Analog input voltage can be 3.3V

[SarenT]

According to this tutorial and my personal test with my Amica, 1.0V delivers 320 and 3.3V delivers 1024 at 10-bit resolution. I guess 3.3V is possible.

[devyte]

The voltage range that can be applied to the ESP8266 ADC is 0-1.0V. If you go above 1V you will fry the electronics.
Some boards incorporate a voltage divider* so that a range of 0-3.3V applied to the board ADC pin scales down to 0-1V on the chip ADC. Some boards don't have this divider and break out the chip ADC directly to the board pin.
So the voltage range allowed on the board ADC pin depends on the board.

*many users don't know about, or at least don't know the implications of, this internal voltage divider, and apply a voltage to the board pin from another external voltage divider. This interacts with the board divider and can produce apparently wrong readings from the ADC.

[SarenT]

Can we write it in a way that people need to be careful when trying to read ADC? I was looking around but this issue isn't addressed at all and I was confused that my readings at 1.0V are 320. As a safe method, one can start reading 1.0V to make sure that they don't fry with 3.3V. If the readings are 1/3 of 1024 -> then they can be sure about 3.3V.

[duncan-a]

Or people could have the sense to look at the schematic of the board they're using...

…

On Thu, 9 Aug 2018, 14:41 SarenT, ***@***.***> wrote: Can we write it in a way that people need to be careful when trying to read ADC? I was looking around but this issue isn't addressed at all and I was confused that my readings at 1.0V are 320. As a safe method, one can start reading 1.0V to make sure that they don't fry with 3.3V. If the readings are 1/3 of 1024 -> then they can be sure about 3.3V. — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#5020 (comment)>, or mute the thread <https://github.com./notifications/unsubscribe-auth/ALP8xQky-lcbIgOrEolpKuLmg0eDzAFEks5uPC3kgaJpZM4V1Bxj> .

[SarenT]

In my case, I didn't find the schematics and many new electronic hobbyists might need further explanation/warnings to avoid fatal mistakes.

[duncan-a]

Yes, I agree - RoboIndia don't seem to provide much information about anything. You could trace the track from the A0 pin but that really shouldn't be necessary. The problem with your suggested method is 'how many people have a 1V source?' I think that something about "1V on the actual ESP8266 - breakout boards MAY implement voltage dividers" should be sufficient. On 9 Aug 2018 16:06, "SarenT" <[email protected]> wrote: In my case, I didn't find the schematics and many new electronic hobbyists might need further explanation/warnings to avoid fatal mistakes. — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#5020 (comment)>, or mute the thread <https://github.com./notifications/unsubscribe-auth/ALP8xQ7LDlw69hjVmzFjheDuOff6ZWZcks5uPEHQgaJpZM4V1Bxj> .

[devyte]

@SarenT go ahead and rephrase as discussed.
The suggestion for unknown boards to start with 1V is actually not too bad. Users won't have a 1V power source, but they should have a pot. They can power it with 3.3V and measure the midpoint output with a voltmeter to make sure it's below 1V, then connect it to the ADC pin. If the ADC reading is of the order of 1000, then there's no divider on the board, and max voltage is 1V.
OTOH, if the reading is about 1/3 of 1024, then the divider is there and max voltage is 3.3V.

[jindrichsirucek]

There also should be mentioned, that many boards (it means basicaly all boards which are not bare esp) has voltage divedr in theirs design, so ESP.getVcc() will produce totaly wrong results - possible resistor connected to A0 pin has to be removed before using this function