NerdKits - electronics education for a digital generation

You are not logged in. [log in]

NEW: Learning electronics? Ask your questions on the new Electronics Questions & Answers site hosted by CircuitLab.

Everything Else » Digital flow measurements?

November 14, 2012
by JimFrederickson
JimFrederickson's Avatar

Has anyone here done anything with measuring "liquid flow rates digitally"?

What have you used?
Did it work well?

I read somehwere, maybe here, that someone used some relatively strong magnets to ionize passing water and then measured current/voltage produced by the ionized water flowing past a sensor?

Mostly, I am interested in measuring/metering water that is flowing through 1/2" PVC lines.

I have "full control" over the water lines, so I can easily attach/embed what is necessary.

I have found a few "laboratory precision flow meters", and some others. The "laboratory precision flow meters" are in the range of $100+ and that is NOT going to work for me.

I was thinking something more along the lines of $20 per line to be measured.

November 14, 2012
by Ralphxyz
Ralphxyz's Avatar

Hi Jim, I'll have to think more about this. Generally "cheap" flow meters use some sort of spinning wheel to measure flow.

I have used ultra sound sensors but those were expensive, seems like there must be way especially since we have the microprocessor to handle any signal processing.


November 14, 2012
by pcbolt
pcbolt's Avatar

Jim -

Some of river flow meters I've used before are based on the Doppler principle and simply measure sound frequencies. You can judge water flow in a confined environment by charting the velocity vs. sound frequency profiles and calibrate accordingly. In a non-confined environment, you listen to one frequency with the listening device pointing "upstream" and one pointing "downstream", subtract the two to eliminate unaccounted noise and come up with a calibration scheme. The difference in upstream to downstream dominant frequencies should increase with speed. Of course now you have to come up with a waterproof underwater transducer.

November 14, 2012
by Ralphxyz
Ralphxyz's Avatar

The ones I used were external to the pipe so they did not need to be waterproof.

I wonder what would happen with a high frequency transmitter tight to one side of the pipe and a microphone on the other what would you hear?

Also I wonder if some sort of vibration sensor might detect flow?

and then letting my mind really run wild what if you had a piece of flexible tubing that would expand proportionally to the pressure? Possible you could measure the expansion and determine the flow.

You can calculate flow thru a 1/2" pvc pipe just using pressure, at X psi you get Y gpm.

Just get a pressure gage, 5 gallon bucket and a stop watch.


November 16, 2012
by JimFrederickson
JimFrederickson's Avatar

Thanks for the suggestions.

The Project is going to be at my house in the Philippines.

Water use there has been quite unpredictable.

While this, by itself, does not impose any type of "financial hardship" I am just quite "curious" as to why this is.

I am at the Philippine House about 2-3 months out of a year. (Sometimes less.)

There are people staying there to watch over the place and take care of the plants. (Although the "effectiveness" of the "plant care" has been an EXTREME SORE POINT for me...)

In the United States, my water usage is quite predictable and quite consistent. So these "wild swings" are not something that I am very used to. (Water usage can change 2x to 3x between what is the "norm/average". We are NOT talking about a simple month-to-month average, but rather a combination of usage by month over about 3yrs with month-to-month factored in as a base.)

I don't think that the "pressure/volume calculations" will be as accurate as I would like. Pressure here, the Philippines, is QUITE inconsistent and NEVER to what is considered "normal" in the United States.

All of the places where there is water available has already been split into separate water lines. (Master Bathroom, General Bathroom, Kitchen, and Outside...)

Most likely I will end up using several different monitoring methods.

What I have in mind now is:

1 - Water metering for the incoming water line.
2 - Water metering for each of the aforementioned locations.
(With the addition that probably the 3 outside faucets will each have separate
meters as well as a meter for all of the outside)
3 - Toilets will likely end up having some sort of data collection as to when water is
running into the toilet aside from metering.
4 - There is be metering on rain water collection and usage as well.

My intent on the "doubling up" on some of the metering is so that I can validate the accuracy of the given metering, as well as once that has been done possibly validate the accuracy of future changes in metering methodology/technology.

Ultimately in the Philippines I don't think that "public water" is really even necessary.

Water collection from rain water seems as though it is more than enough to supply the House. I just need to determine how much water needs to be stored in order to get through the stretches of no rain! (Which really seems only to be the 3 weeks in April...)

This is a longterm project on Data Collection. Whatever I end up doing here in the Philippines will be brought back to my house in the United States as well.

I intend to install the first portion of metering, whatever that ends up being, in May 2013.

Then later in 2013 I will make some additions.
(Perhaps even complete what I have posted here.)

I will update this thread, but not really on any schedule. Mostly just when something gets done.
(So likely around May 2013...)

Thanks for the input and the options...

(Just as a NOTE: I am NOT a "true bean counter". I can't tell you how much I have, but I know it is positive each month. I am REALLY only curious as to where the "beans" come from and where they end up going when and why... The "actual beans" don't have do much meaning to me, as long as they don't run out! :) )

December 06, 2012
by mongo
mongo's Avatar

Here are a few methods:

1, A weir. Basically, a box with a vertical channel or slot from which the water or other liquid flows. The higher the volume, the higher up the channel the liquid flows. I have used this method for calculating flow rates in waste water flume installations. Reliable but not precise linear function.

2, Turbine flow meter. Usually a small turbine in the flow of the liquid. I find them to be a bit finicky as they are a constant maintenance issue. The edges of the blades are usually sensed by a hall effect transducer and frequency is converter to flow rate.

3, Eddy current/flow disruptive measurement. This one is a flow obstruction which creates eddy currents and harmonics as it passes. Most are in the ultrasonic range and the output is usually an analog signal.

4, Coriolis flow measurement. This one is very common in the petroleum and chemical industries. It uses the Coriolis effect in a tube that vibrates at a set frequency. The passing of fluid, including gaseous as well as liquid, is detected by phase shift between two points in the tube. Output is scalable in both digital and analog modes.

5, Positive displacement. There are several types. One is like your water meter outside, which is a rotating wobbling disk. Rotations are translated into output. Another is through geared cavities, like a Roots blower. Mostly used for measuring things like natural gas in high flow systems like boilers. And there is the piston type. Each stroke is a specific quantity and each stroke counts this amount. Works in both liquid and gaseous media. And then there is the rotary vane, mostly used for thicker liquids like shortening and syrups. These are found in food processing plants but not exclusively. They have been found in numerous other applications too.

Here are some company names that you can research. LC (Liquid Controls), Roots, Dynapar/Veeder Root, Emerson-Micromotion, Honeywell... There are a lot of them out there. These are just a few that I have worked with.

Post a Reply

Please log in to post a reply.

Did you know that a microcontroller can measure an RC time constant? Learn more...