0

1.8" TFT shield display problem
Moderators: adafruit_support_bill, adafruit

Please be positive and constructive with your questions and comments.

1.8" TFT shield display problem

by Zuhd on Wed May 10, 2017 6:51 am

Hi folks,
I rec'd a 1.8" TFT shield (w/ the SD reader & joystick) in Dec of 2016, and on receipt all seemed to be working fine after messing around with the graphicstest sketch and writing some example display code myself. I put it aside and just recently got around to wiring it to my project. In doing this I made a couple of simple modifications- using the through-holes inboard of the headers to reroute pins 11, 10, and 3 to other pins on the shield below it (which I mention only because it's the only thing I can think of that's changed since I first tested it).

When I finally got around to writing display code for my project, I found that the LCD display was powering on (displaying all-white), but not otherwise doing anything. I removed it from the project, disconnected the wires from the header-inboard holes, and wired it to a spare Uno using the default pins. When I used the graphicstest.ino sketch from the ST7735 library, I found that the display again did not work, only displaying white.

After I reset the spare Uno a number of times and was fiddling with the TFT shield to check for loose connections etc, the display randomly began to work, running through the graphicstest sketch before going white again after reaching the loop portion of the sketch. After verifying that this was not because of loose connections from the arduino itself, I reset the arduino a number of times and observed that the sketch would sometimes display and sometimes not display on the LCD.

I suspected that there is a loose contact between the display and the shield itself, but when I pried up the display off its adhesive strips to look, I couldn't identify anything (though it's a little hard to tell what's going on with the ribbon connector). There doesn't seem to be any particular rhyme or reason to the display going all-white- it seems to happen more often during the loop portion of the graphicstest sketch, and once it does, it seems to be difficult to get working again. It'll occasionally start working again after I disconnect the arduino's power or something like that, with no change to the physical position of any of the circuit's components, so it's not clear to me that it's a loose contact somewhere. I really have no idea where to begin troubleshooting this, or what the likelihood of being able to fix this kind of problem is. Any help would be greatly appreciated.

Zuhd
 
Posts: 11
Joined: Wed May 10, 2017 6:22 am

Re: 1.8" TFT shield display problem

by Zuhd on Wed May 10, 2017 9:07 am

After some more testing I'm pretty sure this is related to a lousy soldering job on the SMD mounts for the ribbon connector to the TFT display. I was able to make the display work a more reliably by dragging an exacto knife between pads with solder domes that looked a little too close for comfort, and heating a couple of others. I'm not sure this is completely fixed, but it seems to be the case that the solder job on the ribbon connector is probably the place to look if people are having this sort of weird problem.

Zuhd
 
Posts: 11
Joined: Wed May 10, 2017 6:22 am

Re: 1.8" TFT shield display problem

by Zuhd on Thu May 11, 2017 6:48 pm

Had a chance to do a proper drag-solder rework on the ribbon connector, but the TFT remains erratic. I tried removing some of the black material on the connection between the LCD and the ribbon connector to see if I could troubleshoot the connection between the ribbon and the LCD, but the pitch is way too fine at that point. I'm going to try replacing the TFT wholesale with one I'm buying off ebay.

Zuhd
 
Posts: 11
Joined: Wed May 10, 2017 6:22 am

Re: 1.8" TFT shield display problem

by Zuhd on Sun Jun 11, 2017 8:42 pm

Replacing the TFT display had no effect. Looking at the layout diagram on the shop page for the TFT shield, i went through and determined that there were no continuity problems anywhere on the board. I soldered jumpers to the relevant leads of the hex inverter to rule out anything between the headers and the hex inverter, and was still unable to get this to work. I don't have a logic analyser so I can't determine if the hex inverter is bad, but I suspect it might be that. I'm going to have to replace the board, unfortunately.

Incidentally, is this not the place to receive support for adafruit products?

Zuhd
 
Posts: 11
Joined: Wed May 10, 2017 6:22 am

Re: 1.8" TFT shield display problem

by franklin97355 on Sun Jun 11, 2017 11:02 pm

Yes it is but if you reply to your own thread it falls off the unanswered list and sometimes gets missed. I've passed this on to the techs that can answer your questions. Could you post pictures of your connections and the soldering you've done?

franklin97355
 
Posts: 19030
Joined: Mon Apr 21, 2008 2:33 pm
Location: Lacomb, OR.

Re: 1.8" TFT shield display problem

by Zuhd on Mon Jun 12, 2017 12:02 am

Sure. Here's an imgur album with some pics of the shield in its current condition (I removed the jumpers from the hex inverter pins as connecting the arduino directly that way didn't make a difference):
http://imgur.com/a/iTgw5

A little bit of the pad for pin 14 on the hex inverter lifted when I was cleaning off the flux to take these pics, but that wasn't the problem as it was in place the whole time prior.

Zuhd
 
Posts: 11
Joined: Wed May 10, 2017 6:22 am

Re: 1.8" TFT shield display problem

by franklin97355 on Mon Jun 12, 2017 2:08 am

I'm not sure what your problem could have been but you seem to have tried the things I could have recommended. Sorry you have had so much trouble.

franklin97355
 
Posts: 19030
Joined: Mon Apr 21, 2008 2:33 pm
Location: Lacomb, OR.

Re: 1.8" TFT shield display problem

by Zuhd on Mon Jun 12, 2017 2:11 am

franklin97355 wrote:I'm not sure what your problem could have been but you seem to have tried the things I could have recommended. Sorry you have had so much trouble.


