How to measure water flowing from a washing machine pump?

[Brad_805]

Hi all
Am trying to find a way to measure the vol/time flow of water from the washing machine. Is there an electronic component that can be used like a coil of copper wire over the tube and the do something with a microcontroller perhaps and it figures out much flow? by electromagnetic measurement?

 

[davenn]

Hi there
Welcome to PF

Is there an electronic component that can be used like a coil of copper wire over the tube and the do something with a microcontroller perhaps and it figures out much flow? by electromagnetic measurement?

water is neither magnetic nor electromagnetic
there's zillions of flowmeters available ...

https://www.bing.com/search?q=water+flow+meter+suppliers&src=IE-SearchBox&FORM=IENTSRDave

 

[Bystander]

there's zillions of flowmeters available ...

Dynamic range is another matter; admittedly I've not been all that active recently, but "flow-rates" have, in my past experience, seldom exceeded 10 percent accuracy outside a very limited dynamic range, single O(m).

 

[PeterDonis]

Moderator's note: Thread level changed to "I".

 

[Brad_805]

I want to make it from parts, cost maybe $25.00 I know there are flow meters for sale. wanted to make it from scratch

 

[pedro the swift]

Stopwatch and a bucket. Thats less than $25.

 

[Brad_805]

Stopwatch and a bucket. Thats less than $25.

So the requirement here is to have an electronic sensor and microcontroller that measures the instantaneous d/dt flow, sending out flow measurements with a time stamp. The real question is what mechanism is used in this industry to measure water flow? what electronic apparatus, device, sensor, circuit or thing would be put inline if needed or around the hose or at the point where water exits the hose to measure the realtime flow. Or if there is no way known by science to measure water flow electronically? then that's the answer.

 

[anorlunda]

https://en.wikipedia.org/wiki/Ultrasonic_flow_meter

It sounds like you are seeking something like the ultrasonic flow meter. They exist, but I believe it will cost much more than $25 with all the supporting electronics. Probably closer to $25000.

 

[Dullard]

There are several ways to measure your water flow. These include:

a 'Mag Meter' (requires that the water be slightly conductive) -these are commercially available - I've never 'rolled my own.'

A dP device - insert an orifice and measure the flow-related pressure drop - this is a pretty standard approach.

A thermal device - measure the water temp and measure the power required to maintain a heated probe at a slightly higher temp. This is a common way to precisely measure gas flow - it will work with water.

I don't think you'll get near $25 with any of them, unless you have a pretty good junk box and mad skills.

 

[Nugatory]

Am trying to find a way to measure the vol/time flow of water from the washing machine.

You will likely get better answers if you tell us more about what problem you're actually trying to solve. By far the most cost-effective and practical way of measuring the flow from a washing machine is a bucket and a stopwatch... but if that answer isn't what you looking for, then there must be other constraint as well.

Are you by any chance trying to adapt a washing machine motor for some purpose other than washing clothes? If so, tell us what you're trying to do and see what the collective experience of this group can contribute.

 

[Brad_805]

There are several ways to measure your water flow. These include:

a 'Mag Meter' (requires that the water be slightly conductive) -these are commercially available - I've never 'rolled my own.'

A dP device - insert an orifice and measure the flow-related pressure drop - this is a pretty standard approach.

A thermal device - measure the water temp and measure the power required to maintain a heated probe at a slightly higher temp. This is a common way to precisely measure gas flow - it will work with water.

I don't think you'll get near $25 with any of them, unless you have a pretty good junk box and mad skills.

Well the dp device sensing pressure drop may be the best to start with. problem is that in the washer, water flow is say 100% then it drops and tapers off so thermal may not work on the partial flow. Used to have a lot in the junk boxes. people called me a hoarder? I just needed a larger shop. Sadly I have next to no junk now.

 

[Brad_805]

You will likely get better answers if you tell us more about what problem you're actually trying to solve. By far the most cost-effective and practical way of measuring the flow from a washing machine is a bucket and a stopwatch... but if that answer isn't what you looking for, then there must be other constraint as well.

Are you by any chance trying to adapt a washing machine motor for some purpose other than washing clothes? If so, tell us what you're trying to do and see what the collective experience of this group can contribute.

