Calibration help for hall sensor?

[jcbennett (Jan)]

I am using this Hall sensor.

The specs for it are as follows:

• Stainless Steel 1" NPT Hall Effect sensor flow meter.
  Flow Range: 4 - 56 L/min (1-15 GPM)
  Flow Characteristics: F = 5Q-3 ±10% where Q is the flow rate in liters/min and F is the pulse rate in Hz (pulses per second)
  Flow Characteristics: See also the chart provided as one of the images.
  Cumulative Flow Conversion: 1L = 225 pulses ±10%
  
  

From this response, I have come up with the following calibration formula:

  - platform: pulse_counter
    name: "Water Pulses"
    id: water_hall_effect
    pin: 4
    update_interval: 250ms
    accuracy_decimals: 2
    filters:
     - lambda: !lambda |-
        if (x < 1020) return (x / 510);
        return ((x / 300) + 0.6);
    unit_of_measurement: "L"

However, after measuring out 1L Home assistant is showing over 12L, so I’ve messed something up somewhere.

Am I at least on the right track?

I am utilizing the Utility Meter integration to monitor.

[neel-m (Neel Malik)]

You might want to read this VERY long thread, which probably has some good tips in it for a sensor similar to yours.

Using ESPhome to build a water flow rate meter

Greetings fellow humans, I wish to build a thingy that tracks the flow of water from our borehole pump to the tanks. I am using a Wemos D1 R2, and a YF-DN40 water flow sensor (the big one). I notice that ESPhome has a Pulse Counter, which is basically how a flow meter works: there’s an inline rotor with a magnet in it, and a hall sensor that notices when the magnet comes round again. Before I go barking up the wrong tree, is the Pulse Counter the right component to use? I could hack something …

[Karosm]

According to your Flow Characteristics, conversion would be:
- lambda: return (x + 180) / 300;

but since the rate is not constant that cumulative conversion doesn’t make sense…

@jcbennett

Cumulative Flow Conversion: 1L = 225 pulses ±10%

[thusassistint (Ed)]

@jcbennett

From this response, I have come up with the following calibration formula:

  - platform: pulse_counter
    name: "Water Pulses"
    id: water_hall_effect
    pin: 4
    update_interval: 250ms
    accuracy_decimals: 2
    filters:
     - lambda: !lambda |-
        if (x < 1020) return (x / 510);
        return ((x / 300) + 0.6);
    unit_of_measurement: "L"

According to the specification, the given Flow Characteristics formula is:
F = 5Q - 3 with F in Hz (pulses per second) and Q in L/min.
So in this case the formula for the Flow is Q = (F + 3) / 5
And because F = x / 60 we know that Q = (x / 300) + 0.6
This is with:

• Q in [ L/min ]
• F in [ Hz ] or [ pulses/sec ]
• x in [ pulses/min ]

The useable flow range of this sensor is specified as 4 to 56 L/min.

The given flow formula is, similar as was shown in my previous post (the one you are referring to), again a linear function that does not go to zero when the pulse rate is zero.
But we want the flow sensor to show zero when the pulse rate is zero.
So we can again add a filter that makes sure that the flow rate goes to zero below the given minimal useable flow rate of 4 L/min.
We know that x = ((5 * Q) – 3) * 60
So at 4 L/min the x value is x = ((5 * 4) - 3) * 60 = 1020
So Q should be 4 when x = 1020 and Q should be 0 when x = 0
This can be corrected with a linear function for the range below 4 L/min or 1020 pulses/min:
Q = x * 255 (coming from 1020 / 4 = 255).
So the filter should be:

    filters:
     - lambda: !lambda |-
        if (x < 1020) return (x / 255);
        return ((x / 300) + 0.6);

The graphs below show the results of this.

The full range:

Zoomed in around 4 L/min to visualize the correction filtering:

The specified flow characteristic is shown in blue, the corrected flow characteristic below 4 L/min is shown in green and your flow characteristic filter is shown in orange.

In the table below all the results are shown.

@jcbennett

However, after measuring out 1L Home assistant is showing over 12L, so I’ve messed something up somewhere.

There is a mistake in your filter for the correction below 4 L/min, but I don’t think this can cause a twelve times larger value.
Do you have an idea about the flow rate that you were using while testing? Was this rate within the specified useable flow rate of 4 to 56 L/min?

Like @Karosm already states, the given Cumulative Flow Conversion of 1L = 225 pulses is misleading, because this value is not constant but changes with the flow rate.
In the table you can see the applicable number of pulses per liter for the given flow range. The value of 225 pulses per liter is applicable only with a flow rate of 2.4 L/min or x = 540 pulses/min, so this value is below the specified minimum useable flow rate of 4 L/min.