Known issues on early Framework Laptops

just received my Batch 2 laptop configured the insides installed my OS, attempt to put in my wiFi password noticed my keyboard not working properly. opened it back up keyboard/touchpad shorted… just saw forums about this… not happy; would’ve been nice to have something on the envelop that said please open before usage. envelop literally is just a plain sealed envelop

The shielding is give us more margin to pass EMI compliance. It should function fine without it, but it should also be no problem to take your two pieces and sort of stick them down in roughly the right locations to mostly shield.

2 Likes

@nrp to @End 's point.

Maybe add a link in the ‘fixing to ship’ email to this thread?? Just a thought. I’m trolling the community daily but some people may not be. In my opinion for being only on batch 2, the list of ‘issues’ is not really that long, being identified and are fixable.

Thank you guys for being so transparent. Keep up the great work!!!

3 Likes

PCB designer here - Do you have evidence it was actually impactful on a particular test, or was it designed in as a matter of best practice?

Agreed. I use Noctua in most everything I build or work on. So quiet compared to the routine fans out there. And they come with both mounting screws and rubber pins for vibration-reducing mounting. Seriously overpackaged, though!

1 Like

Yeah, you’re going to get people claiming that “noctua is overpriced” or “x fan is just as quiet”

People don’t realize that noctua fans are super high quality. The other point is not just the volume of noise but the frequency. They supposedly build their fans to avoid certain grating frequencies and it shows. I’m slowly coming around to the idea of a full noctua conversion in my PC but it’s expensive and my dang 2080ti is noisy enough to overshadow the fan noise anyways.

I’d love for noctua to make some smaller fans for things like laptops and GPUs. Then again, they’re still having issues making high end fans that aren’t B R O W N

2 Likes

Yeah, I have been using Noctua fans for almost a decade it seems. They are so UGLY… but work SOOOo well… reminds me of something?

image

1 Like

I love Noctua’s brown fans. Gives me a great excuse to shop for PC parts that don’t aesthetically match or anything- just throw an opaque case panel on everything

@Alex_Whittemore

PCB designer here - Do you have evidence it was actually impactful on a particular test, or was it designed in as a matter of best practice?

Is there something you’re concerned about related to that? I’m thinking about buying one and I’m trying to get as much info as possible about what I might be getting myself into

Absolutely not, even the opposite. My gut is that it’s intensive butt-covering, but I’d love to find out that the minute impedance discontinuity going into the connector actually does radiate enough to be a problem in testing.

2 Likes

We didn’t test it without, so I guess you could call it butt-covering! It is indeed best practice taken from many previous notebooks our manufacturing partner has worked on.

3 Likes

We’ve found a workaround for the S4/S0 LED blink issue (which also resolves some other cases where the laptop won’t boot and the LEDs on the side are blinking). We’ve added it to the bottom of the article here: My Framework Laptop is not powering on

We’re continuing to work on determining the root cause of the issue in order to resolve it in production.

4 Likes

There are some ACPI issues on Framework laptops. It reports the TPM2 eventlog over the TPM2 table, but the kernel returns the function which causes the /sys/kernel/security/tpm0/binary_bios_measurements file to never be created.

It can probably be fixed with a BIOS patch, but the function shouldn’t error without reading it off from the EFI configuration table so I have submitted a patch to try correct the general behaviour.

https://patchwork.kernel.org/project/linux-integrity/patch/20210920203447.4124005-1-morten@linderud.pw/

Should be noted that this isn’t isolated to Framework. I have seen this from other vendors as well. But I have never been capable of reproducing this personally until now :slight_smile:

5 Likes

Our gain - thanks for helping to diagnose!

@Foxboron Could you elaborate a little bit more on the ACPI issue you are seeing? We can look at fixing it.

I’m not quite sure about the actual issue. It’s a little bit beyond me and hard to debug without having to build and run the kernel on the Framework multiple times.

Essentially it seems like Minimum Log Length and Log Address in the TPM2 table is wrongly reported? The kernel is unable to take the values and map the memory for the log which abort the function.

It can be fetched from the EFI configuration table just fine though.

The issue is that it would be easier to debug if I knew what a proper ACPI TPM2 table with an eventlog looked like :slight_smile:

I seem to be having this issue now. My touchpad has worked reliably for a few months, then yesterday I noticed clicks stopped working (crept up on me since I usually tap-to-click by habit).

If I push firmly center-bottom, it goes back to working normally, but it seems after a bit it stops registering again.

I have to say, it’s REALLY nice to search for an issue like this and very quickly find “yes, it’s a hardware problem we know about and haven’t figured out yet and we’ll send you a new input cover if you want one.” Also the whole “if there’s a problem we’ll send you the parts and you can repair it yourself.” This is exactly what I signed up for.

7 Likes

Let us know how the temps are, if you add kryonaut to the Framework

I’m one of the batch 1 DIYers facing the S4/S0 issue after five months of hiccup-free experience. I’m curious if there’s any update on the root cause of this issue.

An RTC mainboard reset fixed it partially — laptop boots up now provided all expansion cards are unplugged. It gives S4/S0 blink codes if try powering it on with all four expansion cards plugged in.

My replacement mainboard is on its way. I’ll be happy to send my current mainboard back for analysis.

1 Like

I have also had an issue, for several months, with the trackpad not accepting clicks. It has been difficult to determine, though, if it is a hw issue or software. I am using Ubuntu 20.04.1. Configuring the trackpad is confusing. (Does it accept 3.find clicks? Does a bottom-middle click act as right click? etc)

It would help if there were more tools to test/check the hardware. For example, put an icon into the Favorites bar. Have that icon light up when the hw receives a trackpad signal. Then I can look at it and click around the trackpad and see where it might not be responding.