Following on from Part 1 (additional weatherproofing), this (rough and ready) electrical modification replaces the single reed switch inside the Fine Offset anemometer with 4 Hall sensors to enable a 1 mph resolution at all wind speeds in a 0.33 second gate measuring period. Actually, not much information is gained in measuring wind speeds at time intervals less than about 0.5 seconds due to the physical response function of typical cup anemometers to varying winds. Although not necessary for near real time wind speed monitoring, the additional Hall sensors do improve the wind speed resolution at very low wind speeds measured over small fixed gate periods.
For ease of removal 4 Hall effect sensors (Honeywell SS443) spaced 90 degrees apart around a 40mm radius, are soldered to a piece of veroboard that contains 2 holes that align with the existing reed switch mounting points. The Hall sensors are electrically linked in an AND gate configuration using 4 diodes and a resistor.
If any of the Hall sensors goes logic low (when it senses a magnetic field above its specified threshold) the AND gate output goes low. The signal is sent down a wired connection via an open collector transistor, and, as the attached oscilloscope photo shows, works fine over a 60 meter cable run with a 'hairdryer blowing on cups simulated' wind speed of about 25mph. The wired connection also includes the output from a single Hall sensor (as shown on the oscilloscope trace) just in case it's needed. By positioning the Hall sensors 90 degrees apart (and with a little positional tweeking) there can be equal logic high and logic low signals from the AND gate output across a 360 degree anemometer cups revolution.
This 50% duty cycle signal means that we can count all logic changes (both logic high to low, and low to high) instead of counting either logic high to low changes, or low to high changes. This doubles the wind speed resolution for a fixed measuring gate period, and, for the modified Fine Offset anemometer (now having 8 logic transitions/revolution), gives a 1 mph resolution at all wind speeds in a 0.33 second gate measuring period. It does however add about 30mA of current consumption compared to an unmodified anemometer.
Part 3 will give some details of a homemade anemometer also tailored towards near real time wind speed measurements using mainly household objects - should anyone wish to get their hands 'dirty'.
For ease of removal 4 Hall effect sensors (Honeywell SS443) spaced 90 degrees apart around a 40mm radius, are soldered to a piece of veroboard that contains 2 holes that align with the existing reed switch mounting points. The Hall sensors are electrically linked in an AND gate configuration using 4 diodes and a resistor.
If any of the Hall sensors goes logic low (when it senses a magnetic field above its specified threshold) the AND gate output goes low. The signal is sent down a wired connection via an open collector transistor, and, as the attached oscilloscope photo shows, works fine over a 60 meter cable run with a 'hairdryer blowing on cups simulated' wind speed of about 25mph. The wired connection also includes the output from a single Hall sensor (as shown on the oscilloscope trace) just in case it's needed. By positioning the Hall sensors 90 degrees apart (and with a little positional tweeking) there can be equal logic high and logic low signals from the AND gate output across a 360 degree anemometer cups revolution.
This 50% duty cycle signal means that we can count all logic changes (both logic high to low, and low to high) instead of counting either logic high to low changes, or low to high changes. This doubles the wind speed resolution for a fixed measuring gate period, and, for the modified Fine Offset anemometer (now having 8 logic transitions/revolution), gives a 1 mph resolution at all wind speeds in a 0.33 second gate measuring period. It does however add about 30mA of current consumption compared to an unmodified anemometer.
Part 3 will give some details of a homemade anemometer also tailored towards near real time wind speed measurements using mainly household objects - should anyone wish to get their hands 'dirty'.