The peripherals fail on Framework Laptop 16, Windows 11

I dual-boot my Framework Laptop 16 (AMD Ryzen 7040 Series) to Windows 11 and Fedora 43, and I had an issue with my Framework Laptop 16 with Windows 11 (25H2 version 26220.7344). I am participating in the Windows Insider program, and receiving updates from the Beta channel.

About two weeks ago, I have updated the BIOS to version 04.02.2025.11.13 under Windows, but had to boot in to Linux to update the firmware of the keyboard and numpad. I also updated the drivers while I was at it.

While I’m using my laptop with Windows, half of the peripherals gets disabled on occasions. This affects all my expansion cards, Bluetooth devices, keyboard, numpad and touchpad. This in itself would just be mildly infuriating, if not for the SD card I am using, and the Ethernet card not resuming. I don’t fully understand the extent of the conditions this issue occurring, but startup and resume both seem to trigger it.

I use the Blank ISO (1st gen) keyboard and numpad (1st gen).

The laptop is an AMD Ryzen 7040 Series, and the expansion bay has the AMD Radeon RX 7700S (1st gen). An ASUS Strix XG27WCS display is connected to it’s USB-C (HDR: off).

The following expansion card configuration, described in the order you can find here:

  1. USB-C: Power (180W)
  2. Ethernet
  3. USB-C: Logitech Brio 500
  4. Display Port: ASUS Strix XG27ACS (HDR: off)
  5. MicroSD Card reader
  6. USB-A: Corsair Cooler Master laptop cooler: Logitech Bolt

(Not sure if 3rd party devices are relevant, I have only included them for reference.)

I set the power consumption to performance from power saving when I start working for my firm with it, and restore it to power saving in the Settings (not the quick action power saving mode) so I can get all the performance out of the laptop.

Today, while I was troubleshooting, I’ve noticed that it also stopped charging after re-plugging it, and did not charge even when I plugged it in to the 2nd socket. So I completely powered it off, removed all connection, and after a back-and-forth between 1st and 2nd connection, it started charging again. After this, I decided to unplug all expansion cards, and reconnect it.

Some troubleshooting later, it seems, that excluding the corsair laptop cooler and replacing the 5th expansion card to a USB-A: Kingston MobileLite Plus UHS-II MicroSD card reader resolves the issue. (For now?)

I prefer Linux to Windows, and only keep Windows around when an application doesn’t play nice with (or have no port for) Linux. When I teboot to Linux with this configuration, I face no issues there.

Does anybody have any similar issue with a similar setup? Am I putting the laptop under too much stress?

1 Like

Yep, that is probably drawing too much current from that port. Goodness knows what the power rating for the cooler is, but that is the most likely culprit I see in your list.

1 Like

I see. I’ll avoid it for the time being, and check if that still causes any issue.

I also messed up, it isn’t a corsair cooler, it is a cooler master, and probably the Notepal U3 Plus though not sure, as I don’t have my receipt for it. I have updated my post.

Although the FW16 has 6 slots for usb devices, it is limited in the power it can provide. If you connect more than one device that needs 15W, it can be unpredictable.
See text describing it here:

1 Like

Windows is a malware at this point, so it would be much easier to find alternatives for Linux than trying to work with it. Unless you need high end video editing or 3d software that doesn’t really have a viable alternative. But on the topic. I have had some instances where I needed to fully power off the laptop or even reset the cmos to fix USB port issues. Its likely you have combination of bad driver and weird resume / sleep power state behavior of the laptop that causes your devices not to initialize. I am not sure how you can debug this on windows.

I was looking for Watt limits on the Framework Laptop, but didn’t find any. The link that you have shared (I think I also shared it in my original post) does not detail this 15W. Currently, your statement is the only information I have in output/input regarding Framework’s product (except their power-brick).

I’m afraid that is what happened last time. Luckily a shut-down and leave it off seems to resolved it, but also happened after losing the devices several times, so if I avoid the root cause I should be OK, I guess?

That would be good. I assume you can’t debug it from Linux, if the error occurred under Windows. I could have shared the Windows Event Logs for debugging, if I had thought about saving it. Would anyone be able to do this debugging?

For the time being, I’ll try to avoid this by skipping the laptop cooler. Will post if the issue resurface.

