Posted by Kevin on February 24, 2006, 10:52 pm
An accurate measurement of flow rate is one of the biggest problems
here. If you have an AC pump connected to a differential controller,
you could make a fairly educated guess as to your flow rate. If you
know the amount of time the pump is on, and the Delta T, BTU/hr is a
Perhaps a small turbine water meter with a counter might work.
Posted by Steve Shantz on February 24, 2006, 11:32 pm
I agree whole heartedly about the small turbine water meter. I've got
one in my water softener. But can they be purchased? And can they
take the heat?
As for the AC pump connected to a differential controller... I've got a
DC pump connected to a solar panel. Won't work for me.
Also, to critique my earlier idea about measuring the pressure drop
across the pump...
The flow rate would be somewhat proportionate to the pressure drop,
(especially in the moderately low flow rate in my system) but the
proportionality constant is unknown. Knowing just the pressure change
across the pump isn't quite enough to calculate GPM.
One could make an educated guess based on delta Temperatures at noon on
a sunny day, but even this is not really accurate, as the temperature
and efficiency of the system would have to be factored in.
I like your turbine idea best.. if available.
Posted by daestrom on February 25, 2006, 5:37 pm
If the flow is very slow, the differential pressure needed is proportional
to the flow rate (laminar flow conditions). But that is *very* low flow,
and you probably don't want to be operating in that range anyway. For
turbulent flow, the differential pressure is proportional to the flow rate
Flow = K * sqrt(dP)
If you can find just *one* flow-rate versus differential pressure data
point, you have it made. But keep in mind that the exact constant will
change with the viscosity of the fluid (ratio of water/glycol as well as
temperature), and any degradation of the pipe interior roughness
(scale/corrosion build up). So it's best to recheck your 'calibration'
every year or two.
You might try these folks. I've worked with them and they probably have
something that suits your needs. Of course, price varies with accuracy
Then you have to interface it to 'something', either a recorder, or
datalogger, or computer....
Posted by Steve Shantz on February 25, 2006, 6:47 pm
Fluid dynamics is never quite as simple as I wish it would be. I
should have known better!
Excellent source. $9 for a turbine sensor that goes up to 4 GPM.
Handles temps up to 100 dC.
I'm thinking about using a PIC to interface with temperature and flow
sensors, and interface to a PC. It would be a good learning project.
I just need to find some time. So many other cool things to work on!
Posted by Kevin on February 25, 2006, 11:27 pm
Apparently the this discussion thread happened many times before and someone
decided to produce a product to solve the problem.