ultrasonic anemometer
Вставка
- Опубліковано 28 сер 2024
- This is my experimental ultrasonic anemometer.
The difference in the ultrasonic flight-time with the wind and the ultrasonic flight-time against the wind is used to calculate the wind speed.
The dimensions have been chosen such that the windspeed is numerically equal to 1/3rd of the flight-time-difference.
The code is a variation of the code used in this experiment www.instructab....
Came here from your laser anemometer. Love your projects, why is your channel name quality photos?
Long story ... LOL :)
Years back I ventured into photography.
Hello! It is very interesting what you have done here. I am working on something like this for my bachelor degree. I encountered your video and read the instructions you gave. I tried to implement those but I have some parasite values. Can you, please, explain what you mean by "The dimensions have been chosen such that the windspeed is numerically equal to 1/3rd of the flight-time-difference." ? I also didn't understand: windspeed =(TIME1-TIME2)/K. How to determine the K factor? It really confuses me. I want to learn more about this project. Can you, please, help me with this, sir? Thank you very much! Great project!
Assume two points A and B.
The time to travel from point A to point B against the wind will be:
Time1 = Time-in-still-air + windspeed.
The time to travel from point B to point A with the wind will be:
Time2 = Time-in-still-air - windspeed.
Subtract the times:
(Time1 - Time2) = (Time-in-still-air + windspeed) - (Time-in-still-air - windspeed)
Solving
windspeed = (Time1 - Time2)/2 ............... (1)
If the sensors are optimally spaced then equation (1) will be valid.
Otherwise the equation (1) becomes:
windspeed = (Time1 - Time2)/K ............... (2)
where K = calibration factor
In my case the sensor spacing worked out at (Time1 - Time2)/3
Helpful information:
The speed of sound in dry air at 20 degrees Celcius is 343 m/s = 343000 mm/s.
A windspeed of 1km/hr equals 0.277778 m/s = 277.778 mm/s.
One wavelength at 40kHz = speed-of-sound/frequency = 343/40000 = 8.575 mm.
Arduino microprocessors can only resolve time to 1us.
A more sensitive method is to measure the phase change between the transmitted and received waveform.
@@qualityphotos1 thank you very much for the info!
@@qualityphotos1 I didn’t get how u calculated the k factor 😕
please post the code and instruction as soon as possible.
The code and circuit is a variation of that shown in my article www.instructables.com/Dual-Sensor-Echo-Locator/. Just trigger BOTH sensors simultaneously and look for a response on the oppposite sensor.
@@qualityphotos1 so if i understand ... if i want to do an anemometer and add a perpendicular axe to calculate the direction of the wind i could use your multiplexe echo locator and trigger every captor at the same time ?
@@samuelperianincarpin4021 My method as shown in the video ... (call this the North-South-axis ):
(1)Do NOT tape the sensors. (2) Point the sensors towards each other. (3) Space the sensors about 2m apart (4) Trigger BOTH sensors together as shown in the code. (4) Listen on sensor1 for the signal from sensor2 ... this is TIME1 (5) Wait ... allow time for any echos that might get back to the sensors to dissipate. (5) Trigger BOTH sensors again together as shown in the code. (6) This time listen on sensor2 for the signal from sensor1 ... this is TIME2 (7) Now do some mathematics ... windspeed =(TIME1-TIME2)/K where K is a fiddle factor depending on sensor spacing
Explanation: TIME1= (time-in-still-air) + windspeed ; TIME2=(time-in-still-air) - windspeed; therefore TIME1-TIME2=2*windspeed; Juggle K until the windspeed is correct ... you can work this out using (sound-in-still-air)=343m/s and 1mph windspeed = 0.44704m/s
If you don't want to point the sensors into the wind you will need to duplicate the above method for the West-East-axis and apply Pythagoras.
Expect some jitter. The HC-SR04 sensors send eight 40KHz pulses then listen for an echo after which they stop listening. It is entirely possible that they miss the first echo and respond to the 2nd or 3rd pulse in which case TIME1 and TIME2 will vary slightly.
@@qualityphotos1 in step num 4
do u mean by trigger both sensor is making the trigger pin at both high at the same time?
@@lolasolomon1054 Correct
So if you use 4 of them you could actually make out wind direction?
Yes ... subtract the opposite NS and WE sensor readings and apply Pythagoras.
Have delayed posting the above code as am working an improved circuit ... the concept is the same but is much more responsive.
thanks but how does the left sensor detects wind with its back facing
The annemometer works on doppler.
We alternately measure the time taken for sound to travel a fixed distance "into" the wind then measure the time taken for sound to travel the same distance "with" the airflow. This is achieved by alternately using the HC-SR04 transmitter at one end and the HC-SR04 receiver at the other.
In still air the sound flight-time taken is the same in both directions ... if we subtract the two readings we get zero.
In moving air the two readings are different ... the number we get when the two readings are subtracted is proportional to the wind speed.
Can you give me that code and instructions? I need it emergency for my varsity project. My data is fluctuating so much. Please help me as soon as possible. I have no need so accurate.
Try smoothing your data.
An extremely fast method for smoothing N samples using one addition and one subtraction is described in www.dspguide.com/ch15/5.htm.
Sum (once only) N items.