Well... I don't know how to explain it better, If I said what its for? then the discussion would move away from trying to measure water flow, it is not some hack of the washing machine, its a standard washer.

having an electronic sensor thing that sends digitized flow info 24/7 I think the bucket idea may not be ideal. to calibrate it perhaps? sure. I wanted to know the mechanisms for sensing water flow thru a pipe, what reacts with the water? if a coil was placed over the pipe and the coil was part of a tank circuit with some freq range, and water passes in the pipe would the freq change?

 

[jbriggs444]

If I said what its for? then the discussion would move away from trying to measure water flow

The problem is that unless we know what it is for, we do not know what cost/benefit tradeoffs are relevant. One cannot choose a good design unless one knows the constraints under which one is operating and the design goals one is working to optimize.

Just to add another possibility... flexible tubing and a spring scale to measure centrifugal force around a curve in the tube. Of course, good calibration would probably need to compensate for pressure in the line as well.

 

[Bystander]

https://en.wikipedia.org/wiki/Mass_flow_meter

 

[Brad_805]

Well I mean its a microcontroller and then some wires going to some kind of sensor, or some kind of sensor unit that's powered and all it does is quantify flow. Is there nothing water reacts to? magnetic field zero, charge zero are there no fundamental forces known that water reacts to? yes the spring idea is perhaps very workable, how much deflection happens to some plate as the hose comes up and bends down into the drain pipe, add a device the water hose plugs into, water hits a plastic flap held shut by a spring. water dripping onto the flap would push it some amount, then the trick is to find the conversion sensor to measure that flap

 

[davenn]

Is there nothing water reacts to? magnetic field zero, charge zero are there no fundamental forces known that water reacts to?

as far as I'm aware, nothing in a way that would help you measure the flow

building your own for $25 probably isn't going to happen

What is your electronics experience ( be honest, don't exaggerate) ?
You will have to buy the flow sensor and that could easily cost more than $25 on its own
( I haven't specifically priced them, have you ??) let alone the rest of the electronics, the
time programming your micro to work with the data from the sensor

I gave you a whole page of links

 

[Asymptotic]

Another consideration is you propose to measure gray water flow rate from the washer whih will be loaded with dirt, fibers and other debris, and tend to clog certain types of flow meters.

 

[russ_watters]

So the requirement here is to have an electronic sensor and microcontroller that measures the instantaneous d/dt flow, sending out flow measurements with a time stamp.

How often? And are you sure you mean instantaneous and not cumulative?

 

[Brad_805]

What is your electronics experience ( be honest, don't exaggerate) ?

extensive. about 45 years, a lot of school, a lot of design, test. Worked with and owned tv repair shops, fixed tube-transistor and newer, a lot of electronics work on a lot of things for many companies over the decades. still learning, always new things I didnt know.

You will have to buy the flow sensor and that could easily cost more than $25 on its own

--No, I want to make the flow sensor, there has to be some way.

( I haven't specifically priced them, have you ??) let alone the rest of the electronics, the
time programming your micro to work with the data from the sensor

--I also do programming, RT embedded is fun.

let me look for that list. One of my problems today is that I had to give up just about ALL of my junk, things, tools, spools of wire large/small, connectors, circuits, chips, passives... a small fortune in parts. don't have access to my stereo scope nor hot air soldering tool, or all the automotive tools. most of those are gone. so instead of just try something? I don't have anything to work with, its a very depressing feeling, literally. Then being employed part time, there isn't a budget to buy anything

 

[Brad_805]

How often? And are you sure you mean instantaneous and not cumulative?

Need both but need to rate d/dt and then the total, but would summarize

 

[Brad_805]

Another consideration is you propose to measure gray water flow rate from the washer whih will be loaded with dirt, fibers and other debris, and tend to clog certain types of flow meters.

Thats a great point, its gray water. so a plastic flap being deflected over a spring and then a 3 axis resolver chip glued to the flap, they cost about $1 glued to the hinge. then map out xyz and make a correction factor chart, and the microcontroller would calculate based on what it gets from that resolver chip. probably a good way to go

 

[russ_watters]

Need both but need to rate d/dt and then the total, but would summarize

How often?

