,
each probe presents 1 sq. mm (0.01 sq. cm) surface area to the other?

Neglecting edge effects,
R = ρ*L/A = [25 KΩ*5/0.01] = 13 MΩ resistance and 77 nanomhos conductance between the probes.

You might need an opamp circuit.
 
Last edited:
Hopefully the pipe and larger body of water wont sap all the charge.

This circuit depends on the resistivity of the water, not charge. The big metal pipe is an advantage in this situation.

the shape of the 2 sensing probs is not too obvious on the website.

The shape of the electrodes in the kit are irrelevant to your purpose. They are just two 1/2" diameter pads etched on a small PCB that is placed on the floor. When water wicks under the PCB the alarm goes off.

Ken
 
thats pretty good. so long as they are modifiable for a steel rod to be attacthed. plus it looks as if an op amp can be added, although telling by the experiment you did it sounds relatively sensitive so there is probably already 1 in there. i was also wondering if the diameter of the sensing rod would have any major bearing
 
so long as they are modifiable for a steel rod to be attacthed

It just has two wires coming from the PCB that you can attach to what ever you want.

plus it looks as if an op amp can be added, although telling by the experiment you did it sounds relatively sensitive so there is probably already 1 in there

Already has two.

Ken
 
I think that settles it then. im going to buy one, and hope my 3 mm diamter rod does the trick as soon as it contacts. the sensitity is easily adjustable with the twist of the trimpot.

Only problem is , they dont ship to europe!! godammit
 
Last edited:
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top