How can I make a 1-10 liter/hr flow meter?

[Sherwood Botsford]

I'm working with tap water. I'm building a fertilizer injection system, and need to control the rate that water flows through the bypass, through the fert tank, and back into the main line.

At small flow values, reynolds numbers are on the non-flat part of the curve, so making something that is even semi-linear is difficult. I've looked at rotameters, but machining a smooth conical bore is beyond my skill

At this point, my best notion is to have a thin leaf of springy material that when flat, meets the opposite wall, but with increasing flow bends to leave a larger opening between the leaf and the opposite wall.

* What to make the leaf out of?
* What shape of chamber makes it possible to easily tell the difference between say 6 l/m and 10 l/m
* Where can I find small valves for the metering?

Pressure across the the metering system is fractions of a PSI. Pressure compared to outside is 30-75 psi.

 

[BvU]

I'd go for a rotary kind like they use in dishwashers. Dunno about acuracy, but you can calibrate separately. The sensor is cheap, reading out can be a nice https://maker.pro/arduino/tutorial/how-to-interface-arduino-with-flow-rate-sensor-to-measure-liquid.

Do some googling ...

 

[jrmichler]

Here you go: https://www.mcmaster.com/#5079K12. Exactly the flow range you want, rated for more than your pressure, and +/- 4% rated accuracy. Only USD $67.40 plus shipping. I don't know if they ship outside the continental US.

 

[berkeman]

and need to control the rate that water flows through the bypass,

How about using a pump that you can control the flow rate with? Something like a peristaltic pump from IV infusion devices, but sized a little bit bigger.

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

 

[Sherwood Botsford]

Hmm. Don't seem to be able to edit a post after the fact.

I need to add: No electronics. Visual only.
Accuracy isn't important. Repeatability is. In effect everyone will do their own calibration using a conductivity meter.

Trying to make a device prototype needs to use only 10-20 bucks worth of stuff, and can be mass produced for 2-3 bucks each.

Following ideas may also work:

V shaped chamber, made from a triple layer of plexiglass, and coloured chaff denser than the liquid. While the Re would be low enough to be erratic I think I would see a band of chaff where the upward forces on average canceled gravity. 1 mm aluminum beads? This would emphasis the imprecise nature of the measure.

Paddle on a pivot. Problem here is the pivot. At 1 cm² opening 1 l/hour is moving only 3 mm/sec. Make the opening smaller and you get gauge that is hard to read.

Another method is the drip chamber used for medical IV's. This has transient issues: The air chamber shrinks to 1/4 of it's volume as the system starts up (assuming a supply at 45 psi) then at shutdown, it will force a slug of fertilizer concentrate into the line.

One way might be to make a 'pin ball game' Again a 3 plate flat chamber, but with the input coming in at a corner. The inlet flow is parallel to one side of the chamber. Some form of heavier than water shot is placed in. beads roll down the slope get caught in the jet, and are lifted a space, eventually falling out of the flow. At faster flows they get carried higher. Again, I suspect this would be noisy, with balls flying to different heights.

Or a waving ribbon. Put a flat blade in the flow that sheds vortexes. A ribbon is pushed back and forth as the vortexes flow boy.

Flat Variable area flow meter. Use a wedge shape indicator in a V shaped slot. I'm not sure in this situation how to keep the wedge centred, or how to prevent it from sticking. At present my thought is to make it slightly T shaped. If one side approaches the side of the V, the flow is restricted, that side drops, the T tilts and more of the flow is diverted to that side. However at the flow rates involved, inertial forces are small compared to viscosity forces. Not sure if it would work. Some rotameters seem to have bead with a rim around it at the top that I suspect does this.

In addition, if using a flat flow meter, there are forces on the flat faces. How best to keep them out of contact, but still have clearance.

Here you go: https://www.mcmaster.com/#5079K12. Exactly the flow range you want, rated for more than your pressure, and +/- 4% rated accuracy. Only USD $67.40 plus shipping. I don't know if they ship outside the continental US.