 

[Brad_805]

How often?

I think a plastic flap with a 3d resolver chip on the hinge fed to the microcontroller would work, 100 samples/sec should work? then calibrate with known volumes of water, etc

 

[JT Smith]

You can buy a flow meter for $10 or less that will work for you, provided you're not looking for high accuracy. But since you don't want to buy one you could copy that design. If you do an internet search you'll find an example of someone who did that. I turned up two different inexpensive build-your-own-flowmeter designs in five minutes of googling.

 

[Brad_805]

You can buy a flow meter for $10 or less that will work for you, provided you're not looking for high accuracy. But since you don't want to buy one you could copy that design. If you do an internet search you'll find an example of someone who did that. I turned up two different inexpensive build-your-own-flowmeter designs in five minutes of googling.

Ok I found some ideas but one of the purposes of asking here was to ask about the fundamental physical properties of water flow measurement; what reacts with water, or actually gray water. Maybe gray water is more reactive? will gray soap rinse water store an electric charge? but that may not be good,

 

[russ_watters]

I think a plastic flap with a 3d resolver chip on the hinge fed to the microcontroller would work, 100 samples/sec should work? then calibrate with known volumes of water, etc

That's a really, really high sampling rate.

 

[anorlunda]

Ok I found some ideas but one of the purposes of asking here was to ask about the fundamental physical properties of water flow measurement; what reacts with water, or actually gray water.

You're all over the place. It is one thing to tinker with a DIY flow measurement. But if you want to understand science, it takes serious study. Asking questions on the Internet is a very poor learning strategy.

So please decide. Would you like suggestions on a DIY project, or suggestions on a course of studies?

 

[Brad_805]

You're all over the place. It is one thing to tinker with a DIY flow measurement. But if you want to understand science, it takes serious study. Asking questions on the Internet is a very poor learning strategy.

So please decide. Would you like suggestions on a DIY project, or suggestions on a course of studies?

Seriously? you can't follow what I said, and if you read all the posts, came up with an impossible answer solving a water flow problem for cheap? I chose a physics group that has educated people responding in a peer fashion; if one disagrees? they can correct the other. I want to learn far more about science, if not for this need to earn money for rent, bills and food my entire existence would be r/d.

By all means! I want to study and learn more, but at no time have I been "all over the place". In a couple days was able to come up with a concept to measure flow that likely has good accuracy, its now time to prototype and test the mechanisms

If you have suggestions on studying? I am interested.

 

[Brad_805]

That's a really, really high sampling rate.

How about 50 samples/sec?

 

[anorlunda]

If you have suggestions on studying? I am interested.

I'm not trying to be pedantic. I don't know your background. Many people start with Wikipedia.
https://en.wikipedia.org/wiki/Fluid_dynamics
Perhaps that's not relevant to your interest.

But I stand by what I said, that asking questions and reading short answers is a very poor strategy for learning. That is why we have textbooks. Students get to ask questions after reading the book and listening to the lectures.

 

[Brad_805]

I'm not trying to be pedantic. I don't know your background. Many people start with Wikipedia.
https://en.wikipedia.org/wiki/Fluid_dynamics
Perhaps that's not relevant to your interest.

But I stand by what I said, that asking questions and reading short answers is a very poor strategy for learning. That is why we have textbooks. Students get to ask questions after reading the book and listening to the lectures.

Right, so without backgrounds and all, people can be just a bulletin board voice not implying a timeline of abilities. And since time is more important than perfect text in a forum? people can sound one way but mean another.

Im still learning, and like to learn. here I want to refresh my recollection of physical properties involved by those who are freshly into theory. info/speculate, anything is fine. Already did web searches, want those who are involved to comment because it has a peer review kind of element to it.

by all means if you want to share education info I am happy to check it out

 

[russ_watters]

How about 50 samples/sec?

How about 1 sample every 10 seconds? Why do you need it that fast?

Have you researched the types of flow meters people have pointed out? You really should be doing a lot of your own research based the types of flow meters we've told you about. Your requirements don't make sense and don't jive with what you are telling us you want to do or what is available in the industry. You really need to put a lot more effort into this if you want the result of this discussion or your project to be useful.

 

[JT Smith]

Ok I found some ideas but one of the purposes of asking here was to ask about the fundamental physical properties of water flow measurement; what reacts with water, or actually gray water. Maybe gray water is more reactive? will gray soap rinse water store an electric charge? but that may not be good,

That wasn't clear from your original post. It seems like you're asking two questions: (1) how do you measure the flow at some unstated resolution; and (2) can you measure it non-intrusively. Perhaps you can't break into the outlet line to add a flowmeter? Or maybe you're just curious? Who knows?

I googled and very quickly found answers to both questions I assume you are asking, although I think to do it non-intrusively is going to be more expensive than your budget would allow. It's not hard using the web that way and far more effective than asking questions on forums. But maybe effectively obtaining answers isn't your primary goal.

 

[arydberg]

Look up Peristaltic pump and use it backwards.

 

[arydberg]

Sorry for the caps i cut & pasted it feom Wikipedia

 

[Brad_805]

How about 1 sample every 10 seconds? Why do you need it that fast?

Have you researched the types of flow meters people have pointed out? You really should be doing a lot of your own research based the types of flow meters we've told you about. Your requirements don't make sense and don't jive with what you are telling us you want to do or what is available in the industry. You really need to put a lot more effort into this if you want the result of this discussion or your project to be useful.

You may want to look into the concept of Sampling theory. Water flow is being measured in realtime 100hz is not unreasonable. Measuring rainfall? maybe 10 seconds is ok. I looked up everything folks posted, why the rudeness here? Nobody is forcing you to post are they? Are you just trying to complain, just because?

What do you know of my requirements? measure water flow from a washing machine, some very cheap sensor and a microcontroller that outputs flow values. Even $25 is a high cost but acceptable for the entire thing. I was asking how can water flow be measured, notice I did NOT ask what kinds of water flow systems are for sale commercially? That would be unfair of me because I can look up what's for sale. I wanted to ask those involved in physics theory to speculate, conject, guess or explain how water (or gray water) interacts with sensors so it can be measured. I did not ask to copy someone else's work or design, did I? no. I asked people about the physics behind sensing water flow, asking for ideas.

We came up with a plastic flap where exit water pushes the spring loaded flap, and for a sensor I suggested a basic MEMS gyro chip. With that and other ideas, the design can now go into prototyping and reality check. I had asked if a coil of wire used in some LC tank oscillator is affected by water flow? no answer.

There was a couple and the man complained alot. If his wife cooked eggs scrambled? he wanted them poached. If poached? he wanted them scrambled. One day the wife had what she thought was a great idea, I'll scramble one and poach the other. She presented the food, her husband sneered down at the plate and said blast it woman can't you do anything right? you scrambled the wrong one!

 

[jim hardy]

Since you're obviously a tinkerer

Cheap will be the difficult part. A few thoughts...

..............................
Look up "Nutating Disk" . it's the basis of your house water meter.
I see dozens of them in the brass bin at my local scrap metal yard.
I'd wager your local water company maintenance shop has a pile of old meters and they'd swap you one for a hot pizza around lunchtime.

That's a positive displacement measurement, like a pump run backwards.
..............................You know from your electrical background that a charge moving in a magnetic field feels a force
and that's the principle behind a magnetic flow meter.
This guy does a good job of explaining the principle but he failed to get one working


i think had he added a magnetic core around his pipe (near the end of the viseo) he'd have succeeded .
Beauty of that approach is your signal is already a voltage.
............

Look up Rotameter,
basically a weighted float in a vertical tapered tube that's lifted by velocity of water.
It should be self cleaning because clearance gets bigger as you go up.
Tapered tubes are sold as rain gages.

but you have to measure how high is the float in the tube ...

..................

There's a marvelous book published by ASME titled "Fluid Meters" that's a comprehensive introduction to principles of flow measurement.

https://www.asme.org/products/books/fluid-meters-theory-application-sixth

....................

Who knows - you might even make a pitot instrument from the pressure transducer out of a washing machine water level sensor

...................

