ESP8266 Community Forum

- Wed Nov 14, 2018 8:51 am #79143

I'm the kind of person that follow the rules set by those that are knowledgeable; I'm not an anarchist.

For the power requirement, the "Rules" state:

4.1 What are the general power supply requirements of the ESP8266?

Digital voltage requirement: 1.8 V - 3.3 V
Analog voltage requirement: 3.0 V - 3.6 V (The lowest possible analog voltage is 2.7 V.)
Peak analog circuit current: 350 mA
Peak digital circuit current: 200 mA
Note: CHIP_EN works at 3.0 V - 3.6 V, please use a level converter to ensure compatibility
with digital logic at 1.8 V.

And, throughout the document, they are pretty strict about it.
Not far further in the FAQ, about powering the ESP using batteries:

4.4 Can I use a LiPo cell or 2xAA cells directly to power the ESP8266?

2xAA batteries can power up the ESP8266. Li-Po cells have a higher voltage swing while
discharging and therefore must not be used to power ESP8266 directly.
The ESP8266 contains precision RF circuitry which is affected by temperature or voltage
drifts.
Directly connecting any RF circuit to an unregulated power source is NOT
recommended.
Use of a DC-DC converter or LDO regulator is highly recommended to ensure that the
ESP8266 performs as specified.

So you might get away with powering the ESP with a Li-ion cell at 4.2V (which is only the starting level), though it is absolute discouraged, since at least three quarters of the time the voltage is above the maximum allowed.

Sure, I am not debating what's in the specs. I know what the specs says and what they "recommend". My post was more in the lines of; Has anyone tested over voltage and actually concluded that it damages the chip?

Because my own tests shows no damage with 5 ESPs and I have talked to several people online that has run their chips on 4.2v batteries for years without regulator or diode. Sure, just because it works for us, does not mean it will always work. But I have not heard of ANYONE that can actually say that they ran their esp8266-12e chips on 18650 and they broke after X amount of days. I just wanted to see if anyone can verify this.
After I read your post, I did some extra checks on the 5 ESPs for RF (wifi) signal strengths and they show same numbers as on my non-tested ESPs, thus no change or damage there either. At least so far.

Oh, and one more thing. Printed on the chip, it says that the wifi-signal is 25 dBm. That would mean that only the RF-part would use peak 316mW. According to the data sheet, the peak current is 200 or 360mW for the total chip? Impossible. Things like that makes me not trust data sheets from China 100%.

Last edited by buzzy on Wed Nov 14, 2018 8:55 am, edited 1 time in total.

- Wed Nov 14, 2018 8:54 am #79144 I've never had one brake by using a bit more than 3.3v. One thing to be aware of though is that the current drawn during deep sleep does increase dramatically with the supply voltage. An old post about that here

- Wed Nov 14, 2018 8:56 am #79145

I've never had one brake by using a bit more than 3.3v. One thing to be aware of though is that the current drawn during deep sleep does increase dramatically with the supply voltage. An old post about that here

A bit more? How much more? The chip can handle 3.6v according to the data sheet, so you might still be within specs. My post was specifically about using 18650 (4.2v peak) batteries without damaging the chip.

- Wed Nov 14, 2018 9:11 am #79146

A bit more? How much more?

Similar to you with 4.2v from a Lipo cell or occasionally connecting them to a bench power supply accidentally set to 5v.