alt Hacker NewsFixing a bricked Framework laptop
Comments
I had the same thing happen to me with my Thinkpad Nano, and researching how to fix this made me realize pc manufacturer really don't care. PC bricked from BIOS update are still ultra common.
I uploaded a dump of the flash [1] in the bad state and they somehow analyzed it and made a fix. Great for others I guess, but I'm still mindblown that the regular path would have been that this perfectly working laptop to this day become ewaste.
Being an embedded dev, I don't understand why there isn't a special USB-C protocol to recover bios over SDU/CC pins. Or some DFU mode you can boot into.
> Framework chose to not provide a header for flashing the BIOS, which, when combined with issue #3, forced the use of pogo pins.
Op probably missed https://github.com/FrameworkComputer/FrameworkDebugger#jspi. Though, the connector is not populated for cost reasons.
Otherwise, I agree with the author, this is a pain.
A bit off-topic, but kind of relevant. I firmly believe that if installing custom firmware voids the warranty of a device, installing official updates should extend the warranty.
I'm tired of sloppy updates or change of behavior that mess with how I use MY device.
At this point, I'm convinced Google is doing this on purpose. My Google TV has become slow since the Gemini update, which I wasn't asked if I wanted it. My pixel battery life gets worse with every update, etc.
I agree with their conclusion.
I somewhat regret buying my framework laptop at this stage, because the answer to every fault is "buy a new component, from framework".
I'm not sure I blame them for this really, but there is no competitive market of parts (or really any market at all), which means that the reality is that you're just locking your money to them. They also often have stock issues, so woe betide you if your component fails at the wrong time.
It's very hard to predict costs here, but framework laptops do seem to fail more than other brands, and their laptops do tend to be more expensive. In comparison to MacBooks I've used over the years, which other than screen delaminating, have been faultless, at least.
The non-zero chance of a BIOS update bricking my computer is why I NEVER do them to a machine that's out of warranty, unless I'm experiencing some problem that a BIOS update is _well documented_ to actually fix.
I really don't understand how in 2025 a competent company can't implement a trival A/B flash update. Do they depend on some obscure contractor for the firmware and have no power here?
It's interesting to read this. Because I recently had something similar happened to me and I had mixed feelings like the author here.
I have the original Framework 13, but I have almost changed everything inside of it (motherboard AMD 7840U, wifi 7, ram 96GB, ssd, better hinge, the CNC shell, hdmi dongle v3, APU Fan, speakers v2, webcam v2). I still was using the original battery and it was at 60% health. So, I decided to update to the 61W battery.
After changing the battery, I decided to turn on the battery charge limit to keep the battery at a cool 60% charge to increase its lifespan.
First interesting thing I faced was realizing that Framework has 2 methods for the battery limit implementation. You have to use framework-tools and set it on the ec instead of the standard Kernel module. The kernel allows you to set the min and max, but the ec just allows you to set the max, and it will automatically set max-5 as the min.
But what was more frustrating, was realizing and when it gets to that threshold it constantly discharges 5-9W until it gets to the min and then charges the battery again back to the threshold. So, if you set a threshold it keeps charging that 5% once an hour, instead of parking the battery and waiting for the natural discharge to get to the min (which should be days later, since when you're plugged in the battery is not used for anything, it's like your laptop is off, not sleep, off).
But the good thing is the ec code is open source, I found the code, dug into it, found a bug, fixed it, flashed it, and for the past 2 weeks my battery is working perfectly.
I opened a PR, then found a multi hundred forum post of people talking about something similar with no proper solution, or any fixes in the ec. And my PR has been sitting there with no attention from Framework. Setting a battery threshold is actually worse for your battery because of this bug, but nobody is in a rush to fix it!
So, I'm mad that Framework operates this way, their circuitry, power delivery, and software is flawed (just read the forum thread[1]). On the other hand I'm happy I can see the code, find the bug, and fix it at least for myself! Feels like the hardware version of Linux! If you are technical, you can literally change how your kernel behaves. You can't recommend it to your Mom, but you definitely prefer it to the closed alternatives.
[1] https://community.frame.work/t/tracking-battery-flipping-bet...
The lack of support from Framework is so damning. Nirav has been a fun CEO to watch, but this feels like a huge, huge miss. Can anyone defend this? Why would anyone purchase anything from this company ever again?
I also managed to brick a motherboard, though mine was a Supermicro, by missing a line in the README that tells you to move some hardware jumpers on the board before applying the flash. I found one jumper, but the other one was in another corner of the board, and the README says "if you have both, move both", so I thought I just had a board with only one jumper.
