Dead Pinecil that will go into DFU and take a flash - can it be recovered?

[deftdawg]

Pine Microsystems Inc. Support #4 directed me here after I opened a support ticket for a Pinecil that won't come on again after a single session of soldering.

According to them, because I am able to get it into DFU mode (holding down - while connecting to USB on Windows) and the Pinecil takes a flash from the firmware updating tool, the Pinecil is "working". To me a soldering iron that doesn't solder isn't really "working"...

In any case, is there a way to get the Pinecil to come back to life or something I can do to debug it?

Here's the output from flashing

Pinecil Firmware Updater v1.3
Looking for firmwares and latest version...
Downloading firmware...
Flashing...
dfu-util 0.10
Copyright 2005-2009 Weston Schmidt, Harald Welte and OpenMoko Inc.
Copyright 2010-2020 Tormod Volden and Stefan Schmidt
This program is Free Software and has ABSOLUTELY NO WARRANTY
Please report bugs to http://sourceforge.net/p/dfu-util/tickets/
Opening DFU capable USB device...
ID 28e9:0189
Run-time device DFU version 011a
Claiming USB DFU Interface...
Setting Alternate Setting #0 ...
Determining device status: state = dfuIDLE, status = 0
dfuIDLE, continuing
DFU mode device DFU version 011a
Device returned transfer size 2048
DfuSe interface name: "Internal Flash  "
Found GD32VF103, which reports a bad page size and count for its internal memory.
Fixed layout based on part number: page size 1024, count 128.
Downloading element to address = 0x08000000, size = 48800
Erase   	[                         ]   0%            0 bytes
Erase   	[=                        ]   4%         2048 bytes
Erase   	[==                       ]   8%         4096 bytes
Erase   	[===                      ]  12%         6144 bytes
Erase   	[====                     ]  16%         8192 bytes
Erase   	[=====                    ]  20%        10240 bytes
Erase   	[======                   ]  25%        12288 bytes
Erase   	[=======                  ]  29%        14336 bytes
Erase   	[========                 ]  33%        16384 bytes
Erase   	[=========                ]  37%        18432 bytes
Erase   	[==========               ]  41%        20480 bytes
Erase   	[===========              ]  46%        22528 bytes
Erase   	[============             ]  50%        24576 bytes
Erase   	[=============            ]  54%        26624 bytes
Erase   	[==============           ]  58%        28672 bytes
Erase   	[===============          ]  62%        30720 bytes
Erase   	[================         ]  67%        32768 bytes
Erase   	[=================        ]  71%        34816 bytes
Erase   	[==================       ]  75%        36864 bytes
Erase   	[===================      ]  79%        38912 bytes
Erase   	[====================     ]  83%        40960 bytes
Erase   	[======================   ]  88%        43008 bytes
Erase   	[=======================  ]  92%        45056 bytes
Erase   	[======================== ]  96%        47104 bytes
Erase   	[=========================] 100%        48800 bytes
Erase    done.
Download	[                         ]   0%            0 bytes
Download	[=                        ]   4%         2048 bytes
Download	[==                       ]   8%         4096 bytes
Download	[===                      ]  12%         6144 bytes
Download	[====                     ]  16%         8192 bytes
Download	[=====                    ]  20%        10240 bytes
Download	[======                   ]  25%        12288 bytes
Download	[=======                  ]  29%        14336 bytes
Download	[========                 ]  33%        16384 bytes
Download	[=========                ]  37%        18432 bytes
Download	[==========               ]  41%        20480 bytes
Download	[===========              ]  46%        22528 bytes
Download	[============             ]  50%        24576 bytes
Download	[=============            ]  54%        26624 bytes
Download	[==============           ]  58%        28672 bytes
Download	[===============          ]  62%        30720 bytes
Download	[================         ]  67%        32768 bytes
Download	[=================        ]  71%        34816 bytes
Download	[==================       ]  75%        36864 bytes
Download	[===================      ]  79%        38912 bytes
Download	[====================     ]  83%        40960 bytes
Download	[======================   ]  88%        43008 bytes
Download	[=======================  ]  92%        45056 bytes
Download	[======================== ]  96%        47104 bytes
Download	[=========================] 100%        48800 bytes
Download done.
File downloaded successfully
Your Pinecil was flashed successfully! You can disconnect it safely.

