I'm doing a project to measure wind speed using thermal anemometer principles. So far, I've created the device using a heater and an lm35 sensor (temperature sensor). If I wanna measure wind speed using the temperature loss of the device, how do I calculate the wind speed?
Just calibrate the sensor. Pick a windless day and drive around at various speeds, while recording the speedometer and sensor readings. Then fit a suitable curve to the results.
The readings will depend strongly on the ambient air temperature as well as the wind speed, which makes this sort of anemometer uninteresting to most people.
Professionals do use them, but only after extensive calibration in a laminar flow, temperature controlled wind tunnel.
1 Like
If I were to used mathematical equation to measure the speed, what would be the variables I needed to do that?
What mathematical equation?
Hi,
Exactly, calibrating empirically, using the real world, like a motor vehicle, will give you a more accurate calibration curve than using an algorithm.
Can you please post an image(s) of your project and the anemometer elements.
Did you Google ;
arduino thermal anemometer
Tom... 
I don't have the schematics for all of the components but here are the components I've used so far.
I used 2 LM35 sensors for 2 purposes first sensor is to measure the constant temperature, the second sensor is to measure fluctuating temperature (using arduino serial monitor and multimeter showed the same result in windless environment). 10mV = 1 degree Celcius. For the heater, I used 3d printer ceramic cartridge heater (I don't have the datasheet because I couldn't find any). This heater is controlled using pid controller to maintain constant temperature. The differential amplifier is used to calculate the heat loss based on voltage difference (then the value is amplified 100 times to make sure arduino could read the low voltage from measurement)
Hi,
Thanks for the process explanation, you could have done the differential amplifier in software and two analog inputs.
That way you could monitor the performance of your project.
Thanks.. Tom...
1 Like
Is it possible to use algorithm (or mathematical equation) to determine the wind speed from either heat loss or voltage loss?
No. There is no explicit algorithm or mathematical expression relating those quantities for your device.
The basic thermodynamic and fluid dynamics laws (differential equations) can be used to model the behavior, but that step depends very strongly on the details of the physical construction and the properties of all the materials involved in the device. In any case the results would have to be verified by experiment.
To calibrate your device you will need a third sensor to measure the ambient air temperature.
You have come far enough along with the construction phase of this project. Don't you think it is time to start doing some research of your own to learn how they actually work?
Yes. There are thermal air sensors like THIS
$25USD
The Rev. P6 wind sensor is the latest revision of our wind sensor, revised to increase stability and accuracy. We also added a high precision potentiometer to allow easier calibration. All boards are calibrated in the shop, so the potentiometer shouldn't need to be changed by the user, although it's there to experiment with if you so desire.The Rev. P wind sensor is an improved hot wire anemometer, similar to our Rev C wind sensor, but including hardware compensation for ambient temperatures. The P refers to positive temperature coefficient thermistors that are used in this model. These thermistors are more stable and more absolutely accurate than the thermistors on the Rev C.
These thermistors require a higher voltage to heat the sensing element to operating temperature. Thus the sensor needs an 9-12 volt supply. One pleasant side-effect of this increased voltage is that there is plenty of power to heat up the thermistor, so the sensor can sense hurricane force winds without saturating (going to a flat maximum value). (It does not necessarily mean the sensor can survive being in hurricane, apologies storm hunters).
The new mounting of the thermistor sensing element means that the Rev P is much less directionally sensitive than the Rev C. sensor, closer to the 15% range. The sensor is physically more robust than our older sensor due to the loop that holds the sensing element. Another new feature of the sensor is a dedicated ambient temperature sensor that is independent of supply voltages. The temp sensor analog output is also scaled to 3.3 volts. The voltage (wind) output of the sensor has also been scaled to a maximum 3.3V output with high precision (.1% resistors). This is to accommodate the increasing use of 3.3 volt boards, and microcontrollers.
It works great with any Arduino or clone, right out of the box, however you need to provide at least 9-12 volts from an external power source, with 12 volts being ideal.
The Rev P Wind Sensor is now available through our UK-based partner Digital Smarties. Non-US customers can order them here.
Specifications:
- Supply voltage: 10-12 volts
- Current: ~40mA - increases a bit with higher wind speeds
- Wind Speeds Measured: 0-150 MPH
- Ambient temperature compensation
3 Likes
This topic was automatically closed 180 days after the last reply. New replies are no longer allowed.