So I recently stumbled upon the CM4 MATX Project to create a Micro ATX motherboard for the RPi Compute Module 4. Does anyone know if this will be compatible with the SOQuartz or if anything similar is planned by Pine64? I've been waiting forever for an affordable, low power desktop ARM board.
Posted by: nixcamic - 02-05-2022, 11:38 PM - Forum: General
- No Replies
Hey I don't think the sign up question for the wiki is that good, A: Some of us don't know what the first community edition of the Pinephone was. I own a bunch of Pine64 products but had no idea. B: The answer isn't that easy to google, I found it but it wasn't "first result of first search". C: It seems to be the same capatcha every time? At least refreshing didn't give me a new capatcha which means that once spammers figure it out once they can make unlimited accounts.
I don't think it's limited to Mobian or Phosh, but I can't get the F1 to F10 keys to work.
I can use the Fn + number to get the other characters/symbols, although it's a known issue that some of those are incorrect at the moment. As far as I know, you use the Pine/Super L key as the modifier for the actual functions, but that just outputs the number, as if you hadn't used the Pine key.
The Pine key registers as Super L in xev, so it's doing something at least.
Curious if anyone has or hasn't got these keys to work as expected, and if they have got them working, what was involved in doing that.
Today I received my Pinephone keyboard case and after waiting for it to reach room temperature and then completely charge, I was ready to try it.
The first power up was a nice surprise: it worked perfectly fine with Mobian (on uSD). It also works with PostmarketOS (on eMMC). Great!
But it didn't charge the phone in either OS. Bummer!
To cut a long debugging story short, here's THE PROBLEM:
- the pogo pins on the Pinephone don't all make contact, hence the title.
Quality control on the Pinephone needs to be improved.
The DCIN pin in particular was responsible for not charging the phone.
See attached picture: 01_problem_pinephone_pogo_pins.jpg
THE FIX is to make all the pogo pins make contact.
In my case, just the DCIN pin since the I2C worked fine.
My solution was to use a copper tape which has adhesive (not conducting!) on one side.
However, a piece of aluminum foil/thin copper or other metal will work just as well.
THE EXPLANATION - a picture is worth a thousand words. Here's three more, in order as numbered:
02_explanation.jpg
03_fix_before_application.jpg
04_fix_applied.jpg
And just like that it worked! Charging the phone as intended.
Note #1:
Whatever you use, make sure you:
- fix the metal piece(s) in place firmly with some THIN tape
and
- make absolutely sure you don't short-circuit any pogo pins (with each other)
and
- make sure the fix parts don't touch any metal parts of the phone internals
Just cover any non-contact parts with THIN insulating tape.
Note #2:
An additional explanation for my fix:
The metal is longer in the bent over itself part do reduce the resistance of the metal piece itself.
Note #3:
The proper fix is twofold:
- Pine64 needs to improve the quality control for the Pinephones and make sure all the pogo pins are the same length once assembled
- the keyboard case contacts need to be shallower (less deep) to ensure differences in pogo pins length don't matter as much as now
I hope this helps more than one person with similar problems.
Now, to print some spacers as recommended in another thread and fix the first row of keys...
Updates
-------
In the end I added a plastic shim (0.37mm) under the POGO contact on the lid while still keeping the copper foil patch in place because I was losing keyboard functionality when the processor got hotter. A slight press on the back of the screen brought it back for a short while (before shimming) and after shimming seemed to work fine w/o any need to do anything. The shim raised the INT and DCIN pins primarily and all the others by a smaller amount.
It stopped working again once I started pushing the CPU AND at the same time increasing the charging current limit (from the case battery) to 1.5A.
This was done by writing 1500000 to
/sys/class/power_supply/axp20x-usb/input_current_limit
as pointed out in a number of places, for example in this thread: https://forum.pine64.org/showthread.php?...881&page=2
I have a script which does this and it has to be done after the battery starts charging because the OS will reset the limit to 500mA.
- turn on the case battery and wait for charging indicator
- write the desired limit value
- print the actual value to verify it's still correct
Or just leave running the graphical OS utility which displays live charging info.
Continuing, it was the same behavior as before shimming so I supplemented the shim with a second piece of plastic (0.80mm) and this holds for now. The original copper patch is still on but the POGO pin deformed it as expected so it's less effective.
For POGO pins shimming I just cut two small rectangles of ~1cm x 0.5cm and pushed them about 3mm under the base of the "U" shaped edge which surrounds the POGO pins piece of plastic on the kbd lid. I inserted first the slim one then the thicker one by barely lifting the kbd plastic with a spudger (also plastic, but sharp). This tool was inserted less than 1mm under the base of the "U".
See picture 04_fix_applied.jpg for the "U" I mention.
There is nothing special about the shims sizes or the shim pieces.
They're just solid plastic from product clamshells or any other throw-away packaging you might have around the house at any time. Just insert the slimmest piece first and don't go too deep.
The technique is mentioned in some other thread I can't find now and also in the kbd troubleshooting table on this page: https://xnux.eu/pinephone-keyboard/faq.html
Top row shimming - as explained here: "PinePhone keyboard number key row issue - solution" needs a safe method to pull the keys w/o damaging them or the silicone membrane below.
I pulled the keys using a DIY ghetto key puller modeled after a full size key puller. It's made of CAT3(phone) wire stripped of insulation and some heatshrink (optional). Not the best material but it fits between keys. Two loops of fishing line/nylon/thread would work just as well.
Hindsight: if you use wire, size it for the keys before twisting the wires together and make both loops the same size.
To use, slide one loop from the left, the other from the right completely under the key and pull straight up.
See picture: 05_key_puller.jpg
The key shims are a mix of 0.20mm and 0.15mm (I made a set of each) depending on how each key feels.
Curious how viable the PineNote with android is right now, and some other users have reported KOReader is working well for reading. How's the mobile version of Xournal++?
The only network you'll have service on is based on T-mobile by the end of Feb 2022.
Coming into 2022 service did work on AT&T. AT&T has decided on a whitelist for network access and the modem in the PinePhone is not on their whitelist. Because AT&T liked to lie about 4G its possible the older 4G display on the phone was really 3G they called 4G. It doesn't matter if they were supporting the phone in 4G before this switch or not based on "search to see if your phone will work" tools - come Feb 22nd or so.
Using the Verizon search tool the PinePhone is not listed as supported.
Pure Maps cannot get the actual position. It install with flatpak and maps can be seen. It works. Sorry. It was my fault due lack of knowledge. It is necessary to put the APN of the provider(I called my provider and gave it to me) in the cellular network and data and apn and I could walk several streets with the signal on the map
