Grove GSR sensor - exact skin resistance calculation?

Hello. I am using a Grove GSR sensor (Grove - GSR Sensor | Seeed Studio Wiki) for my PhD project, and I need to extract the exact skin resistance data in Ohms. Unfortunately, I am having problems understanding the formula provided with the sensor.

Human Resistance = ((1024+2*Serial_Port_Reading)*10000)/(512-Serial_Port_Reading)
Could someone please explain how this formula works? Thank you.

im not exactly sure but 1024 is the maximum analog reading for an 8 bit ADC Analog to Digital Converter so the 512 is half that value so

1024 + (2*sensor)*10,000


512 - sensor

the maximum reading would be 1024 @: 1024+(2*1024)*10,000=20,481,024
divided by 512-1024 = -512

equals -40,000 ?

the maximum reading would be 0000 @: 1024+(2*0000)*10,000= 1024
divided by 512-0000 = 512

equals 2?

anyway that doesnt make any sence to me so i would say just experiment withknown resistnce
and see what you get
the main thing is you want to adjust the circuitry so that the maximum value you want to read will give a sensor value (Serial Port Reading) of 1024 and the lowest value will be 0

i hope that helped CWG

As far as I can understand:

  1. 1024+2×Serial_Port_Reading: This part of the formula is related to the raw reading from the sensor. The sensor output is a value between 0 and 1023, and this part scales and adjusts the reading.
  2. 512−Serial_Port_Reading512−Serial_Port_Reading: This part is used to create a ratio that represents the difference between a mid-range value (512) and the actual sensor reading.
  3. The entire expression is a ratio of the two values obtained in steps 1 and 2.
  4. Multiplying by 10000 is a scaling factor, bringing the result to a suitable range for skin resistance.
  5. The final result is the calculated skin resistance in Ohms.

Thank you fellas.
The product has a variable resistor that is supposed to be calibrated so the default output of the device is half of the bit ratio, in this case - 512. 10,000 does look like a scaling factor, given that human skin resistance is pretty low and should be amplified to get the result in Ohms.

I would just find some known values and plot them and see if it is a linear relationship or what and write the equation that way… then go back and test and see if it is tracking