[HanaO00]

Hi, there are surely more competent people than me here, but I can possibly bring my stone to the building,

• Does the Pinecil heats up pressing "+" ? It may be a screen problem, if the Pine64 support indicates that the Pinecil is "working"
• Have you tried to flash another version of the firmware as a test ?

[deftdawg]

Your stone is appriecated 😄

• sadly + doesn't do anything, I tried tapping it a number of times, holding it down, etc; tip stays cold. I left it plugged to a power supply for a few days, when I picked it up, the screen woke, I moved to another room plugged in and then it was dead.
  I'm think the screen should be physically ok, it wasn't dropped or jostled in anyway.
• For PSUs:
  • First one I tried was a Dash charger, but it only puts out 5V - DC too low (Dash != QC); that was before my soldering session
  • Next I used an Asus Chromebox AC adapter -> 19V @ 3.42A which seemed fine for soldering until I disconnected it
  • After it died I also tried a HP AC adapter -> 19.5V @ 11.8A
  • Voltage of the two AC Adapters was validated by a multimeter
• I have noticed that to bottom area under the screen does get a little warm when the pinecil is connected to USB or a power supply, I wasn't paying attention to it before (was only gripping on the silicone sleeve) so that may or may not be something of interest
• Tried updating to 2.14.1 and 2.15, the flashing output is the same as above for both. Mass Erase fails with DFU-Util exited with error code 74!, but I understand that is typical for many pinecils

[Ralim]

Hmm, I have no special wisdom here really, if the MCU is flashing correctly and the screen isnt coming on that really only leaves the I2C bus. The parts on the I2C bus are: FUSB302 and the OLED itself.

I dont use the flasher tool so not 100% on the output, but could you try reading the firmware back out of the MCU just to confirm its a 1:1 copy (aka flashing worked correctly).

If the firmware has flashed correctly its possible its the OLED or the FUSB302.
If your comfortable it may be worthwhile opening it to see whats getting warm, as nothing should get overly warm from usb power.

[deftdawg]

Count down to the sixth pin from the top of that image. Soldered, soldered, soldered, soldered, soldered, marginal, soldered, soldered... It was a dicey proposition softening the foam adhesive to swing the OLED out of the way so I could inspect its solder joints, but as soon as I did, the problem was right there and the fix was easy enough. Have a look, and good luck!

Thanks, that's encouraging... I used a flat head to stab the adhesive foam away from the board side, the OLED came loose rather suddenly at the end, but everything seems intact.

I'll have a go at it with my multimeter when I get back to the house tmr to see if there's continuity from ribbon to pad across each solder joint and if there's any bridging (continuity) with joints above and below each line. Hopefully that will show me what to try to rework.

[myself248]

Wow, unless I'm mis-reading your photo, it looks like MOST of your joints are bad. I've annotated my image because I think it's clearer to start from, and because I was able to examine it from multiple angles to confirm my understanding:

All the green lines show a continuous bead of solder from the board up to the OLED flex. The red line shows where there's a junction that didn't get made. It would sometimes make contact if the pressure and temperature(??) were just right, but sometimes not.

Looking at your image, it looks like only the first and ninth have that nice continuous bead, all the rest appear to have a step profile rather than a swoosh profile, if that makes sense.

I would pull the OLED completely off, clean the pads with braid, and reapply fresh flux and solder. I have some spare OLEDs here and I'd be happy to take some step-by-steps if that'd be helpful.

[deftdawg]

I got back to looking at it... I'm afraid there's no way I'm going to be able to fix this (assuming this is the issue), it's simply at a scale that is too small for my eyes and my hands to work with. I'm not even able to see the lines as I try to pin them with my multimeter, though I am able to get continuity across some while working blindly.

In any case, I think I'd need access to a microscope set-up (something like out of Louis Rossmann's workshop) and a way steadier hand than I have...

Thanks all for the help.

[myself248]

Wow, you got those photos without a microscope? I'm impressed!

Yes, you'll definitely need optics for this. I'd try it with a swing-arm magnifier or a head-mounted visor thingy, both of which are considerably more affordable than the gold-standard binocular or Mantis scope. Does your local hackerspace/makerspace have any of the above?

Or, since I think we can conclude with high confidence that it's an assembly defect, perhaps push back on Pine support and pursue a return/exchange. Best of luck.

[pwllheli]

I had the same problem. Thanks for the suggestion - it fixed it!