I purchased an amplified digital load cell (TE FX293) and now I need to calibrate it. All the research I did I only found ones using the HX711 amplifier. I need help writing a code to read and measure weight without the amplifier just the load cell and the Arduino using an AC/DC adapter.
thanks
I'm not hopeful for your chances of success in that endeavor. The differential voltage produced by a load cell is a handful of mV. And, it can't be loaded down by too low an input impedance of the device you use to measure it. That's why people use an instrumentation amp (HX711 or similar) load cell applications.
ive uploaded a figure from the data sheet. I was told that the load cell is already amplified and would not require an HX711. Do you think its not possible to calibrate without it?
You could measure the output with no load and then with an object that's close to the maximum for the cell. Assuming a linear relationship, you would then use those two points to solve for the equation of the line (y = mx +b) between them.
You could also get fancier and make the measurement more than two points and do a linear regression to come up with the best-fit line equation.
I have a dumb question how about a link to technical information on the load cell? You might also consider the ones that went before you with the HX711 might have a good reason for using it.
here is the link im using FX293. I might have to change the whole load cell and get one that's compatible with the HX711 but I was just hoping I could calibrate the already amplified one.
Your load cell has an output of 0.5 to 4.5 VDC for whatever its specification is, the remainder of the part number? For example you have a span of 4.0 volts, your load cell is a 0 to 50 Lb version. So now 0.5 to 4.5 volts = 0.0 to 50 Lbs of force applied.
To check the calibration you would note the zero force output voltage (should be 0.5 VDC) and then apply 50 Lb of force to the button (measurement point of load cell) and you should see 4.5 VDC out.
Since the zero force out is 0.5 VDC and full scale span out is 4.5 VDC the span for 0 to 50 Lb is 4.0 VDC. So we get 4.0 / 50 = 0.080 or 80 mV per Lb of force applied. Actually a full calibration would involve applying for example 0, 10, 20, 30, 40 and 50 Lbs of force and measuring Vout at each point.
That or send it to a lab and let them chart it for you or just assume it is accurate out of the box. These are commonly called a button type compression load cell.
Writing the code depends on the Arduino used and knowing the ADC reference voltage. Just for example and a rough idea if the ADC reference is 5.000 VDC and you have a 10 bit ADC then 0 to 5.0 Volts becomes 0 to 1023 bits. This is where you map the inputs to your units of measure. So you get 5.0 / 1024 = 0.00488 or 4.8 mV / bit. We start with 0.5 volt in the example so 0.5 / .00488 = 102.45 bits or just call it 102 bits. The full scale of 4.5 volts becomes 4.5 / 0.00488 = 922 bits. So map it and 102 to 922 becomes 0 to 50 Lbs.
There are 3 models. One that puts out mV that you must add an amp to, one that puts out a high level signal that you wire to an Analog Input, and a 3rd that has SDA/SCL interface. #1 needs additional circuitry #2 you can read with an AI directly #3 you need an #include (library)
Thread starter mentioned they had the amplified output. You would only need the library if you were using the I2C interface (SDA Serial Data & SCL Serial Clock). This is the data sheet for that line of sensors. TE FX293 is not a complete part number, there is a suffix. The complete part number defines exactly which version and range. Thread starter has provided an incomplete part number? Since they mention amplified version I replied based on that.
There are THREE models, one with naked strain gauge that needs a separate external amp, one amplified to .5-4.5V that can feed an analog in directly, and one with I2C that needs a library.
If you have one with its own amplifier, you can read it with the analog inputs direct. Note that product line has 3 output options, millivolt, amplified and digital.
Calibration is conceptually identical with any version, take average reading of the tare point, take average reading with a test mass, then you can subtract out the tare and scale appropriately.