« first day (1 day earlier)      last day (17 days later) » 

1:58 AM
I read the comments of the 2017 tutorial you are following. It appears that there are comments about your hx711 unreliability problem, you might like to read them to see the comment are helpful to solve you problem. I also found another gitHub python 3 only library you might try.
(1) Felix - Build a digital Raspberry Pi Scale (with Weight Sensor HX711) - 2017

(2) Gandaf15 - hx711 gandalf15 python 3 library - 2019jan02
2 hours later…
3:53 AM
I also found that the pyPi hx711 python3 library I am using is based on gandaf15. Also that your green module is different from mine, which has rate pin jumpers. Gandaf15 has a tip to set rate, as shown below.
3 hours later…
6:45 AM
Now I am going to calibrate the 5kg load cell. My test plan is the following.
(1) Test 1 hx711 module with 2 channels each of which handling 1 load cell.
(2) Test 2 hx711 modules, each Channel A handles one load cell.
Now I am searched for my old calibration record of the 5kg Load Cell.
(1) HX711 Forum Discussion Notes 1/4

(2) HX711 Forum Discussion Notes 1/4

(3) HX711 Forum Discussion Notes 1/4
1kg = 3.2mV

(4) HX711 Forum Discussion Notes 1/4 (calibration Notes)
A very rough summary of the 5V load cell calibration results is the following:
For Vcc = 5V:
(1) It is about 5mV for 5kg max load,
(2) Or 1mV/V sensitivity (I forgot what this exactly means! :)
My rough rult of thumb is the following:
Either Vcc = 3V3 or 5V0. The voltage output for max loading, either 500g or 5kg, is about 5mV. In other words, the HX711 load cell amplifier should be set to the following:
gain factor = 3V3 / 5mV = 3300 / 5 ~= 600. This 600 gain factor is what I am going to set in the coming Hx711 ADC test.
The calibration record for 5kg load cell:
2 hours later…
9:08 AM
Now I have connected the 500g load cell bridge output (wht, grn) wires to HX711, and ready to use the PiPy hx711 load cell amplifier library function to amplify a couple of mV to a couple of Volts.
Ah, locking down jogging and supper break time. Will come back late this evening or tomorrow.
One thing I forgot to mention is that: (1) What is an "airfoil"? (2) I googled and know it is just the "wing" of an aeroplane. (2) But how can the wing lift the plane? So I googled the following Youtube: How Does A Wing Actually Work? 2012aug03, 1,224,732 views
https://www.youtube.com/watch?v=aFO4PBolwFg. See you later.
3 hours later…
11:55 AM
airfoil is more like mimicking a wing or "flying object". Anyway the whole experimetn is rather a disaster and tomorrow I'll be taking everything apart, enough.
Too many mistakes and wrong choices were made. My tests are completely useless. I also have 5kg load cells and they barely register any force. So ideally they should have been more 1 or 2kg cells but worse the rod used is made of brass which way too soft meaning before any force is send to the load cell the brass rod will nullify it by bending plus the fan we use is waaay too weak.
I get an average wind speed of 4 m/s. Nonetheless many lessons learned from choosing right tools, hardware (next time arduino) to coding which I almost enjoyed most hahaha
3 hours later…
2:38 PM
(1) About load cell spec - Just now I used a 12V 0.7A CPU fan to blow my 500g load cell (5mV full scale), and found there was not any noticeable different in DMM reading. I at once gave up trying any bigger fans. I guess you need a 100g load cell to do your experiments. My 500g load cell is from this dirt cheap weight scale:
AliExpress Mini Digital Weight Scale 100g ~500g, 0.01g ~ 0.1g - US1 ~ US$4
(2) Falling back to Arduino - I very much agree you return to Arduino, because Rpi python has too steep a learning curve for you now. You remind me two years back, I over ambiguously thought I could learn ESP8266 and LUA quickly. But I did not know that LUA is event driven and perhaps easy for GUI design.
But very surprisingly it took me more than 100 hours to learn the IDE and another 10 hours just to blink a LED. For Arduino it took me less than 30 minutes to read the example and blink a LED. My conclusion was that I did not have the prerequisite knowledge to do NodeMCU LUA event driven programming.
So I cut loss and gave up and went back to Rpi. That is why I think your case is very similar to mine (choosing the wrong tool). Learning a new language seems easy, but there are many things you don't know that you don't know. Cheers.

« first day (1 day earlier)      last day (17 days later) »