Author Topic: Pro OLED Shield appears dead  (Read 7267 times)

ej.cardinal

  • Newbie
  • *
  • Posts: 6
Pro OLED Shield appears dead
« on: December 01, 2014, 10:28:48 am »
My brand-new OLED shield for the Pro is totally black after assembling and uploading the "demo" linked from the wiki page (https://github.com/Defragster/OLEDPRO). The code uploads without error, and the digispark pro does a simple "blink" test fine.  I'm pretty experienced in soldering, and the joints all look solid and clean. Can someone provide specs and docs on the shields, alternate libraries, or super-simple demo code so I can tell if it is truly Dead On Arrival?

Are the three (5V, Gnd, VCC) end "compatibility" pins used in the OLED shield, or can I pull them off and use the shield on a breadboard?

nxocorp

  • Newbie
  • *
  • Posts: 2
Re: Pro OLED Shield appears dead
« Reply #1 on: December 10, 2014, 03:24:50 pm »
I have the same problem. I have solder it like in the tutorial and have tested the digipro with simple sketches.
when i dock the OLED shield on the PRO, the power LED on the PRO goes out and the board slowly heats up. im thinking that there is some type of short on the shield board.

Can someone upload some instructions on wiring the OLED to a breadboard and doing some simple test.
Appreciate the help

defragster

  • Sr. Member
  • ****
  • Posts: 467
Re: Pro OLED Shield appears dead
« Reply #2 on: December 15, 2014, 03:45:05 pm »
I got this working with the Beta and breadboard and should have made better notes - looking this up I find:

Using the UNO I connected:
Red/Black show 5v on OLED across VCC/GND
SCL to ANALOG 5
SDA to ANALOG 4
Lines SDA/SCL each also go across a 4.6K resistor to 5V

AND same 4 wires to the PRO – VCC/GND measured 5v, and SDA>PB0 and SCL>PB2

*NEW library code in the Wiki solved the issues I found.  Thanks Marcus.  Works for me on UNO and PRO.
« Last Edit: December 17, 2014, 11:47:02 pm by defragster »

defragster

  • Sr. Member
  • ****
  • Posts: 467
Re: Pro OLED Shield appears dead
« Reply #3 on: January 06, 2015, 12:59:16 am »
I have now built three of the OLED's onto the PCB and all worked first time and for many hours now.  Did you guys get yours working?  I have posts and pics in other Topics here showing my units and some code, as well as what is in the example.  Maybe you got something bad or created a solder bridge?  If you have an UNO you can wire the OLED up there and the examples will work if the OLED does.

After I first saw these posts I ran my DMM on beep across before soldering and mine seemed clean and right and came out fine.

cborn

  • Newbie
  • *
  • Posts: 11
Re: Pro OLED Shield appears dead
« Reply #4 on: February 11, 2015, 03:59:07 am »
It appears there are some bad PCBs in the field.

I started putting some of my purchases together today. First the DigiSpark Pro, which happily blinked away, I modified the Sketch so it blinked faster and at varied rates, all good. Then I loaded up my first Combo Shield, pretty straight-forward, added the OLED display and prepared to load a test program.

The power LED came on, but it wasn't loading, hit Reset, still nothing, then my finger started burning (I was holding it as i plugged in the USB) and I saw smoke, pulled the USB out and the LD1117 flicked off the PCB and onto my desk!

I loaded the schematic and board into Eagle - that chip shouldn't even have been doing anything!
Ok, unplugged the Shield and plugged the USB in again - happily flashing its LED, so main board is at least partially ok, although missing the LD1117.

So, check out the Combo Shield - and I find Vin, GND, and connector pins numbers 7,8,and 11 are all dead shorted to each other.
Nothing at all obvious visually on the board, not easy to follow the copper without a PCB layout.

So, I go to the hot-air rework station and remove everything but the two resistors, and check again, just in case.
Nope, those lines are all shorted somewhere on the PCB.
My guess is a ground fill misalignment. Hard to see under the microscope (especially without a layout) but the drill-spots do appear off-centre on the pads.

When we get boards made here (I work in an electronics company and spend quite a bit of time on board analysis and rework) we usually get 100% e-test, looks like that wasn't done here. Perhaps the others who had similar problems could check their shields to see if the same lines are shorted?
« Last Edit: February 11, 2015, 04:02:48 am by cborn »

cborn

  • Newbie
  • *
  • Posts: 11
Re: Pro OLED Shield appears dead
« Reply #5 on: February 11, 2015, 04:02:13 am »
As a follow-up, I then built up my second Combo Shield (after verifying it had no shorts!) and fitted the same OLED display and DigiSpak Pro (minus LD1117), loaded the software, and the display worked beautifully!

digistump

  • Administrator
  • Hero Member
  • *****
  • Posts: 1465
Re: Pro OLED Shield appears dead
« Reply #6 on: February 11, 2015, 10:04:46 am »
Anyone experiencing these issues please email us at support at digistump dot com and we will replace any faulty boards.

cborn

  • Newbie
  • *
  • Posts: 11
Re: Pro OLED Shield appears dead
« Reply #7 on: February 12, 2015, 02:35:26 pm »
Anyone experiencing these issues please email us at support at digistump dot com and we will replace any faulty boards.

Email sent one day ago, with a followup containing thermal imaging which may help locate the problem, let me know if it helps.
Would appreciate a brief response just to let me if it got through or not, thanks.

cborn

  • Newbie
  • *
  • Posts: 11
Re: Pro OLED Shield appears dead
« Reply #8 on: February 20, 2015, 03:17:26 pm »
Anyone experiencing these issues please email us at support at digistump dot com and we will replace any faulty boards.

Email sent one day ago, with a followup containing thermal imaging which may help locate the problem, let me know if it helps.
Would appreciate a brief response just to let me if it got through or not, thanks.

Following up on this, I received a reply from Erik and unfortunately they received a bad batch of PCBs.
They are working through it now, hopefully the problem won't be too extensive.
Many thanks to Erik for his open communications on this issue.