Place these items into a circular queue
Point to the first (oldest) queue item
Subtract the first item value from the sum
Add the current sample value to the sum
Replace the first queue value with the current sample value
Advance your pointer
Average equals sum/samples
The code for a filter using this technique may be found here www.instructables.com/Laser-Anemometer/
Regarding my ultrasonic code ... it is not yet ready for release
Hlo sir, NIce video
im working on sonic anemometer project
using HCSR04 so can plz share the data of the project?
Sorry ... I have abandoned this project. The only details are those in my replies to the previous comments below.
@@qualityphotos1 Can u plz share the data which is available?
bcz im strucked in project examinations
im not able to get any information regarding this sonic anemometer project
plz help me
@@manupatil8762 Sorry ... I have no further data on this project
@@qualityphotos1 Can u share atleast wat data u have?
plz 🙏🙏🙏
I need to measure a wind speed generated by a fan. I have a space of 30 cm. How can I do it?
Measure the transit time in both directions and subtract the readings. In one direction the windspeed adds to the transit time of sound in still air. In the other direction the windspeed slows the transit time. (Still + airspeed) - (Still - airspeed) = 2 x airspeed. Sound in travels at 343.2 m/s in still air. 1 km/hr = 0.277778 m/s. Just do the numbers.
Otherv methods include ... a wind vane and a load cell ... or a Pitot tube ...
@@qualityphotos1 What I want to do is to measure the speed of the wind blown by a fan. to place it in front of the fan, with a distance of 15 cm between the sensors. If the margin of error is not very high, I think to implement it. Could you please describe it to me step by step?
@@berat8949 I already have ... see my first answer
can it detact a breath of human???
I want to put it on mask so it detact breath,can it do that???
Not with this circuit as (1) too much jitter (2) the transducers need to be spaced some distance apart.
Phase detection is more sensitive. Try using a CD4096B phase detector with the acoustic sound path fed into one input (pin 14) and the sound source fed directly into the other (pin 3). Adjust the signals for 90 degrees in still air ... this will produce a square wave output (pin 2). Air movement in the acoustic path will cause the mark-space ratio of the output to vary. It should be possible to adapt the sonic anemometer circuit diagram shown in www.instructables.com/Sonic-Anemometer/ .
hi i tried to go on the site but i can' t find the code and it was a very cool video btw
Sorry about that .. the code is a variation of the code used in this experiment www.instructables.com/Dual-Sensor-Echo-Locator/.
I will release the final code once I have sorted the jitter problem.
My method as shown in the video ... (call this the North-South-axis ):
(1)Do NOT tape the sensors. (2) Point the sensors towards each other. (3) Space the sensors about 2m apart (4) Trigger BOTH sensors together as shown in the code. (4) Listen on sensor1 for the signal from sensor2 ... this is TIME1 (5) Wait ... allow time for any echos that might get back to the sensors to dissipate. (5) Trigger BOTH sensors again together as shown in the code. (6) This time listen on sensor2 for the signal from sensor1 ... this is TIME2 (7) Now do some mathematics ... windspeed =(TIME1-TIME2)/K where K is a fiddle factor depending on sensor spacing
Explanation: TIME1= (time-in-still-air) + windspeed ; TIME2=(time-in-still-air) - windspeed; therefore TIME1-TIME2=2*windspeed; Juggle K until the windspeed is correct ... you can work this out using (sound-in-still-air)=343m/s and 1mph windspeed = 0.44704m/s
If you don't want to point the sensors into the wind you will need to duplicate the above method for the West-East-axis and apply Pythagoras.
Expect some jitter. The HC-SR04 sensors send eight 40KHz pulses then listen for an echo after which they stop listening. It is entirely possible that they miss the first echo and respond to the 2nd or 3rd pulse in which case TIME1 and TIME2 will vary slightly.
@@qualityphotos1 Thank you
@@qualityphotos1 Hi there, I am new to Arduino and have been attempting to create such an anemometer without any libraries or software other than IDE. I have followed all the steps you listed but my sensors appear to be unreactive to any wind. Placing objects in front of sensors will create a change in the values given though. I'm unsure where exactly I may be going wrong, do I need libraries for better functionality of HC-SR04?
@@NickyG0123 Hi Nick ... libraries aren't needed. Make an ultra sonic range finder such as this www.instructables.com/Ultrasonic-Range-Finder/.
Once you have it working add a second HC-SR04 as I have done here www.instructables.com/Dual-Sensor-Echo-Locator/ but don't blank the sensors. Define the extra sensor pins in the arduino setup(). Place object at different distances in front of each sensor and trigger them at the same time (or immediately after each other) ... you should be able to measure each distance by listening selecting from which echo-pin to make the measurement.
Now separate the sensors and point them at each other. If we trigger both sensors at the same time they should hear each other. Each echo-pin is now measuring the distance (time-of-flight) is opposite directions.
A better method is to remove the RX sensor from each HC-SR04 board and position it at the distant end. Now trigger each HC-SR04 in turn and measure the time-of-flight. Be sure to wait a small amount of time so any distant echos have had time to dissipate.
I found that jitter is a problem. If you look at the waveforms in this article www.instructables.com/Enhanced-Ultrasonic-Range-Finder/ you will see that the echos aren't clean ... but rather a cluster of pulses which means the time-of-flight can vary slighly if the first received pulse is missed.
Good luck with your project.
how u make this
Have discovered a couple of minor issues ... will publish the code and construction details when the design is finalised
@@qualityphotos1 thanks for this. I wait for the code