Simple flow sensor wanted

Try RS Components.

Is the liquid clear or coloured?, can you use an LED/phototransistor
combo?.

It varies between clear and relatively opaque.

The application is a crop sprayer where various chemical mixtures
(herbicides) are transferred from an auxiliary tank mounted on the
front of the tractor.  The electric motor driven roller pump used
doesn't like being run dry for any length of time so I need to be able
to accurately detect when the tank is empty.

Conductivity is a possibility but past experience tells me that
keeping electrodes sufficiently clean is likely be a problem.

Restricting the flow and sensing the resulting pressure is also a
possibility but isn't ideal as it will both reduce the flow rate and
increase the motor's current draw.

I note that RS have an extensive range of sensors and switches (thanks
I.F.), which I haven't fully checked out as yet, but most of them are
more complex, and hence more expensive, than what's probably needed.
The simpler ones are also physically smaller than I want  (mostly ¼"
diam) but it may be possible to use a bypass line for sensing. The
transfer hose is ¾"  (19mm).

All ideas welcome.

plastic
restriction.

Flow is not easy to measure reliably. That is why sensors are a bit
expensive and each type has its drawbacks.

In your application, the most simple and rugged types in the RS catalog
are 256-562 and 256-578. Designed for water, metal bodies with
compression fittings, rugged, simple microswitch output. Price does not
appear unreasonable.

 They work by a spring loaded vane in the flow, IIRC. I have used these
before and they are fine. Like most of these types, they may fail in
either state, unlike the rotary types (but more prone to wear and also
require more electronics in interface).

Either 256-578 ($76.60) or 257-082 ($38) look suitable for the job,
and don't cost an arm or a leg.  The latter is a plastic body which
isn't necessarily a disadvantage.

I'm a bit confused with the plumbing fittings though.  22mm is the
approximate OD of 3/4" galv pipe, whereas I'd have expected them to
compatible with copper tube... or am I just being confused by the
terminology used?

Presumably the compression fittings will be a match for 3/4" copper?



This sounds to be a very suitable system for my application as I don't
need to measure flow, only detect it's presence or lack of.
Reliability, or lack of, is something I'm used to living with.  :)

Not sure, but I would not assume it. I'm no plumber so don't understand
their stupid pipe histories but you might have to be prepared to make an
adapter, preferably soldered copper tubings. I think somehow we had the
right size copper tube when we used one.


I would recommend this type over complicated DIY electronic solutions
which will be unreliable and probably frustrating to calibrate.

If you are mechanically handy and have time, another quick and dirty
system is to have a section of your pipe made from a flexible pipe such
as poly, canvas or whatever. You can put a spring loaded welding clamp on
the hose and rig a microswitch so that as the pressure drops, the clamp
closes a little and the handle parts of the clamp close the switch. How
that is done is up to your clerness. Relies on a pressure drop on the
output of the pump in event of flow stopping though. Will not detect a
blockage on the outlet either.

the threads beneath those compression nuts should match the threaded adaptors
used for 19mm plastic pipe too...
19mm is just the metric name for 3/4" pipe. it's still 3/4".

Bye.
   Jasen

I've got problem with that, but WTF is 22mm -- which is the size these
come in.

The 1/2" equivalent appears to be called 15mm, and the 3/4" equivalent
is called 22mm!

It's the size of the threaded part.  pipe is sized by the diameter of the hole
(as that's the most important part) but the couplers are sized by the thread
(as that's where the action is)



Yeah that sounds about right I haven't dealt with that stuff for over a
year now.

Bye.
   Jasen

 A simple mechanical float and microswitch arrangement at
the bottom of the tank perhaps?


G

Just put a pressure switch after the pump. when the tank is empty the
pressure will drop.

Bye.
   Jasen

What sort of pressure drop do you have in mind?

The frictional loss in 19mm tubing flowing 20 litre/min is something
less 3 metre of head per 30 metre.  (A google on Hazen-Williams should
find the various tables for frictional losses)

I'm pumping through 6 metre of hose, which will give around 600mm head
at the pump. This works out to be around 0.1 bar (1.5 psi) pressure.

I'd also suspect that the problem would be compounded by the fact that
I'm feeding to the bottom of the receiving tank (to minimise the head
pressure, and hence the pump load).  This will have the effect of
significantly varying the static head at the pump outlet, depending on
the level of the main tank and the slope the machine is working on
(which can be relatively steep on occasions -- 30° or so).

If you want a DIY solution, can I suggest a plastic ball or slug with an
embedded magnet operating a reed switch on the outside of a short piece of
tubing next size larger than your delivery tubing. The ball or slug will
need to be slightly larger than the ID of the delivery tube,  and have some
kind of restraint to limit its travel. If you cannot arrange the flow
vertically up, you would also need a light ss spring ahead of the ball/slug.
Food for thought.

Perhaps a T piece in the pipe which feeds a pressure sensor?  No
restriction in flow but requires some pressure in the pipe.

Altenatively look at impeller driven units.

Alan

http://www.discovercircuits.com/PDF-FILES/NewPDF/Capmeter1C.pdf

Hope this helps.
Kim

John_H wrote:


Check out
http://qualitrolcorp.com/productdetail.asp?ID36 %
This is a flow gage to verify oil flow in oil cooled power
transformers. Use to verify that oil is flowing when the pump is
running.

Dan

John,
     A nice simple solution would be to place a thermistor in the liquid
flow, pass some current through it and it will heat up. When the liquid
flows it will cool the sensor, no flow, no cooling. A simple comparitor
circuit can then turn off your pump and the circuit itself, so not too
much heating.

The whole circuit with a Relay to drive your pump should be under $20.

Regards,
         Brenden Ede


John_H wrote:

Brenden wrote:



This would only work if the liquid being pumped was at a different
temperature than the liquid that was stagnent because the pump wasn't
working. The fact that the liquid is moving doesn't mean it is cooler
when moving.

Dan

Dan,
     Not quite, if you apply current through the sensor and the fluid
isn't moving, it heats the surrounding stationary fluid. When the fluid
is moving the heat is passed into the moving fluid which keeps the
sensor cooler. This technique can be used to detect movement in any
fluid or gas that is in contact with the sensor.

Brenden



Dan H wrote: