Current spikes during chip boot

[jonsmirl]

When the ESP8266 boots we are seeing current spikes up to 300mA. We think this is when the chip calibrates the RF circuity.

Is there anyway to reduce those spikes so that boot current stays under the 30mA pulse output of coin cells?

The application here is to clone the Amazon Product Ordering button.
http://www.esp8266.com/viewtopic.php?f=6&t=2240

[ESP_Faye]

Hi，

Thanks for your interest in ESP8266 !

Could this API help you？

Code: Select all

Function:

Call this API before system_deep_sleep to set what the chip will do when deep-sleep wake up.
Prototype: 

bool system_deep_sleep_set_option(uint8 option)
Parameter: 

uint8 option :

deep_sleep_set_option(0): Radio calibration after deep-sleep wake up depends on init data byte 108. 

deep_sleep_set_option(1): Radio calibration is done after deep-sleep wake up; this increases the current consumption.

deep_sleep_set_option(2): No radio calibration after deep-sleep wake up; this reduces the current consumption.

deep_sleep_set_option(4): Disable RF after deep-sleep wake up, just like modem sleep; this has the least current consumption; the device is not able to transmit or receive data after wake up. 
Note:

Init data refers esp_init_data_default.bin.
Return:    

true  : succeed

false : fail


[jonsmirl]

The application works like this:

1) push the button
2) pushing button starts the power supply
3) ESP boots (this is where the problem is)
4) ESP sends message
5) ESP uses GPIO to turn itself off

This allows the chip to last for about 2000 button pushes spread out over up to five years on a coin cell.
Since the power is turned off the device can sit for long periods of time while not draining the battery.

So the deepsleep calibration API does not help since we have to boot each time the chip is powered.