Use this forum to chat about hardware specific topics for the ESP8266 (peripherals, memory, clocks, JTAG, programming)

User avatar
By Prodigity
#33476 Hello fellow IoTers,

I have noticed a discrepancy between the datasheet of the ESP8266EX (version 4.4) and my measurements.
According to the datasheet input leakage current should be 50nA yet I'm measuring 25uA.

That's a factor 500x difference and has a big effect on big resistor values;
50nA running through a 1MOhm creates a voltage drop of 50mV
25uA running through a 1MOhm creates a voltage drop of .. 25V (if it were an ideal current source)

My measuring setup:
ESP-WROOM-02 running nodemcu on pcb with GPIO14 and 3.3V line hooked up to a uCurrent.
uCurrent hooked up to cheap, non-calibrated(!) DMM
Code running on ESP - "gpio.mode(5, gpio.INPUT)"

I've also noticed a voltage drop of ~1.5V when connecting GPIO14 to 3v3 through a 500K resistor while measuring with a Hantek DSO5102B (also not calibrated unfortunately)


Can someone confirm this behaviour?
User avatar
By Prodigity
#33541 The input leakage is up to spec, I made a mistake with my setup.
My apologies. :roll:

<edit>

I still have the same problem and can now say with certainty something weird is happening.
I have a 470KOhm resistor hooked up to GPIO14 and +3.3V.
When I first connect the oscilloscope to GPIO14 and then configure the GPIO as input, the voltage stays at +3.3V.
However when disconnecting and reconnecting the oscilloscope probe the voltage drops to +1.2V and the gpio pin starts pulling 3.6uA.

If the internal push-pull configuration for the output setting is left floating while the GPIO is configured as input then rapidly pulling the GPIO to ground (through the capacitance of the probe for example) could explain this behaviour; the capacitance between the pin and the gates of the FET could be enough to make one of the gates conduct.. or I'm overthinking it ;)