Notably, the flash process itself does not check for the positions of the jumpers and exit, or anything like that, it just proceeds merrily along, and renders the board unbootable. I've been around with Flashrom a dozen times and can't figure out the right sequence of flash-this-rerun-that to recover the thing. (It's an X9DR3, if anyone here has hints.)
So I wish I could be mad at Framework, but only the tiniest bit. It's sadly common.
Good story, and surprising that there was so little assistance from the OEM. As an aside, I got to test out Asus crash free BIOS a couple of weeks ago. A strong thunderstorm hit and before I could get to my office a rapid series of deep voltage drops occurred. I shut down my systems and waited out the storm. When I powered my Asus tower back on it failed to POST, automatically restarted, recovered the BIOS (CMOS settings I assume), restarted again and booted Windows. Pretty slick, and instantly removed that cold lump of lead that had formed in my gut.
I really like what Framework is trying to do but this story is very hard to ignore. If a failed BIOS update can brick the machine, and the official solution is a CA$500+ motherboard (that's about $350+ USD I belieive), there's still a pretty big gap between being modular and being truly repairable
Not to mention there are no "alternatives" that you can look into to purchase these replacements. You are essentially just locked to purchasing from Framework only
The entire USB-C situation with the Frameworks (I have both the original gen 11 and the AMD 7040) is a massive train wreck. They have clearly given up on fixing it. Only some of my ports work, and not reliably. Th EC controller is a bug fest. If you don't believe me, check their forums. I wish I could get back all the hours I spent debugging this, to no avail.
Framework seems pretty popular here in general, but nobody really seems to talk about how repairable enterprise laptops are. Some laptops have clips around the edge of the bottom panel to hold it in place, which frequently break. On my brother's T480s, those clips were made out of metal and held in with screws, meaning they were themselves replaceable. More recently I worked on a Dell Precision 7550 laptop, and if you crack it open everything is labeled just like a Framework. Also, before selling it I ran the onboard diagnostic. I was surprised when it told me to check the touchpad cable, because the touchpad was working fine. Took it apart again and yep, the touchpad cable was loose.
Framework really caught everyone's attention with the fun DIY repair angle, but Dell/Lenovo/HP have been selling repairable laptops to IT departments for decades. In addition, they're a much more mature product -- been selling laptops for much longer at much higher volumes. Thinkpads are the prototypical example but I actually like Dells better. Many of them are Ubuntu-certified as well.
My support experience has been quite similar. Lots of required evidence, slow responses, and then just links to buy replacement parts in the store.
The kicker is that the things they were telling me to buy didn't even fix the problem. My fingerprint reader was very flakey, and after replacing it and all cabling connected to it, it still didn't work.
They then told me to buy a whole new input cover which was the wrong solution again. The issue was the touchpad, which routes the signal for the fingerprint sensor through its circuit board.
They also encouraged me heavily to update the BIOS and firmware of everything aggressively before really digging into any of the kernel messages I was seeing.
All that said, it's really hard to complain that for $50 and some waiting I was easily able to fix the problem.
Compared to replacing the battery on a MacBook pro which took me 5hrs and required complete disassembly of the chassis sans fans it was a dream.
Here's to hoping that framework does better, they do seem to take well to community feedback.
Was seriously considering getting a Framework to replace my aging ASUS. Upgrade- and repair-ability were my main motivations.
But now I'm reconsidering. Not standing behind their own BIOS update is pretty horrible.
Tip: if you already have Raspberry Pi, your can skip purchasing CH347.
RPi has SPI pins and is supported by flashrom.
> For three years, the laptop has served me well, until Framework suggested via newsletter to install the latest BIOS update, version 3.20, with a bunch of security fixes. Unfortunately, the system hung and displayed a corrupt image on the screen, signifying a failed BIOS flash.
> Naturally, I reached out to Framework support, who told me to unplug the laptop, let the battery drain, and power it back on again afterwards, hoping the laptop would recover by itself. Unfortunately, it never did, and after giving Framework a bunch of information, they informed me that since my 1-year warranty has expired, I have no option but to purchase a new Framework motherboard, which will cost at least CA$500.
This is very disappointing.
1) Framework failed in not implementing a bootloader which takes a BIOS file on a USB-Stick and flashes it w/o requiring a working BIOS/Firmware or even a CPU/RAM. Every $70 Desktop-Mainboard worth its salt has this feature.