I'm trying to make the entire injector for $40. The idea is that a pipe section with a small restriction in it will generate a pressure difference across the restriction. A small hose tapped into 'before' and 'after' can run through a pressure tank (withstand standard hose pressure) filled with soluble granular fertilizer. Water enters at the bottom, exits through the top. returning to the 'after' connection. Pressure tank is a section of 4" PVC pipe with a cap on one end and a clean out port on the other.

How about using a pump that you can control the flow rate with? Something like a peristaltic pump from IV infusion devices, but sized a little bit bigger.

Needs proportion. When the timer stops the flow of water, the pump would continue attempt to pump concentrate into the line. Could starve it for input, but the pump keeps running. Requires a power.

Which brings up another design point:

In normal use it's set. Water is started, and stopped. by an inexpensive battery operated timer. Under normal use, once adjusted it may not be touched again for months.

***

A venturi asperating solution from a bucket is the traditional way this is done. But:

* Very dependent on flow rate. Increasing the mainline flow rate by 10% makes a MUCH larger than 10% difference in the injection rate.
* Requires a large pressure drop (~15psi) across the asperator.
* For short runs that don't use all of the solution in the bucket, there is a risk of siphoning the remaining solution when the timer shuts the water down.
* For long runs that use all the solution it will entrain air into the line. This oxidises the iron and makes it plate out on the drippers clogging the drippers.To control the flow rate, I need a gauge of some sort with a needle valve. Typically for a 1000 liter per hour hose run I want 1 to 10 lph of saturated dissolved fertilizer.

The device is used far from the nearest electrical source, so an electronic method is out, both for power and for complexity reasons.

 

[jrmichler]

A rotameter does not necessarily need a tapered bore. You can use a series of drills to make a stepped bore. I did exactly that once using a piece of acrylic (Plexiglas) rod. You would have to do some experimenting, but McMaster-Carr sells 304 and 316 stainless steel balls in many sizes. Try a 3/16" diameter ball, then drill a stepped bore with number drills from, say, #11 (0.191") through #1 (0.228"). Use a piece of 3/4" to 1" diameter acrylic rod, and cut male pipe threads on each end. Drill through with a hole smaller than the ball, and add a screen on top to keep the ball from being blown out by excess flow.

 

[Sherwood Botsford]

Cool idea. Will a drill leave a smooth enough surface that the ball doesn't catch, and be reasonble to read? My experience with acrylic/plexiglass is that any machining results in a very milky surface.

 

[jrmichler]

Not perfect, but good enough, especially with water in it. Make sure your drill bits are sharp. Try it and find out. McMaster-Carr (www.mcmaster.com) has 3/4" acrylic rod for $6.08 per foot, with a one foot minimum.

 

[jim hardy]

Perhaps one could adapt a tapered tube that's already mass produced for something else?

First return from a search on "Tapered Rain Gage"

 

[Sherwood Botsford]

Progress report

Rain gauge is an interesting idea. I'm a bit skeptical about it's ability to support 30-75 psi internally, and adapting fittings would be hard. For the low flow for my project the ball wouldn't rise much.

I have found that there are companies that make tapered drill bits -- quite cheaply too -- for drilling pilot holes for wood screws. Sequence would work something like this:

Drill 1/8" hole through block to guide subsequent
Drill tapered hole to near bottom of block
Drill bottom hole so that ball can't wedge. This is just bigger than ball
Target right now is to use a 3/16 ball bearing, so 7/32 inch hole
Drill top of hole and bottom of hole for threaded 5/32 ID (5 mm nominal) hose fitting.
Tap top and bottom hole for hose fitting.
Cut disk of plastic screen wire to fit bottom of top hole, to keep ball from closing off top hole with heavy flow.

Options.

Buy two tapered drill bits. Use one in reverse with water and a fine grinding compound to polish the taper.

 

[jim hardy]

I'm a bit skeptical about it's ability to support 30-75 psi internally,

wow i missed that requirement. My bad.

 

[rokytnji]

What is the flow rate?

What is wrong with using one of these? They are cheap and plumb able.

Just curious because I run a home made Mexican solar still for distilled water out of 2x4's and plywood and a old plasma tv screen.