Parasitic Intermittent Battery drain - I went down the rabbit hole, and there is a bug in firmware.

[lighttech3490]

I have a HP laptop with "intermittent" battery slow drain issue when powered off when no A/C supply is plugged in.  This slow drain when it occurs, is "unmetered" and does not show a reduction of state of charge.  I have verified this using the UFEI Diagnostics, and looked at each cell's voltage.  When this problem occurs (I will show later what triggers it), a fully charged NEW battery will completely drain to dead in about 3 to 4 weeks if the laptop is not left plugged in.  I have been through several batteries, and even installed a new mainboard.  Before anyone starts telling me the obvious things to check, I have confirmed the following:

1..  The problem can be induced EVERY TIME if you shut down the laptop (not sleep or hibernate) and then close the lid as soon as it turns off.  If you wait exactly 8 seconds or more before closing the lid, the slow battery drain does not occur.

2.  I have ruled out the OS by removing the NVME drive and powered the laptop on, then turning it off, and if the lid is closed before 8 seconds, the problem occurs.

3.  I have used a scope and meter to see what exactly is going on.  If you leave the lid open, shut everything down, and then unplug the AC adaptor, all switching power supplies on the main board turn off except of a LDO regulater to keep the BIOS settings alive.  That is expected.

4.  If you immediately close the lid after the system turns off (no LEDS on) one of the 5V switching regulater stays on, along with another regulater down to 3.3V which can be read at the VCC pin of the BIOS serial EEPROM.  It never turns off.  However, if you re-open the screen, it will shut off in 8 seconds.

I figured that maybe its a firmware/BIOS issue, so I downloaded and installed the latest version from the support website.  That did not change anything.  My next move was to change out the main board.  That did not fix it either. 

The only way that I found to completely eliminate this issue is to remove the magnet behind the screen bezel, so the lid/screen being closed is never registered.

Please forward this to your software engineers, and I am more than happy to give further details.

Thank You,

Craig Sherman

(I am an IT professional with an extensive background in electronics and design.)