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.

Water Drop Sensor

Status
Not open for further replies.

atntias

New Member
hi,

if I have a constant drip of water and i want to know exactly when each drop falls, so you can say the water is like a resistor and put two wires aligned exactly below the drop and when it falls it will trigger a 741 op amp circuit to make a line go high to 5V, so it can be detected by a PIC microprocessor.

so i have this sketch and its kinde works but not so good i wanted to know some advice about how to approach this to get it to work better, meaning what changes can be done in the circuit to adjust it to work better.

**broken link removed**
 
hi,
I would suggest is to use a CA3140 CMOS opa, they have a very input impedance, the 741 is a dinosaur.

Do you have a picture of the system you could post?
 
Yeah if they get wet then remember to use something like graphite (?) I believe because it won't oxidize.
 
hi

yes the water drops do touch the contact and it covered with gold so it wont rust, the thing is i am getting some changes in the voltage but not good enough to be detected by a PIC. any ideas?
other methods are acceptable to the goal is to know the drop rate with good precision.

here are some pictures of the system...
 

Attachments

  • DSC06090.JPG
    DSC06090.JPG
    147.7 KB · Views: 498
  • DSC06092.JPG
    DSC06092.JPG
    145 KB · Views: 425
  • DSC06089.JPG
    DSC06089.JPG
    141.2 KB · Views: 406
atntias said:
yes the water drops do touch the contact and it covered with gold so it wont rust, the thing is i am getting some changes in the voltage but not good enough to be detected by a PIC. any ideas?
other methods are acceptable to the goal is to know the drop rate with good precision.

What opamp did you use?.

Also I would suggest using it as a comparator, rather than this rather crude way.
 
Nigel Goodwin said:
What opamp did you use?.

Also I would suggest using it as a comparator, rather than this rather crude way.

Yes, I would also suggest using a comparator (or an op-amp as a comparator). You want a logic-level output, and comparators are good for that.
 
well i asked this quesion at another forum (dutchforce) and Søren came up with this solution which gives a ~10µs pulse out when the drop ceases to bridge the input.
**broken link removed**
 
Here's one that puts out a logic HIGH as long as water bridges the gap.
 

Attachments

  • water droplet dtector.PNG
    water droplet dtector.PNG
    21.5 KB · Views: 662
Suggestion.........

atntias said:
hi,

if I have a constant drip of water and i want to know exactly when each drop falls, so you can say the water is like a resistor and put two wires aligned exactly below the drop and when it falls it will trigger a 741 op amp circuit to make a line go high to 5V, so it can be detected by a PIC microprocessor.

so i have this sketch and its kinde works but not so good i wanted to know some advice about how to approach this to get it to work better, meaning what changes can be done in the circuit to adjust it to work better.

QUOTE] [snip]

How about just using a waterproof microphone? And amplify the sound it makes when the drop hits. Or put a regular mic. inside a small waterproof drum.
 
that would be allot more complected because the drip is fast and ul have to deal with noise reduction and voice analysis the 2 solutions above provide a much more accurate and simple solution.
 
Rejection.........

atntias said:
that would be allot more complected because the drip is fast and ul have to deal with noise reduction and voice analysis the 2 solutions above provide a much more accurate and simple solution.

I can understand your outright rejection of my suggestion if you think that it is necessary to apply "noise reduction and voice analysis" to the mic. output. (I think the s/n ratio will take care of that)
And I might even understand why you don't think the mic. would not respond to a "fast" drip in an "accurate" manner. (limited frequency response?:D)
But I do understand that you probably never have heard a mic. being hit by a raindrop (or anything else) and checked the signal on an oscilloscope.
And now i understand why you posted the in the first place.
 
Rolf said:
I can understand your outright rejection of my suggestion if you think that it is necessary to apply "noise reduction and voice analysis" to the mic. output. (I think the s/n ratio will take care of that)
And I might even understand why you don't think the mic. would not respond to a "fast" drip in an "accurate" manner. (limited frequency response?:D)
But I do understand that you probably never have heard a mic. being hit by a raindrop (or anything else) and checked the signal on an oscilloscope.
And now i understand why you posted the in the first place.

hi,
Rolf is quiet correct about the sound amplitude of a water drop hitting a mike.
If you choose your mike carefully, you will require minimal amplification.