The link had this text:
“Ports 1,2,4,5 are shared on a first come first serve basis. So they are 1.5A by default, but one port from any of these can negotiate 3A over PD. Once that is taken, the other ports are limited to 1.5A. Ports 3,6 are 900mA.”

These are all at 5V DC. Volts x Amps = Watts.
So the watt limits are:
3A == 15W
1.5A == 7.5W
900mA == 4.5W

1 Like

Windows event logs are full of cryptic codes and not really targeting the owner of the OS, but mostly Microsoft and their support. You can try a technical AI like Gemini Pro to analyze it for you, but you have to be careful in the prompt, instructing it what sources to use, how to analyze and to support its claims by quoting the sources. Trouble is that public information on MS logs is very limited so neither the AI or searching yourself are going to come up with a lot of useful documentation.

Oh, I did not realize that! Thank you.

It happened again, twice today. The laptop cooler wasn’t connected this time, and the storage device also was disconnected when this occured. I’ve installed and run HWMonitor, and wanted to find out what used the most wattage, and found this after the second drop:


Is this normal for max? I actually blame Citrix Workspace Desktop. Also, I would like to upload the HWMonitor report, but can’t upload text files.

Update: I have installed USBLogViewer too, this way I was able to capture the exact time when the devices got disconnected. I crosschecked with the event viewer, and saw nothing signaling the issue, only the fallout of disconnecting the devices. Some of them are the USB hub (within the Framework Laptop), the fingerprint reader, perhaps all HID devices (not sure which) and the Bluetooth adapter.

Process Explorer was running too when the issue occurred, and noticed a spike in CPU usage around the time, and while I was not how it correlated (during, before or after) noticed Steam and Everything search in the processes. I have realized that too many.

I have no longer any idea what causing this issue at all. Under Linux, this setup works without issue. Under Windows, this setup causes disconnect the integrated USB hubs. :person_shrugging: I have disconnected one of the monitors, to mitigate possible load, though not certain this could be the issue.

Nope, it disconnects the device in this setup too. Now I just removed everything, using the built-in screen, web camera, speakers/mic, keyboard, touchpad. I (don’t) want to see disconnecting now. I don’t know what it would mean if it disconnects too. I guess I reinstall the drivers?

I started getting the same thing after updating to the new BIOS and doing the firmware updates that I was told to do before updating to the new BIOS. Now and then all my USB devices, my ANSI keyboard, and my numpad keyboard just disconnect and then reconnect after a couple of seconds. So yeah, I bump this.

This problem is arguably more annoying than when I’m playing games like Battlefield 6 and my CPU is not even 50% utilized, yet I’m stuck at like 58 frames and stuttering. This always happens after like 30 minutes of gaming normally on 100+ fps at max resolution at 165 Hz and then it gets stuck in this like second mode of stutter land. I’ve tried lowering my game resolution to the minimum and it doesn’t change my FPS at all. I’ve just come to assume that AMD graphics suck for most games and others seem to not have issues. So when I can, I’m going to upgrade to the new NVIDEA one and see if that brings me peace.

Update: It happened right at login. I decided to report this to Microsoft as a driver issue. I did that twice, first thinking it was due to Citrix Workspace but I created the second report as a clarification for driver issue (Windows wasn’t a category; the report is in Hungarian).

While I’m also suspecting graphics driver issue, I completely ruled out hardware issues, as this configuration works on Fedora 43 without issues (hence I made the report to Microsoft).

I recommend USBLogView from NirSoft to monitor USB connection; that is what I use right now to see what is happening. Currently, I am running it in the background as I work. The device ID’s are criptic, but I see USB hubs with the vendor name Genesys Logic, Inc. which I assume is the integrated USB hubs (as I have none connected externally).

I have not considered BIOS, or AMD to be the culprit. I was also considering to upgrade the laptop, though I wasn’t looking to upgrade it after a year.

I figured out my issue at least. My Ethernet card is no longer working at all after updating to the most recent BIOS and doing the firmware thing that framework had on their drivers page and for some reason would cause my keyboard and all other USB devices to disconnect. Now I’m not using the Ethernet card and I don’t have the issue anymore. Thanks Framework. :folded_hands:

I wish it would be as simple for me. I had issues upgrading the keyboard/numpad firmware through Windows, but was successful through Linux. Later, I’ll check my Ethernet adapter’s firmware, but the issue was present even when I didn’t have that plugged in, so I don’t thing that will work.