2) Considering 1) and that it was an official BIOS that caused the issue, Framework should repair (replacing/re-flashing the BIOS chip) for shipping fees max.
I'm on my third mainboard trying to combat what seems to be an AMD platform issue that causes random freeze-then-reset events. It doesn't just affect Framework. If you search AMD Data Fabric Sync Flood you'll find a lot of discussion across various forums. The open Framework issue is: https://github.com/FrameworkComputer/SoftwareFirmwareIssueTr...
I recommend people who have a soldering iron to start experimenting with enameled wire. It's easily sourced from any random coil/inductor/switching transformer/broken speaker even. Dip the tip of the wire in some flux and keep it on the tip of the iron for a bit so enamel clears away and tins the wire.
This way you can solder the thin wires in crammed places, don't have to stress about them shorting out anything they touch, easily routable and they don't put stress on the solder joint, like a regular thicker wire would.
Working with this type of wire made things easier for me in these sorts of situations. And it's perfectly adequate for signal (withing reason, if not very high speed), and even some power, few mA needed for these kinds of chips is no issue.
My framework was sitting in a drawer for two years and now will not turn on no matter how long I charge it. I assume there’s a dead coin cell in there somewhere or something that needs changed but it’s pretty disappointing.
I had considered buying a Framework laptop.
This systemic lack of support for the product for two problems of their own making, both the original design flaw of inadequate recovery from a broken BIOS and producing a bad BIOS for some models and actually encouraging its installation, is more than alarming.
My primary concern about Framework was if they would survive so they were capable of following up on their goal of modular upgrades. That was a risk that I was willing to take.
However, that a company that targets a market segment of fairly sophisticated technologists would fail this badly on _both_ engineering and customer support is a deal killer.
Framework is not even on the shopping radar now for me.
I can see the potential, but for the average consumer, they would prefer an integrated solution, but because Framework is using a "Focus market" as their business strategy similar to Instacart and grocer, they have a short to mid term window to earn outsized profits and build something more defensible.
Anyone else bothered by the author’s constant return to whining about how Framework did not do enough for this specific repair? Each time I found myself thinking “ok, rant over, now on to the interesting technical bits” only for the rant to resume a few paragraphs later.
I know that firmware (and software in general) has bugs. But I wonder why modern firmware for UEFI has to be so complex, that it needs updates on a regular basis. It would be nice to have a dead-simple hardware-initialisation system that doesn't need regular updates and that only has very limited attack surface.
It's a shame Framework don't offer support for component-level repairs, but sadly neither do other laptop manufacturers, if you don't count leaked resources. They could certainly do better, but I wouldn't see it as a reason to switch to e.g. ThinkPads.
$20 of tools, and a lifetime of skills. In this case, cheap has quite a few prerequisites.
If the motherboard has so many parts that could fail independently, and each part is small money, then maybe the motherboard should also become modular?
I have the same AMD 7040 Framework laptop and just had the keyboard fault explained in this thread where the "-", "0", and "P" keys stop working: https://community.frame.work/t/tracking-framework13-keyboard...
While it's nice to know that I can buy a $100 replacement keyboard (I'm about 6 months out of warranty) it's also frustrating that Framework wouldn't own up to what is obviously a manufacturing failure.
I bought a Framework 13 in 2021. I was so stoked. I used it for 5 years, and of those 5 years only 1 was heavy use. I had to replace the battery and the input cover. I never had to do that before. I never had a laptop I wasn't able to use for 5 years so the repairability was actually not that much better for me. I gave up on Framework because after 5 years they still didn't give a crap about terrible battery performance on Linux distros.
I was using a Macbook Pro before. Went back to a Macbook Pro again. I don't want a Mac. I just can't find anything else remotely worth my money.
Don't bother with the motherboard, the new BIOS is worse. I have two that just randomly shut off whenever they feel like it. I just bought a thinkpad instead, shrug.
After months of waiting on preorder (in which time I could've cancelled it), my Framework 16 finally arrived just yesterday, so this is a very unfortunately timed article for me.
Mine is using the AI 300 chip and is presumably on the latest bios so this specifically wouldn't be a problem for me, but it doesn't fill me with future confidence either, though I suppose if you look hard enough you can find similar stories about any manufacturer
> Still, I was very disappointed that Framework chose to entrust sensitive information to a third-party who obviously couldn’t keep the data safe.
