Kite SAIO LCD washed out / text illegible

[rogerimp]

Hey all. Let me preface by saying total noob, first post, first GBZ build. I tried to find a similar problem in the forums, but if my search-fu failed, I apologize and will happily accept a link to a similar problem.

With that out of the way, I'm midway through my Kite SAIO 0.6c build, just got the GPIO all soldered up and the LCD connected. The test image shows all colors but the blacks are totally whitewashed - there's no shadow and no contrast. When I boot into normal mode the emulationstation text is basically illegible as it appears light gray on white.

PhotoShow

Any ideas? Particular GPIO pins to try resoldering? Thank you in advance.

[YaYa]

Have a read on the wiki, LCD troubleshooting chapter.
https://github.com/geebles/Super-AIO/wi ... leshooting

You have the pinout settings used by the LCD. You will need to rework them and make better soldering of them.

Could you post a close up picture of your solderings ?

[rogerimp]

Yeah, I've been using the pinout guide here, but I'm a bit unsure where to start, as the red, green, and blue colors all appear equally well on the test image. I suppose I'll just start with the pink pins and go from there testing all the way... Maybe I'll begin by reflowing all the GPIO solders with a heat gun. Will post pics and updates in a few.

Thank you!

[YaYa]

Yes, start with the pink pins !

[rogerimp]

Wellp. I've reflowed all the pins twice with a heat gun and gone over specifically the pink pins with the iron. Reseated the LCD ribbon a couple times, cleaned the contacts with alcohol. I'm gonna test every pin for continuity with a multimeter next. I don't want to blame the hardware but a nagging little voice is wondering whether the LCD display is bad. After every step listed above the screen appears the same way with high brightness. Another hint may be that there seems to be some kind of banding where every other row seems a bit brighter.

Here are the solders currently.

frontShow

backShow

Side Note: I was following kite's soldering video and PDF manual and so pre-tinned the SAIO-side connections before realizing the current recommended method uses header pins. By that time I just decided to go through with the direct-contact method. As a small suggestion the original Support Thread says use the PDF manual as the primary documentation, however it seems that the WIKI is the most current recommended build guide - and probably should be recommended over the PDF?

[YaYa]

The pdf is about to disappear. It’s too difficult for Kite to maintain it. I’ll give a whistle to Kite so he look at your problem. There is nothing more i can suggest. But kite is very good at troubleshooting and he knows very well his hardware

[kite]

Hi may I ask why you soldered through the holes rather than using the pins shown here? https://github.com/geebles/Super-AIO/wi ... SAIO-v0.6c You could have had a better time using those pins if you followed the PDF then that may be the final nail in the coffin and I will delete it from github

You can ignore the pink pins, you would see nothing (black/grey screen) if there were problems with those

Try re-soldering with an iron the BLUE pins, use a multimeter to probe each side to make sure they are actually connected. Do you have any spare screens (any 3.5" 'reversing camera' screen is the same type, otherwise search for LQ35NC111 as it MAY be the LCD itself..

[rogerimp]

Hi Kite,

First off thank you for an amazing board and your reply.

To answer the Q, I was following a combination of the PDF, wiki, and your excellent soldering video. I did notice that the soldering video and wiki were more recent. When I got to starting soldering the Pi Zero onto the SAIO board, I started referencing your soldering video for that portion as it contained not only lots of detail but really good asides... thank you for your narration there. Anyway I was aware that the current version used header pins. Yet when I followed the step in the video and pre-tinned the SAIO side I thought this step also applied to the header pin method. Only later did I realize that the correct method lets the pins touch through to the tape on the bottom, and that instead I had made a solid bottom plate that those pins wouldn't guarantee a good solder with. I had to make a choice, either to clean off that solder or to continue with the direct-contact method. In the end did the latter since 1. your video provided excellent documentation thereof 2. no risk of cutting the board when cutting the pins 3. nicer final look 4. I had achieved some success with this method already on the USB & SD card pads, and was emboldened to try it for the GPIO pins as well.

Unfortunately there is good wisdom contained in each of the PDF, the wiki, and your videos - so it's hard to recommend simply letting one of them go. Perhaps going forward it might be good to disclaim up front which version of your board any instructional medium refers to. And as I noted above, indicating in the original support thread post that the wiki is the most up-to-date and should be the primary reference, might help.

Anyway, obviously my mistake and I'm solely responsible for understanding the instructions correctly.

Back to the screen. I think it may be a clock/timing issue. The blue pixels seem to display a bit to the left of where they're supposed to, and this alternating rows thing I mentioned is only visible in photos, optically all the pixels seem equal in brightness. I'm uploading a video and a closeup.

Video (externally hosted)

still closeupShow

I did try resoldering all blue pins with no noticeable change. I tried especially to eliminate bubbles, pushing the iron tip directly through the hole rather than the usual oblique pad method. No change. Just tested all blue pins for continuity between sides of the solder and all good.

I will try to look for an old digital camera or somesuch to sacrifice though I do wonder about the odds it will have the same interface / DPI config. I'm a bit beyond my depth there. Tried a PocketCHIP screen but it uses a much smaller ribbon connector. Any hints where to look for a replacement screen? digikey, aliexpress, sparkfun not turning up anything, google only suggesting Thai retailers, it would seem. [EDIT] Aha! I assume part LQ035NC111 is the same, adding a zero there comes up with lots more results. [EDIT] Ordered one, so, unless there are other things to try, will have to put this on hold for a month or so.

Again many thanks for your help!!!!

[keffcat]

@kite I'm having the same issue, and it's because my GPIO28 isn't properly connected - I accidentally ripped off a pad while soldering (on both sides somehow...)

Is there an alternative place where I can bridge the connection?

[kite]

@kite I'm having the same issue, and it's because my GPIO28 isn't properly connected - I accidentally ripped off a pad while soldering (on both sides somehow...)

Is there an alternative place where I can bridge the connection?

Hi there isn't a "GPIO28", which pin in that image are you talking about?