-->
Page 1 of 2

some ESP8266 getting really HOT

PostPosted: Mon Jan 21, 2019 8:52 am
by 7pot
Hi,

some of my ESP8266 chips are getting really hot during (normal) use to the point of burning my fingers by touching. Powering via USB, I have seen currents of up to 0.7A whereas others just take 0.07A. Using 3.3V does not improve the situation. All are running the same LUA firmware and were freshly formatted.

LoLin NodeMcu V3 is getting quite hot (0.5A) but ESP12E DEVKIT V2 from www.doit.am is by far the hottest.

Am I doing something wrong? Is there really such a variety in power usage?

Regards
Christian

Re: some ESP8266 getting really HOT

PostPosted: Mon Jan 21, 2019 2:00 pm
by jankop
Am I doing something wrong?

Yes, no doubt.

Re: some ESP8266 getting really HOT

PostPosted: Tue Jan 22, 2019 3:56 am
by 7pot
jankop wrote:
Am I doing something wrong?

Yes, no doubt.


Thanks for you answer, jankop!

Actually I was hoping for something more specific like a tip (e.g. connect GPIO15 to Vcc). Do you have any idea about that?

I have just connected USB and flashed the standard 2.2 LUA firmware. No other pins e.g. are connected, nothing is running, wifi is not active. Two ESP8266 get really hot, one does not. That seems strange. If I am doing something wrong I have not idea what.

Re: some ESP8266 getting really HOT

PostPosted: Mon Feb 04, 2019 7:01 am
by davydnorris
7pot wrote:
jankop wrote:
Am I doing something wrong?

Yes, no doubt.


Thanks for you answer, jankop!

Actually I was hoping for something more specific like a tip (e.g. connect GPIO15 to Vcc). Do you have any idea about that?

I have just connected USB and flashed the standard 2.2 LUA firmware. No other pins e.g. are connected, nothing is running, wifi is not active. Two ESP8266 get really hot, one does not. That seems strange. If I am doing something wrong I have not idea what.


I have just recently been trying to find the source of a random current draw in my ESP circuit, and have found that GPIO15 seems (at least in the latest SDKs) to be left HIGH after boot. As this has to be pulled to GND at start up in order to boot from flash, this means it's draining to GND through whatever resistor you have pulled it down with.

I've now added a line to set it LOW and disable it on startup and it's reduced my current draw by about 12-15mA.

In addition, I also set up the wifi radio to be off by default - I turn it on only when I need it on by using the *_current functions so that the state doesn't get stored in flash. This saves a lot of power and keeps the chip cooler.

Hope those give you some ideas