ESP8266 Community Forum

Hi All,

I am trying to read the voltage from a 6v solar panel using an ESP-12. I have a voltage divider made up of a 10k and 2k resistor. When I did my testing my voltmeter was reading 5v from the solar panel and analogread(A0) is giving me 803, from that I calculate 803/1024 * 6 to me 4.705v. This is a 6 to 7% difference, which I find pretty big. What could be causing such a large variation?

Mike

Hi Mike,

What is your Measured Full scale voltage for your ADC input? This varies depending on the VCC value so might have some error from there.

Also I think I will choose the resistors a bit higher in values, like 54k/10k. This will give you a confortable max input voltage of about 0.938V for a 6V input.

For more details about ADC and voltage divider might worth to take a look at the Internal ADC - ESP8266 post. Take a look also at the comments, few good examples also there.

trackerj wrote:What is your Measured Full scale voltage for your ADC input?

Are you referring to the actual voltage on the ADC rather than the calculated voltage based in Vin * (R2/(R1+R2))? I have not measured it but I guess it probably is not exactly 1v. I will increase the resistor sizes and check the actual voltage on the ADC pin. I also read carefully the site you linked to. A quick passed showed there was a lot they were considering that I was not.

Thank you for your help.

The 1 volt reference is not made to supply an accurate reference for use with an A/D converter. The 1 volt is used for internal level and biasing functions. It is only being made available for the A/D converter because it already existed. Think of the reference as a byproduct of a different process.