Appreciate the commiseration, it's had me completely baffled. At least I learned some things along the way. I'm hoping I'll have better luck with a replacement board.

Zuhd
 
Posts: 11
Joined: Wed May 10, 2017 6:22 am

Re: 1.8" TFT shield display problem

by adafruit2 on Mon Jun 12, 2017 1:24 pm

its a bit unclear what is going but yeah lets just replace it - email support@adafruit for a new shield

adafruit2
Site Admin
 
Posts: 17162
Joined: Fri Mar 11, 2005 7:36 pm

Re: 1.8" TFT shield display problem

by Zuhd on Mon Jun 12, 2017 4:00 pm

adafruit2 wrote:its a bit unclear what is going but yeah lets just replace it - email support@adafruit for a new shield


Thanks, I really appreciate this- I actually didn't realise that adafruit stood behind their products like this and by "replace the board" I meant that I was going to buy another from a retailer. I will be sticking with adafruit products for this project and future projects.

Zuhd
 
Posts: 11
Joined: Wed May 10, 2017 6:22 am

Re: 1.8" TFT shield display problem

by Zuhd on Fri Jul 07, 2017 6:15 pm

I'm back, unfortunately. I rec'd the replacement shield in good order and verified that the graphicstest sketch was working properly after soldering headers to the shield. I then modified the shield to fit the project stack (it sits in a stack on a protoshield with RTD & wifi breakouts, and 2x whitebox tentacle shields, which are then wired thru an IDE cable to a mega)- routing pins 11, 10, and 4 to 3, 2, and A1 respectively. I then rewired the shield to my testing arduino UNO and tried the graphicstest sketch again- all in order.

At this point I installed the shield on top of the project stack and left it there while working on some unrelated code (running a couple of water pumps). I got around to working on the display code again today when I discovered that the shield had failed in exactly the same manner as the last one- white screen, I can intermittently get the graphicstest sketch to display but it's very difficult (resetting 100s of times etc) and only lasts 10 seconds or so before going white again. The only things that happened between the shield being in working order and being in this same failed state were about an hours worth of pump-running, with no code related to the TFT shield, and then an attempt to use the Arduino TFT library with the shield (failed, at which point I became suspicious and removed the shield to the test UNO and discovered that it had the same problem as the last one).

I can only think of two possibilities here- the display somehow got messed up when I was working with the pumps (seems unlikely, I would figure the SD card reader etc would be affected if it was something related to EMF or similar), or the Arduino TFT library somehow messed up the display. The latter seems vaguely plausible because I recall attempting to use this library right before running into the original problem, although I can't figure out how that might happen.

So- have you guys ever had a problem with the Arduino TFT library on this shield? Is it possible that using it could damage the display in a way that would prevent the adafruit library from working afterwards? This seems like a crazy question, but I can't figure out what the hell else would have caused both shields to fail in an identical fashion.

Zuhd
 
Posts: 11
Joined: Wed May 10, 2017 6:22 am

Re: 1.8" TFT shield display problem

by adafruit2 on Fri Jul 07, 2017 10:28 pm

hmmm - no this is the only thread we're on with this issue. just to check, you've tried a full powerdown, disconnect all extra parts, and restart...?

adafruit2
Site Admin
 
Posts: 17162
Joined: Fri Mar 11, 2005 7:36 pm

Re: 1.8" TFT shield display problem

by Zuhd on Sat Jul 08, 2017 3:48 am

Yeah, the TFT shield is being tested seperately from the project on a dedicated UNO with nothing else attached. I have powered it up/down and restarted many times. I made soldered connections with wrapping wire between the relevant pins (8,9,10,11,13) on the UNO and the TFT shield to make sure the issue is not a loose connection somewhere. It seems wildly unlikely that two shields would have failed in exactly the same way unless it was something that had happened to them both, but literally all I did with this one was have it on a powered project for a little while while I did other things, then make one attempt to use the Arduino TFT library with it.

I'm kind of at my wit's end here, this is the weirdest electronics issue I've ever encountered. If you folks have any more troubleshooting suggestions I'll gladly take them, otherwise I guess I just have to shrug and move on. I appreciate all the help one way or the other.

Zuhd
 
Posts: 11
Joined: Wed May 10, 2017 6:22 am

Re: 1.8" TFT shield display problem

by Zuhd on Sat Jul 08, 2017 6:45 pm

Finally, finally figured this out- at least well enough to fix it (I still don't know exactly what is causing the screen failure). I should have realised that the white screen was related to TFT reset, but I never intended to use the reset button for my project so I never bothered investigating it. By way of apology for taking a replacement when I didn't properly need one (though it did help me do a little "destructive testing" in finding the solution), please accept the following video documenting the problem and its solution:

https://www.youtube.com/watch?v=B2x7X4yje7g

Zuhd
 
Posts: 11
Joined: Wed May 10, 2017 6:22 am

Re: 1.8" TFT shield display problem

by Zuhd on Sun Jul 09, 2017 1:16 pm

As a last note, I was speaking to an EE friend of mine about this (I have no background in the field, can you tell?) and he suggested that there's no mechanical problem with the shield causing the reset, but rather it's just a function of the reset pin floating that may have high resistance to ground- for whatever reason moving the shield around is resulting in voltage on the floating pin, and being in a shield stack that's not got a connected reset pin is a similar situation. This seems more likely to me than there being a mechanical defect with the shield.

Zuhd
 
Posts: 11
Joined: Wed May 10, 2017 6:22 am

Please be positive and constructive with your questions and comments.