AVR Freaks

Are you wanting to measure a continuous flow of water or measure an amount into a container?

Looking online I can find sensors that will measure down to 0l/min but they don't use any moving parts. If measuring into a container could you do it by weight? Can you use a pump? Continuously pump from the bottom of the tank to the top. Take your feed to your flow sensor from the output (higher pressure) of the pump. Your flow rate to the sensor will go up. You will probably need to restrict the discharge into the tank to raise the pressure.

You already seem to have at least two threads running on this:

https://www.avrfreaks.net/forum/m...

https://www.avrfreaks.net/forum/e...

It would have been helpful to at least reference them - you are already getting some of the same answers again that were previously given in those threads!

And why is this in the Evaluation and Development Kits forum? Is this in any way AVR related or do we move the thread to Off topic?

And now there's yet another one: https://www.avrfreaks.net/forum/e...

Well, it could be the characteristics of the sensor -- or it could be your code/approach.

Generally, pulses per liter (or pulses per gallon) are nominal and calibrated at a certain flow rate.  And generally for water at a certain temperature.

With any given flow rate and given fluid, the PPL needs to be recalibrated.

It could be as simple as your orientation of the flow meter.  Some are designed to be horizontal -- e.g. inlet on the left and outlet on the right.  Some are designed to be vertical.  You could get your symptoms with the wrong orientation.

A high-flow sensor with heavy "paddles" indeed may not turn at low flow rate but allow leakage past the paddles.

In any case, use a frequency generator (or create one from an AVR, timer, and CTC mode) and test your code at various constant flow rates to ensure proper processing.

I forget whether you are using the "pulse" type or "square wave" type of flowmeter.  The pulse types are tricky.  IIRC I commented on that in one of your other threads.

Here is the link that you gave for the "datasheet":

http://www.seeedstudio.com/wiki/...

But now you say:

The manual stated 450 pulses for 1 liter of water

???  I don't see 450 in there anywhere.  One can infer, I guess, that it is about that value.  A note says about 50Hz at 6lpm.  3000 pulses in one minute for 6 liters.  About 500 pulses per liter?

awneil wrote:

 

You already seem to have at least two threads running on this:

LOL--OP appears to have abandoned the flowmeter approach to dispensing exactly one liter of water, to ways to fill a container with volume of exactly one liter.

I'll throw another approach into the mix:  For industrial chemical dispense, we've gone away from flowmeters in many cases, and tend to get better results with a vessel and fill it by weight to the desired amount (specific gravity and all that).  These chemical are very hard on flowmeters.  Next thread will be on load cells...  (but OP mentioned water)

Back to the highlighted "exactly" -- tell what that means in this app.  What accuracy and repeatability is needed?  Is it really volume-only, regardless of the density of the fluid?  I.e., does temperature matter?

I have a customer who uses flowmeters to monitor cooling oil flow.  They are the paddle variety and their range is 5 - 50gpm and as noted lower than minimum flow rates, the paddles stop turning but the oil keeps moving. 

Lee, I like the idea of fill by weight, great idea.

OK - not directly as a question, but:

No, I can't see that he ever has stated that tolerance.