Can the output current of the esp32 GPIO be reduced?

[samsam]

The 12mA *is* the maximum amount of current you can safely pull from the ESP32 [2]. A real chip actually can deliver much more: .....

If this is true (and I give you the credit as it seems reasonable) then the spec sheet need to be fixed ASAP as it is quite misleading with this 80mA in the last edition. Unfortunately I cant search it to see if the 12mA still stay somewhere in the text
I was prepared to ask those 80mA are they per pin or per domain, but fortunately found this posting, so it gives some light on the situation.
Anyway I'd suggest your major points from your post have to find place in the official ESP32 datasheet (or at least in the manual), and also I think have to be set there in the data sheet what is the maximum current per domain or for the all GPIO's together, and if there is difference in sink and sourcing current....Thanks

[ESP_Sprite]

Yep. If I didn't find this out at the start of the weekend,the 80mA value would already be gone by now.

[Willie]

I use gpio_pad_set_drv to change pin drive current has the change, but the change of the current volume is very small...

[ESP_Sprite]

What is the setup you're using to measure that?

[rudi ;-)]

hi

back to the question
since 2017 june 9 the datasheet of ESP32 speaks from ~5 mA ( before ~10 mA )
the ~80 mA is remove now.

( if we search - we find not the search text - if we copy a text from it - it does not have the letters from the seeing text )
example text in the pdf "~40 mA" is then "~:6 wH" )

2017_june_9_datasheet_esp32.jpg (36.05 KiB) Viewed 13141 times

best wishes
rudi

[anakyr]

Hi,

I read the previous posts but I think they still are not quite clear.
So the version of ESP32 now is using the 5,10,20,40 mA levels?
Apart from that, this current is the total current the CPU supplies to all the pins? Meaning that in each pin the current is much lower, if more than one pins are used at the same time?

[ESP_Sprite]

No, it's 40mA per pin. Worst case scenario, you can indeed have a situation where you pull 40mA from all pins - while we haven't done any exhaustive testing, a quick test on a few room temperature ESP32s does indicate it keeps on working if you do this. (See the datasheet, section 5.1) And yes, this means the chip uses 4 watts or so in that scenario - I was pretty surprised it happily survived that as well.

[Neuer_User]

Reviving this older thread, I still do not fully understand what is now the maximum current that can be "continuously" drawn from one GPIO?

In my application, the ESP32 needs to drive an external contactor, which draws 29mA from the pin. It works, but I don't know if that can safely be done for 1-2 hours, or if I risk damaging the esp32 and better use a relay (there is only very little space, unfortunately).

Thanks in advance for clarification.

Michael

[username]

In my application, the ESP32 needs to drive an external contactor, which draws 29mA from the pin. It works, but I don't know if that can safely be done for 1-2 hours, or if I risk damaging the esp32 and better use a relay (there is only very little space, unfortunately).

The typical method is to use a transistor or preferably a fet to drive anything with a inductive kick. A SOT-23-3 fet would easily handle the currents your are needing.

[ESP_Sprite]

It should work fine, the ESP32s IO pins are pretty beefy, but if you don't feel comfortable I'd put a mosfet in between, as the post above me suggests.