Then you're going to be very disappointed a lot. When HIBP first came out, I worried over each incident. These days, breaches of non-payment data are so common that I don't even bother to sign up for notifications.
When it comes to PII, we live in a de facto panopticon.
Honestly, this is why I like Dutch consumer law. Dutch law states that the guarantee duration is dependent on the expected lifespan of a product. Continued official BIOS support is evidence that the device remains within its supported lifetime and also dutch law treats defective or incomplete updates as non-conformity. Therefore, in this situation it would be framework fault and the device would also be within warranty.
I have a framework 13 for. Actually the same model as OP (AMD 7040 series). I have never been as conflicted on a laptop as this one. I do love many things about it : The whole expansion card system is awesome. It does feel a bit redundant with a simple usb-c hub that do the usual network, hdmi, usb-a, but being able to set it on the side that fit my current setup is much appreciated. I love the form factor and the screen ratio also. The linux support have been very good for me. And being able to open it up with just one tool and have detailed repair manual is amazing.
But on that last point, it's also the first laptop in a while that I had to get into to fix something. The webcam never worked. That's not a linux issue, it's just that my fw13 is a hand me down from a previous colleague who I believe borked the webcam installation. But I don't use the webcam, so I never cared. On the other hand, when I first got it, the cpu fan was emitting a constant high pitch whine. After some research I wasn't the only one affected, and the support had me do a lot of steps before sending a replacement part (which was very easy to install, but the problem should not have been there in the first place). I also had some random crashes that I was never able to debug. The mic is very meh also. I never use it in anycase with my setup but still.
But at the same time, I don't see myself switching to another laptop. Despite all, this is still the laptop that felt the most "right" to me. I do believe that Framework has to seriously step up their support and QA game. But it does feel like it's a company that is "suffering from success". They got big investment, lots of publicities in some circle and so I feel like they have to produce and sell way more computer than the rest of the business can reasonably handle. I hope it's a “necessary evil” for them to get enough customers and revenue to start being able to attract more investment and being able to invest more in the “quality” side of the biz.
EDIT: I should say, being in the E.U make it so that I have an automatic 2 year guarantee, so I didn't have to experience the whole "just buy new parts lmao"
I had a similar story with Framework support.
I've bought 12th gen i7 FW, and after about a year, a fan bearing had failed. Despite originally sending me a laptop through a freight forwarder (FF), they both refused to send me a replacement fan for free, and to sell me one to FF address.
I had to use other means to buy myself a new fan (It is not sold separately from the heatsink), paying around 50 bucks for it.
After another year my motherboard stopped working completely. Framework refused support besides the post light interpretation, did not give any schematics, and outright ignored emails from my company with requests for schematics. Autopsy shown that CPU just freaking cracked in the middle.
I've got a replacement 12th gen i5 from a friend of mine, and not 3-4 months later it met pretty much the same fate, except it looks a tad bit more alive (no visible damage to the crystal, no blown mosfets, no nothing). I cannot diagnose power circuitry without spending time to reverse engineer it, and no local shops would accept it.
At this point, given that you don't have any documentation nor parts available in most of the world, FW is generally less repairable than a macbook. For macbooks you can at least buy docs at Vinafix (and for most other laptops too), and chineese replacement parts pretty much anywhere.
I don't really care what they are telling about «waa waa cant release schematicc nda» or «scammers use ff u will die using it». System76 doesn't care about FF, and provides warranty replacements. They could have tried to renegotiate NDA terms for older boards, or outright let someone leak them; they could have remembered their Coreboot promises; they could have done lot more to actually make something that doesn't just look repairable and hackable.
Instead, they've spent their money to make a nice to use, yet unfixable piece of trash currently lying in my closet and waiting for some cheap board to show up on aftermarket.
So, heed my warning. As good as their promise sounds, you better off supporting someone who actually cares about the mission of repairability, and not just doing whatever poser shitshow they are doing.
I apparently have a very uncommon viewpoint on this, given the flavor of the comments.
I won't try and say that this situation is totally fine, or ideal, or anything close to it, but I think that in order for us to know how to feel, we need one very important piece of information that is missing: was the actual BIOS update itself bad in some way, or was this just the kind of error that happens sometimes, especially as hardware ages.
If it was the former, then this becomes quite bad: Framework released (even if they were not the ones who made it) a software update that bricked someones computer, and it is incumbent upon them to fix it, even out of warranty.