Mark Twain complained he suffered from "An excess of imagination" .
I suppose it's better than being bored...
Good Luck !old jim

 

[russ_watters]

You may want to look into the concept of Sampling theory. Water flow is being measured in realtime 100hz is not unreasonable. Measuring rainfall? maybe 10 seconds is ok.

I'd be curious to see an example of a problem and a meter where water flow is measured with a sampling rate of 100hz and why. The sampling rate needs to be a function of how fast the flow rate is varying, how long the system needs to record for, and the data processing requirements. For a system like a washing machine discharge, 100hz is needlessly high. I work with building infrastructure systems and the meters I work with don't have sampling rates anywhere close to that high. A random $2000 ultrasonic flow meter I found via google was 10 seconds, which is where I got that number.

Rainfall is measured at sampling rates measured in minutes.

I looked up everything folks posted, why the rudeness here? Nobody is forcing you to post are they?

It's not rudeness, it is constructive criticism. And yes, it is my job to ensure that threads are productive and focused. This thread is problematic because of a vague purpose and seemingly wildly out of alignment constraints (very stringent measurement requirements combined with extremely low cost).

What do you know of my requirements? measure water flow from a washing machine...

Right, that's pretty much all you've told us.

...some very cheap sensor and a microcontroller that outputs flow values. Even $25 is a high cost but acceptable for the entire thing.

I would be curious to see an example. Can you provide one?

I was asking how can water flow be measured, notice I did NOT ask what kinds of water flow systems are for sale commercially? That would be unfair of me because I can look up what's for sale. I wanted to ask those involved in physics theory to speculate, conject, guess or explain how water (or gray water) interacts with sensors so it can be measured. I did not ask to copy someone else's work or design, did I? no. I asked people about the physics behind sensing water flow, asking for ideas.

That's a much larger description than you provided in the OP, though it is jumping around quite a bit.

It isn't clear if you have a specific measurement you want to make (much less what that measurement is) or if the purpose is to design a flow meter that you can then sell.

There was a couple and the man complained alot. If his wife cooked eggs scrambled? he wanted them poached. If poached? he wanted them scrambled. One day the wife had what she thought was a great idea, I'll scramble one and poach the other. She presented the food, her husband sneered down at the plate and said blast it woman can't you do anything right? you scrambled the wrong one!

Yes, that's a pretty good analogy to what is wrong with this thread.

 

[Asymptotic]

I was asking how can water flow be measured, notice I did NOT ask what kinds of water flow systems are for sale commercially? That would be unfair of me because I can look up what's for sale.

Researching commercially available flow sensors is a good place to start. Tailoring tradeoffs between initial cost, lifetime cost, accuracy, repeatability, turndown ratio (serviceable measurement range), operating temperature band, and other factors is why a wide range of different sensor types exist.

For example, paddlewheel flow sensors are one of the less expensive options, operate based on fluid velocity, typically have a turndown ratio of 20:1 (i.e - 20 FPS to 1 FPS), but must be installed in a straight run of pipe (10 to 50 pipe diameters upstream, somewhat less on the downstream side) to provide satisfactory operation. They are also susceptible to damage by water hammer, and surges in general.

Turbulence is a big deal when locating paddlewheel and many other flow sensors, hence the minimum straight run distances (often expressed in 'pipe diameters') from tees, elbows, and other turbulence-inducing devices - at best, being outside of laminar flow reduces accuracy and repeatability, and at worst, it destroys the sensor in short order. A clothes washer isn't likely to afford sufficient straight run distances, and may preclude using a flap-based sensor (which likely also works best in a laminar flow environment).

$25 seems unrealistically low for an accurate sensor with an adequate turndown ratio that will survive in this application.

 

[Brad_805]

Since you're obviously a tinkerer

Cheap will be the difficult part. A few thoughts...

..............................
Look up "Nutating Disk" . it's the basis of your house water meter.
I see dozens of them in the brass bin at my local scrap metal yard.
I'd wager your local water company maintenance shop has a pile of old meters and they'd swap you one for a hot pizza around lunchtime.

That's a positive displacement measurement, like a pump run backwards.
..............................You know from your electrical background that a charge moving in a magnetic field feels a force
and that's the principle behind a magnetic flow meter.
This guy does a good job of explaining the principle but he failed to get one working


i think had he added a magnetic core around his pipe (near the end of the viseo) he'd have succeeded .
Beauty of that approach is your signal is already a voltage.
............

