According to a Web search:
“Lithium-ion batteries have a low idle discharge rate, typically losing 1.5-2% of their charge per month”
I have been playing with the EC source code, in order to force it into IDLE mode.
While plugged into the FW 180 psu, when the Cpu is mostly idle, it should hardly discharge at all.
I am seeing it discharge about 1% in a few hours.
I don’t understand why, but I expect it’s a hardware bug, but difficult to know for sure without the schematic.
I would be a shame if they are carrying the hardware bug into other future models.
It could explain the relatively short battery life mentioned in some reviews.
Can anyone think why this discharge is happening?
Note: I am not 100% sure I am measuring charge correctly so the above statement could be false. For example, I see charge % drop by 1%, but this could be due to the battery recalibrating itself, and not real discharge.
Laptops used to work that way in sleep mode. Microsoft decided to make a change to the specification some years back that, long story short, makes the CPU draw significantly more power in sleep mode. Many of today’s CPUs apparently don’t even implement the older, more power-efficient sleep design anymore.
If you want details, look for “S0 Modern Standby” vs “S3 Sleep.” Be warned, the topic is very confusing.
The top part of this thread might help explain parts of it.
I did not make it clear, my tests were done with the laptop switched on, just not doing much.
I.e. the 180W should have been enough for 100% of the power for the CPU.
But, it did take a small amount from the battery as well for some unknown reason.
I am capturing a lot of EC output now, so try and understand why its happening. Whether its a choice of the EC, or some other thing going on.
Ok, some evidence now:
Sat 19 Apr 15:28:57 BST 2025
Battery 0 info:
OEM name: NVT
Model number: FRANDBAT01
Chemistry : LION
Serial number: 0084
Design capacity: 5491 mAh
Last full charge: 5795 mAh
Design output voltage 15480 mV
Cycle count 79
Present voltage 15895 mV
Present current 0 mA ← Notice no charging or discharging happening
Remaining capacity 3474 mAh
Desired voltage 17800 mV
Desired current 5491 mA
Flags 0x0b AC_PRESENT BATT_PRESENT CHARGING
Sat 19 Apr 15:28:58 BST 2025
Battery 0 info:
OEM name: NVT
Model number: FRANDBAT01
Chemistry : LION
Serial number: 0084
Design capacity: 5491 mAh
Last full charge: 5805 mAh
Design output voltage 15480 mV
Cycle count 79
Present voltage 15892 mV
Present current 0 mA ← Notice no charging or discharging happening
Remaining capacity 3484 mAh
Desired voltage 17800 mV
Desired current 5491 mA
Flags 0x0b AC_PRESENT BATT_PRESENT CHARGING
Note:
Notice no charging or discharging happening but the “Remaining capacity” increased!!!
Notice also the “Last Full Charge” changed.
So, I think this is some sort of battery BMS re-calibration is happening.
The Linux UI also changed from 59% to 60%.
The EC said it went from 59.9% to 60.0%
3474 / 5795 * 100 = 59.9
3484 / 5805 * 100 = 60
Other jumps are from 60.0% to 59.7%
I suspect that this might have something to do with battery temperature, but I need to change the ectool and EC firmware to also report battery temp.
So, the EC should not base charge / forced discharge to this accuracy, there should be a hysteresis to ensure that these re-calibrations do not cause a charge / forced dischage cycle.