If it's the latter, then, I'm sorry, but that's the kind of thing that happens sometimes. Yes, having dual bios chips or some other way to recover from a failed flash would be nice, but I don't know the engineering trade offs to make that happen. Lots of desktop motherboards have it, where space is not at a premium, but do any laptop motherboards have it? My guess is that, at the bare minimum, it's much much less frequent.
I think that, in the latter situation, the main thing that I would like to have seen from a company like Framework, with their ethos, is to provide the pogo pin reflashing tool (at a reasonable price), along with the necessary instructions and files, either for the user, or for a local, inexpensive computer repair shop (depending on the users preference). So even in the latter situation, I do think that Frameworks support is lacking here. This seems like the kind of thing that, with the proper tools (which seem cheap) and instructions and files, that this wouldn't actually be that hard, especially for the type of consumer that would buy an upgradeable, modular laptop.
But, in the case that the BIOS itself was not bad/caused the bricking, then I think the rhetoric around this is overblown. I'm not sure that any laptop I've ever had before has ever told me that I needed a bios update (aka: provided that level of support post purchase).
Again, if the BIOS was in fact bad, and caused the brick (and no, a thread of people who experienced the problem is not enough to conclude this), then that's a very different situation, and the level of failure exceeds even what is described here. And even if it wasn't bad, I don't think think that "buy a new mainboard" is the appropriate service response when "buy these $20 dollars of parts and tools, and follow this guide" would likely have solved the problem.
But, as I understand it, BIOS updates are always somewhat dangerous. If Framework is on the hook for every time a BIOS goes bad outside of warranty even in situations where nothing they did directly caused the issue, then what they will do is just stop offering BIOS updates outside the warranty window. We can hold two ideas in our heads at once: Are they completely and perfectly living up to their ideals? I think that the answer here is obviously, even in the least bad case, no they are not. But are they providing better service than other laptop manufacturers? I think that as long as the BIOS file wasn't bad, the answer is yes. No laptop manufacturer would have offered to fix this out of warranty if they even provided continued BIOS updates, and for most manufacturers, their answer would have been "buy an entire new computer", and your only alternative would be third party resellers or used parts on ebay to get a new motherboard.
-edit- After doing a bit more research, BIOS recovery paths in laptops, while not universal or as common as in desktops, are also not exactly rare, and not having one, in a company with Frameworks ethos and stance, is worse than I initially assumed.
I think that, even if this isn't due to a bad BIOS or updater (which, again, would make them, in my opinion, liable for fixing it), then this is a very significant oversight, and the fact that they don't provide the support necessary to route around it (reasonably priced tools and clear, well written instructions) is even worse.
the conclusion section is a pretty damning and disappointing read.
i'm realizing that framework's marketing swayed my opinion a lot more than i thought. i was seriously considering a framework laptop when my current one dies, but i'll probably just end up going with a thinkpad again, as the level of serviceability seems a lot closer to equal than i was lead to believe, and apparently there is no "feel good" advantage of buying framework.
this is disappointing. I need a new machine soon but I am not interested in Framework if they are going to play the usual games.
Is thinkpad really it? I wasn't wildly excited about that one either.
I've had some rough moments with my Framework 13, but overall I appreciate it because when something like this happens you have a way out other than binning the entire laptop. I agree with the author that Framework should have a better answer than buying a new motherboard, /but/ at least you can buy a new motherboard and move all the components over. Realistically, any >$1000 computing device should have the ability to boot BIOS from USB to provide a fallback if the BIOS is damaged, this is pretty standard on most high-end motherboards, and I think its a fair expectation from Framework. Hopefully they will see this thread on HN and make changes to the product in the future to improve this.
Disappointing to read. I think I'll just keep buying used laptops and throwing Linux on them.
I think cases like this should be brought before a small claims court. It seems unfair. The company (Framework Computer) provided the software, but the software is faulty and broke ("bricked") the laptop. Can you imagine if a GPU driver caused the card to get bricked? Even if out of warranty, surely the manuf would be legally liable in many jurisdictions for this faulty driver.
The author's LinkedIn profile[1] tells me that this person lives/works in Toronto, Ontario, Canada. While the issue is already fixed, for others affected with the same issue, I would suggest the small claims route. Make sure to talk with an LLM to follow the correct steps to gather and prepare evidence for your claim. If you are lucky enough to win, then you should "go nuclear" on social media (and their own forums) to share the win and the exact details about how you did it. Framework Computer will surely (1) cave and provide replacements or (2) be flooded with small claims about this issue.
[1] https://www.linkedin.com/in/guanzhong-chen/