Look up Rotameter,
basically a weighted float in a vertical tapered tube that's lifted by velocity of water.
It should be self cleaning because clearance gets bigger as you go up.
Tapered tubes are sold as rain gages.
View attachment 240897

but you have to measure how high is the float in the tube ...

..................

There's a marvelous book published by ASME titled "Fluid Meters" that's a comprehensive introduction to principles of flow measurement.

https://www.asme.org/products/books/fluid-meters-theory-application-sixth

....................

Who knows - you might even make a pitot instrument from the pressure transducer out of a washing machine water level sensor

...................

Mark Twain complained he suffered from "An excess of imagination" .
I suppose it's better than being bored...
Good Luck !old jim

Jim thanks so much! This is very helpful. I was reminded one of the problems is that this is gray water with soap and fiber and? residue in the flow, but the idea of the wobbling/rotating plate with a piece cut out of it is really clever. I know some folks who work at a city water dept, and know they like pizza,

and I had the idea of the electromagnet and measuring the reaction, but some here said water has NO reaction to electrical nor magnetic fields at all so a magnetic force cannot be used to measure water flow. I think the guy needed to have a better EM setup, run 10Ga wire up to the EM, then use solder pads to connect the copper to the wire, maybe thicker wire and more current, and why only on top? why not put the coil around the pipe?

The book link? This is the kind of thing I am looking for since its from 1971 I am hoping to have an easier time of finding/borrowing it.

Im sure I could do this with some impeller action, but no moving parts would be better, maybe a venturi from an old carb, extend the drain tube vertically and place the venturi in there and measure vacuum, there may be some effect like that once I get the bench side going. thanks again

 

[jim hardy]

but some here said water has NO reaction to electrical nor magnetic fields at all so a magnetic force cannot be used to measure water flow.

water itself doesn't but the ions dissolved in it do.
Tricks are
a strong magnetic field ,
sense induced voltage with a very high impedance amplifier connected to the electrodes
you'll probably need synchronous demodulation
and the ones i worked with used triple shields with a driven guard to keep from loading the electrodes.
Of course mine worked at 60 hz so line frequency interference had to be avoided.
It had probably forty pounds of magnetic core around a 2" pipe.
The electromagnet drew better part of ten amps to produce the magnetic field.
It generated signal of a millivolt or two proportional to flow.
Yours could be smaller. I think my washing machine pump has a 3/4" inlet.

Someplace on PF is a reference to an instruction manual for a Foxboro industrial magnetic flow meter.

maybe a venturi from an old carb, extend the drain tube vertically and place the venturi in there and measure vacuum,

you'll probably want the venturi in a horizontal run so as to not have to correct for elevation difference between the pressure taps.
But vertical is certainly do-able.

I hope you find a copy of "Fluid Meters" - it's a fascinating book and a great addition to any reference library

If you get started , consider opening a thread in our DIY forum.

What are you up to - trying to detect impending "Slow to Drain" error codes (F22?) on your front loader? My last two were ladies's mini sock-ettes in the filter , along with coins bobby pins and ball point pen refills...

Have Fun

old jim

 

[JT Smith]

You found on google how to measure the output of a washing machine's water for under $25?

Yes.

I asked about physics properties of measuring water...

Did you? I read your original post and it wasn't a physics question. I think you're having fun though and that seems to be what counts.

 

[jim hardy]

Yes.

Sounds like a handy device to know about.

Have you a link ?

 

[256bits]

Just run the water through a flap in the pipe.
The more water flow, the more the flap opens up.
Something like a dial sensor would record the angle of the flap, and thus the water flow.
25 bucks, maybe.

 

[ChemAir]

For this, I'd look at power usage by the discharge pump. Power usage should be proportional to mass flow. A power usage or clamp on meter might be used, depending on what other motors may be running during the draining. It would take some playing around and calibration by tape/bucket/scale.

Most flow measuring devices have problems when the pipe is not full.

 

[anorlunda]

The OP now has been given numerous answers to the original question. Thank you to all contributors.

Thread closed.