DIY Liquid Flow Switch

Is this idea feasible:

I want to make liquid flow switch to use in a home brewery I'm building. I'm only realy interesting sensing that liquid is indeed flowing and don't care much about knowing flow rate. Store bought options are usually made in China and I have no idea what all the material are and if they are safe for food contact at the high temps I will need.

My idea, simply use a small length of CPVC with two pieces of copper wire inside, the wire would be inserted into holes drilled into the pipe and not be touching each other. The idea is that if liquid is flowing over the conductors then I should be able to sense that, somehow...? The pipe would be oriented vertically so that there can't be standing liquid.

Is this feasible? Would this be reliable? Would I need some kind of amplification? I assume so, can anyone suggest a circuit or components for the amplification?

All thoughts are greatly appreciated.

If the pipe is clear, and the liquid isn't you can detect that with light.

if the pipe is clear and the liquid is, you could also do that, but would need to detect the bend in the light difference between no fluid and fluid.

You could use a method I’ve seen used on shower pumps. Inside the pipe you have a small capsule. This is like an elongated doughnut. Naturally when liquid flows, it is inclined to move with the liquid but has it’s progress through the line stopped by a fixed ring inside the tube. As it rests against this end stop, a magnet within the capsule causes a reed switch on the outside of the tube switch on.

This whole assembly is oriented vertically, so when no water is flowing, the capsule falls back out of range of the reed switch.

Edited to add diagram.

Can you elaborate on using a light sensor, maybe provide a link to an example?

Regarding the capsule, how much would that impeed flow?

How coloured is the liquid?

In a steam loco, the water level is detected by sight, but has a background that has a line on it which shifts depending on whether water is behind it or not.

If the liquid is dark or cloudy, you can use an LDR and a light. (In an enclosed hounding)

If not you will need to experiment with a small beam on an LDR also in an enclosed housing and look for refraction to throw the beam off or on the LDR depending on whether liquid is there or not.

A pressure sensor installed perpendicular to the pipe will detect even rather small flows, as a result of the Bernoulli effect. There is a significant pressure drop when the liquid is moving.

By the responses I take it my original idea isn't very good?

Its beer brewing so the liquid color will range from light golden to black. Can I use the light methed with silicon tubing? Its sort of translucent, especially if a strong light were shined on it.

I thought about a pressure sensor on the tubing too, though I had dismissed it as I guessed it could be error prone. So this is a viable option then?

I like KenF's capsule magnet thingy.

(As long as it doesn't somehow detach and end up in some punter's pint.)

JimboZA:
I like KenF's capsule magnet thingy.

It's exactly what I found in a shower pump and surprisingly sensitive. As soon as you turn on the tap, the slightest dribble activates the pump. Once the flow has been stopped, (by closing the taps) the pump turns off. The shower pump also has one on each supply (hot and cold) so it has built in redundancy.

I can't think of how to implement kenf's idea with readily available parts. Keep in mind food safe and cheap components are primary goals here.

I gave my idea a preliminary test, I made a voltage divider using a 22k resistor and a pair of 22gauge solid core wire as my sensor. I striped about a 1/16 inch of insulation off the ends of the wires and fixed them about 1/4 inch from each other. Just sticking this sensor under a running kitchen faucet I could clearly see results: 1023 with no running water, 300 to 500 in running water. I know I'll need to orient the sensor in such a way that standing water can't cause false positives, other than that is there any reason I shouldn't bother with this idea any more?

A popular food safe flow meter...
http://store.kegbot.org/products/swissflow-sf800-flow-meter

Hi,

I gave my idea a preliminary test, I made a voltage divider using a 22k resistor and a pair of 22gauge solid core wire as my sensor. I striped about a 1/16 inch of insulation off the ends of the wires and fixed them about 1/4 inch from each other. Just sticking this sensor under a running kitchen faucet I could clearly see results: 1023 with no running water, 300 to 500 in running water.

I think you will find that the 1023 without running water was in the air, so different medium.
Try it in a cup of water and then stir it.
Running tap water has a connection to earth and you could have been detecting a volt drop due to earth currents.
Copper in your brew or water may not be a good idea as any voltage detection will probably be due to current flow and some chemical reaction with the electrodes.

Tom… :slight_smile:

I doubt you'll see any difference between running water and static water. The difference between air and water is much more pronounced.

If the fluid is not pressurized this might be a viable solution. Very simplistic, low cost and easy to build yourself. If the fluid is pressurized the solution cam be modified by a expansion volume above the float.

theboozler:
I can't think of how to implement kenf's idea with readily available parts. Keep in mind food safe and cheap components are primary goals here.

With a simple piece of tube and a rod, you could easily make a mold. Pour in a little epoxy, once it's started to go off add the magnet and more epoxy to fill the mold.

Take a short length of standard plastic tube(as would be used in plumbing). cut a ring from this. snip out a small section of the ring so that it can be compressed into to fit into the tube. Apply a little solvent adhesive before putting it in place. Drop in the capsule you molded earlier and then the second ring.

A simple cable tie would do to hold the reed switch in place. Now just include this piece of tube in your plumbing using the standard fittings.

I suppose a little electrolysis would be happening with my idea, were only at 5 volts and a few miliamps, would/could this cause corrosion or contamination issues?

theboozler:
I suppose a little electrolysis would be happening with my idea, were only at 5 volts and a few miliamps, would/could this cause corrosion or contamination issues?

There are actually many circuits available for something similar. Some use AC to prevent the effects of electrolysis. But these are all just to detect the presence of water, NOT to sense it flowing.

I did a quick drawing to illustrate what I’m trying to achieve. The project is a RIMS setup, liquid flows out of a beverage cooler, down to a pump, then pumped into a RIMS tube where it flows over a water heating element, then it flows back to the top of the cooler. The element is controlled by the arduino and a PID algorithm to maintain very specific temperatures up to 168*F.

I intend to use this flow switch as a safety check, it will not allow the heating element to fire if there is no flow detected.

Another way of detecting flow is measuring he pressure difference on both sides of an internal stricture.

Here’s a circuit I’ve used successfully to detect the presence of water. If you were to place a couple of probes beneath the flow of the water coming in, I’m sure it would work. It gives an active LOW output (ie low= water present, high= no water present)

BTW the transistor used was BC548C

waterSensor.JPG