I have used low cost mike's as impact detectors in rolling mills.
 
i understand that if the water is hitting something hard and making a solid noise then is possible.
but that not the case the water is dropping to a container filled with water and there are splashes that make noise too, the drop rate is quite fast so u will have to analysis the voice and find wich peak is the water drop and what are the splashes it just seems to me much more complected... if u have an idea to simplfy it i will love to hear it im not denying any ideas as it may seem.....

just to note what are the requirements, the water is dropping to a small container willed with about 2 inches of water which an overflow hole to where the pump is... i can place a mic in the container walls but the affect of the drop hitting the water is important...
if u have any other question don't be shy
 
Last edited:
Water droplet analyses.....

atntias said:
i understand that if the water is hitting something hard and making a solid noise then is possible.
but that not the case the water is dropping to a container filled with water and there are splashes that make noise too, the drop rate is quite fast so u will have to analysis the voice and find wich peak is the water drop and what are the splashes it just seems to me much more complected... if u have an idea to simplfy it i will love to hear it im not denying any ideas as it may seem.....

just to note what are the requirements, the water is dropping to a small container willed with about 2 inches of water which an overflow hole to where the pump is... i can place a mic in the container walls but the affect of the drop hitting the water is important...
if u have any other question don't be shy

Don't quite understand what you are doing, thought you were counting drops.
I would think a water drop hitting the mic. or a membrane close to it is going to make the mic. output many times bigger than any acoustical noise from splashing of droplets hitting the water or the sides of the container.
Did some crude checks on droplet rates, anything over 5 or 6 per second tends to make the water go in broken streams. Don't think you can get much more than that even under ideal conditions.
 
@atntias
nice time fountain setup.so ur planning it with drop sensor.
pls do update us once u finish it.and what are u using for pulse generation.i mean is it 555 or PIC?
and for this experiment i don't recommend that flow control knob.its not that accurate.
we have a great thread here
**broken link removed**
 
heres th whole deal..

im working on a time fountain

lately i got accesses to a 3d printing technology, which is an exiting new technology for prototyping... i live in Israel and there is this Israeli development for a 3d printer which basically prints 3d parts from very thin PVC layer by layer fashion. more info can be found here
**broken link removed**

so i got hand on with 3d software disgn (solidworks) i wanted nothing less then drop sensor with pic...
with pic i haven't really got started yet but i'm planning to use a 16f84 wich i have with a homemade burner i actually no expierance with pics but i know C well..... i guss how hard can it be :p

here pictures of the 3d design i have a meeting with the 3d printing guys this week to go over the first draft.

**broken link removed**
**broken link removed**
**broken link removed**
**broken link removed**
 
drop drop drop drip

Well,

Once I did repair a kind of very precisely controlled distiller used in petrol industry analyses, and these guys solved it like this.

The point where the liquid is dripping from, is isolated and by shape they've made sure that there is no electrical contact to the rest of the system. That The drop hangs and drops from a conductor (stainless steel wire, sharp tip) which is held on a high voltage through a resistor to prevent arcing (about- 60 to -100V, 2-10MOhm).

Each drop has it's own capacity and carries some charge with it as it drops.

On the low side, there is a fet op-amp with inputs protected with a low capacity diode and a 10M resistor to the ground. The op-amp reads impulses recpectve to the ground and there are some buffers after it to just shape impulses. Need I say, this electrode is just a small plate also very well insulated from the rest of the system.

On this distiller this reading (Frequaency) is used to control power to heaters and there you get this system regulated.
 
being a scientist

I am terribly sorry for complicating things up. And the answer couldn't be simpler.

Just let it drip on a piezo buzzer (ringer as used in an old banana phone) and amplify whatever you get from it. I guarantee you wont need to filter anything much.

Why doesn't anyone tell me I'm full of s...

Good luck.
 
Problem Time.

hi, so first i completely changed my design of the fountain its not finished yet but when its done il post some pictures...

meanwhile i'v been working on the electronics... and encountered some problems..
im using 30 strip led that come with build in resistor and require 12V.
so i used a old voltage regulator i had (MC 78T05CT).. i connected it like in the scheme.
i started with connecting 1 small led with 470 ohm resistor and it worked..first time without TIP41 second time with... blinked at both... when i connected the strip leds.. it didn't work i tried playing with the connections and then the pic and voltage regulator heated so much i burnt my hand... the pic was burnt out but the voltage regulator is ok.. i replaced the pic it the last one i have.. and i don't know whats wrong any ideas?

btw in the schem theres a values messed up, the capacitor on the voltage regulator wich 0.1uF..
 

Attachments

  • 1.PNG
    1.PNG
    32.4 KB · Views: 450
Last edited:
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top