ESP8266 Community Forum

Hi, I'm making a "mometary ESP" device - somewhat similar to the Amazon Dash button and whilst all the code is working fine I am having a slight problem with the hardware - I've attached the schematic.



The circuit works perfectly well in that everything switches on with a press of the button, the first line of code pulls GPIO2 LOW to keep the circuit powered on and at the end it makes GPIO2 HIGH to power itself off.

The issue is with quiescent current. The HT7333A regulator is drawing 3.8uA when the circuit is off. I really can't see how this can happen as the device has no path to ground once the MOSFET is open. I appreciate that 3.8uA is particularly small but the idea is that this will last for some considerable time on a set of 3xAAA batteries.

Can anyone more experienced that me tell me how this HT7333A is drawing current? How is the circuit completed?

I'm not an EE, so I may be completely wrong here, but your FET has a "Zero Gate Voltage Drain Current" of up to 25uA, so the 3.8uA are probably just leaking through the FET.

You may get lower quiescent with a different FET, but at 3.8uA you are probably already significantly lower that the batteries self-discharge (a 1000mAh battery would last 30 years on that current!).

From https://www.microcontrollertips.com/tea ... ppliances/

Those who have analyzed Dash Button circuitry say it draws 200 to 300 mA when on, 2.3 μA when in sleep mode.

So 3.8uA is in the ballpark.

EDIT:

New Amazon Dash Button Teardown (JK29LP) Posted on July 12, 2016
https://mpetroff.net/2016/07/new-amazon ... wn-jk29lp/

While in sleep mode, the new Button uses ~2.0 μA, down from the original Button’s ~2.3 μA.

Hi both,

Many thanks for taking the time to reply.

I know that 4uA is of the order of the self-discharge rate of the batteries, but I wanted to try and reduce the draw as much as possible.

The BS170 has a much lower IDSS than the IRLZ44N (0.5uA MAX) and replacing the FET in my breadboard has reduced the quiescent current to lower than my uA meter can measure!

The downside is a slightly higher RDSon (5Ohms) and so more power consumed during in each button press but each button press is a very short duration compared to the idle time so I price worth paying I think.

Many thanks again - I now have a working circuit with <0.5uA standby current. Good to be at least 4 times better than the Amazon product!