Can someone post high resolution photos of the L5's M.2 cards?

Well I’m not disassembling mine for you … :slight_smile:


This might help (be good starting point, until you get some decent pictures):

FCC ID XF6-M7DB6 and XF6-M15SB:
Source info:


Perhaps open an issue about the schematics and see what can be done.


AFAIK, you don’t really need to disassemble the phone for that. Just open the back side and remove a small plastic piece attached with a couple of screws.

1 Like

I was considering taking a picture and removing the small plastic piece, but did not because there’s also a metal shroud around the components on the M.2 PCBs (as seen in the PDFs that @Quarnero sent).

1 Like

Argh, @amosbatto I Would love to disassemble mine for you but I don’t think its shipping for another 5-6 months (~sep 2019)

So, I am no photographer and I don’t have a good setup for taking photos like this or even a real camera, but I took a shot at it anyway. Note that this is a Chestnut device though as far as I know, the two m.2 boards haven’t changed in later L5 revisions.

EDIT: I accidentally uploaded a shot of the front of the BroadMobi card with the identifying numbers included. I have now upladed one with those numbers obscured. Hopefully that doesn’t come back to bite me.


EEK!!! You still did. You need to black out the QR code, which gives your S/N and IMEI !!!

Gotta love QR codes. :frowning:


Well I feel foolish. Oh well…

Thanks for the heads up. At least I can buy a new modem if this becomes a problem.


Maybe open source browsers should warn / manage this automatically i.e. when you upload an image check for a QR code in the image - just as they ought to do with image metadata.

Thanks @FamousJameous.
On some of the photos it is hard to tell whether there are just solder balls or a capacitor/resistor is attached.

I count 51 components in this this photo. Can you check whether you see any more?

If the count is 51, then here is my conclusion on the Components List wiki page:

If the test elements aren’t included (since they are printed in the board), then the two Librem 5 boards together have 1174 components. In addition, the RS9116 WiFi/Bluetooth M.2 card contains 88 components and the BM818 modem M.2 card contains 6 components, so the 4 boards in the Librem 5 contain a total of 1268 components.


The spot you have labeled 46 is empty, so the count would be 50. That is the only difference I can see. I’ll try again a little later to see if I can get a better picture that can be included on that wiki page.


@FamousJameous, Thanks for spotting #46. I have updated the wiki page so it now says that there are 1267 components on the 4 boards.

We can now say for sure that the Librem 5 has 2.5 times more components than the standard smartphone.


I think you can strike the “more”.

I would interpret “x times y” as x * y and I would interpret “x times more than y” as x + x * y.

I would interpret it differently from you. I think that what @amosbatto has written is OK (correct).

To clear it up … he gives the number of components for the Librem 5. If he gives the number of components for a “standard phone”, whatever that might mean :slight_smile:, then people can go with those numbers and they can work out for themselves what the ratio (2.5) is.

I think any possible confusion relates to “percent more”. “2.5 times more” means “2.5 times”. “250% more” means “3.5 times”. So “times more” is multiplicative while “percent more” is additive.

@FamousJameous, one more request. Can you read what is printed on chip #43?

It is the only IC which is unidentified in the whole phone, and I think that it is important for users who care about security to know exactly what every chip does and where it comes from.

Unfortunately, I don’t think that Purism can release the schematics for the RS9116 M.2 board, since that is controlled by Silicon Labs, but we can give people some assurance about the purpose of each chip on the board.

It looks like:


The datasheet for LSF0108 from TI describes it as a bi-directional voltage level translator and has a VQFN package option that looks like it could match what’s on the board. So, that may be correct.

Here is the datasheet from TI that I was looking at:


I always take “2.5 times more” to mean “3.5 times as many”, just like with percentages. I don’t think I have ever considered that someone might intentionally write “2.5 times more” with the meaning “2.5 times as many”. That makes the word “more” redundant, and I don’t expect it to be redundant because it seems as though it is in an important position.

So, it looks like there is disagreement on the meaning. Therefore, “times more” is ambiguous.

(Edit: sorry, didn’t notice last post was 15 days ago. This comment is hardly important enough to bump the thread.)


Yes. There is disagreement (among native English speakers no less :slight_smile:) so that suggests a reword would be in order, @amosbatto. (Of course technically it is a Wiki and anyone can go in there and hack it around, right?)

However I would draw your attention to my comment above: as he gives the two numbers (1267 and “between 400 and 600”) there is no confusion or disagreement.

And my other comment: There isn’t really such a thing as “typical smartphones” so the figure of “2.5” is inherently an approximation, so perhaps it doesn’t matter whether it is “2.5” or “3.5”. An impression is conveyed either way. “2.5” is not, and is not intended to be, an exact number. You can work out for yourself the actual range for the ratio based on “between 400 and 600”.

In the wiki, I changed it to “2.5 times as many as a typical smartphone”, so that should clear up any potential confusion. “Times more than” sounds less awkward to my ears than "times as many as ", but It never occurred to me that anybody would interpret “2.5 times more” as 350%. Oh well, clarity is more important than a phrase that sounds nicer.