Took delivery of my batch 3 DIY 7840 this week and have spent the last few days getting things installed and set up.
But I wanted to see if anyone else is seeing this behavior- when using Windows 11 Pro and putting the laptop to sleep (either via power button, software sleep, or lid close), I’m seeing the laptop continue to draw 4-5w continuously from my 65w usb-c PD power supply. I have a usb-c cable with builtin display to show the negotiated PD voltage/amps and instantaneous watts drawn, and this matches another inline power measuring board that I can also use.
This seems high and makes me wonder if I have an issue getting the laptop to sleep. While in this mode, the power button light pulses.
Other troubleshooting details:
I’m using Gskill 5600 40-40-40 32gb x 2 ram and it passed 4 passes of memtest86 with no errors.
SK hynix p41 platinum 2tb SSD.
USB-c x2 in the back slots and USB-A in the front slots. Only USB-C charger plugged in to the back left port.
Fresh retail Windows 11 ISO install, followed by framework drivers bundle and all current windows updates (early adopter button not checked).
-Bios 3.02 arrived with the laptop and I haven’t upgraded to 3.03 yet after reading some issues people had with PD chargers and lack of downgrade options if 3.03 introduces another issue. Figured I would wait a week or so but if I’m the only one seeing this type of power usage after closing the lid for sleep, then I can try the bios upgrade.
in the bios I had set it to only charge to 80% whole plugged in and it seems to be doing that just fine.
Also seeing the issues where unplugging or plugging in the charger or closing the lid in sleep mode wakes up the laptop, as discussed in another thread.
Anyone else? Any other ideas I should try? I have a n100 home server that sips 6w 24/7 and prefer not to be wasting that much with a sleeping laptop. But I also prefer not to have to shut down and re-open all of my apps and windows after startup to protect from waking to a dead battery…
Everything else so far with my first Framework laptop has been overall positive, so I hope we can figure this out.
I am running windows 10 but I saw something similar with 2.5W+ power draw during sleep (up to 9W during one sleep period) on battery. I was able to get it to reduce to 1-1.5W or so by installing the drivers from AMD’s website directly I believe. Most of the major offenders list from powercfg /sleepstudy seemed to stopshowing up after that. I was already running 3.03 though so not sure if that would make a difference (but updating to 3.03 did not itself seem to improve the situation as I recall).
Thanks for the tip- are you referring to the newest AMD video drivers, or are there other updated AMD drivers?
Looking at powercfg /sleepstudy, it only showed an abnormal shutdown when I checked on the computer this morning and it had run out of battery overnight while sleeping and plugged in…
The battery settings showed I had a constant drain overnight from November 3rd into the 4th (perhaps the computer woke back up even though the power was unplugged before I closed the lid and put it to sleep?)
While plugged in and asleep earlier today I saw some 1-2w readings, but I keep coming back to a laptop that has woken up by itself (i.e. put to sleep with the windows sleep menu item, and the computer enters the pulsing power button light state, and then at some time in the future, suddenly the screen is on again or the screen is black but the power button led is constant lit, and at that point touching the touchpad turns on the screen to the login screen).
Trying a few other powercfg commands (I may not be understanding what these do- I thought /lastwake was supposed to tell me what woke the computer up last time):
C:\Windows\System32>powercfg /lastwake
Wake History Count - 0
I haven’t been able to click through every task in Task Scheduler to see if any are set to wake up the computer.
But so far, very frustrating to not have a laptop that just goes to sleep when you ask it and doesn’t wake up until you choose to wake it.
I haven’t tried removing the expansion ports yet (I’m just using USB-C x2 in the back and USB-A x2 in the front, and have nothing plugged in except for the charging cable in the back left).
I did a windows software sleep last night (leaving the lid open as is my usual custom for a laptop that’s still sitting on a desk and will be used again the next day), and it seems to have gone into hibernate fine, saw power draw on the USB cable go to 0w and then unplugged the cable (and the computer did not start back up). I left it unplugged overnight, and this morning plugged it back in and then when I hit the power button, it resumed from hibernate mode as expected. The battery level was 79% compared to 80% when it went to sleep, but 2 minutes later it reads 80%, so there was negligible drain, if any.
If it keeps hibernating and resuming without issues, I’ll be satisfied. I don’t mind waiting less than a minute for it to write ram to disk and hibernate, and it woke back up in about 15 seconds. It’s an overall better experience than not knowing if your laptop will wake up early from sleep in a bag and burn through the whole battery and also lose all open progress/apps.
On AMD’s webpage for the 7840u processor there is a section for drivers, with software to auto detect and install (just drivers is the option I used).
Perhaps you’ve already tried all of these, but here are some other possibilities based on comments I’ve read, how much any of these helped I’m unclear on :
Change to network disconnect sleep (you can view the current setting for S0 by using powercfg -a) (I believe this may have taken a registry edit)
Disabling indexing
Disabling memory diagnostics
Using O&O shutup to disable various windows functions
Removing realtek audio drivers (I’m not sure if the AMD driver bundle had these or was just on the intel one, but there were problems with these in the past, amd drivers should work I’d think?)
Disable fast startup in the power menu
While hibernate does work as I understand it, when it’s set to hibernate on lid close, if you close your laptop often during the day that could potentially cause significantly more wear to your SSD over time. I personally set both a time and power limit for hibernate (it will hibernate if it loses x% power or after y time). So it should hopefully not be writing to the SSD too often each day.
