Notice. New forum software under development. It's going to miss a few functions and look a bit ugly for a while, but I'm working on it full time now as the old forum was too unstable. Couple days, all good. If you notice any issues, please contact me.
I think most of the tracks on the bottom side of the LCD are to do with the touch sensor and SD card.
Me thinks I will try to peel off the TFT from the board.
Whats the worst that can happen, being a brown touch day?
Lou
WhiteWizzard Guru Joined: 05/04/2013 Location: United KingdomPosts: 2944
Posted: 08:41am 22 Feb 2018
Copy link to clipboard
Print this post
WARNING: Peeling off the TFT will likely crack the glass (depends on strength of sticky tape holding it down!).
Best way I have found is a 'wallpaper stripper' with a sharp edge (or a large - but thin - kitchen knife). You need to slide the blade totally under the TFT BEFORE applying any 'lifting' force otherwise it will crack!
Good luck . . .
WW
Grogster Admin Group Joined: 31/12/2012 Location: New ZealandPosts: 9610
Posted: 08:58am 22 Feb 2018
Copy link to clipboard
Print this post
Those photos look like the LED backlight is running full-tilt at maximum. This would produce what you are seeing, if the backlight is running flat-out, as it will overpower the light output of the LCD itself. That would also cuase the SMD transistor to heat up me thinks. Very interested to see what happens when you put 5v on the LED line via a resistor.Smoke makes things work. When the smoke gets out, it stops!
OK, managed to pull the LED line clear and solder a 330R resistor between vcc and LED. Same result as PWM 5 ish. Parallelling a 100R resistor across this produces same result as PWM 10 ish.
Also, I lifted the TFT but other than a flat flex that have some passives on it, there is not much there. Soldering looks OK under magnifying glass.
I think it is safe to assume that LCD module is not well.
Think I mnight remove the LCD before flashing the backpack next time. Still don't see how this would have caused the issue, but like I said earlier, I don't believe in coincidence.
Many thanks to all for your assistance. Warms the cockles of my heart that there are people willing to assist those in need.
Lou
Azure Guru Joined: 09/11/2017 Location: AustraliaPosts: 446
Posted: 10:20am 22 Feb 2018
Copy link to clipboard
Print this post
It looks like there is no color in those photos, just greyscale, could just be the pictures.
Can you try these commands on a command line: [code] CLS RGB(RED) CLS RGB(GREEN) CLS RBG(BLUE) [/code]
Do you see the colours filling the whole screen?
I tried using PWM 2, 250, xx with xx at 1, 5, 10, 20, 50 and 100. All showed clear colours at the appropriate intensity with both the gui test and the cls with a colour.
In case you are not aware how these LCD panels work; the image you see is a result of the LCD panel filtering the light behind it (the backlight). The LCD panel itself does not produce any light it just filters the light passing through it using varying levels (0-255) of filtered red, green and blue at each pixel location.
If you look at the side of the screen you will be able to see the backlight when it is being driven.
I can see colours OK. Just very faint. Thanks for the info re LCD. Is there some common point for how the LCD is driven as I believe it is not letting the colours through very well. Might be the single point of failure I am looking for.
Regards,
Lou
Azure Guru Joined: 09/11/2017 Location: AustraliaPosts: 446
Posted: 11:47am 22 Feb 2018
Copy link to clipboard
Print this post
Gets very complicated. The ILI9341 has 320 x 720 column and row drivers (1 for each RGB in row 240 direction), these column and row drivers have voltages generated to ge the required levels for the LCD material and this is all on the controller chip (over 1,100 pins) that is mount on the LCD panel.
if you google just the chip part number you will find pdfs with the chip spec if you are interested.
I think from all you have said so far, either the LCD is damaged or the chip is faulty. Basically a throw away item.
I have flashed my Backpack V2 quite a number of times always with the LCD connected and it has not failed... (yet).
Grogster Admin Group Joined: 31/12/2012 Location: New ZealandPosts: 9610
Posted: 10:52pm 22 Feb 2018
Copy link to clipboard
Print this post
With the LCD going on the test pattern, press the LCD to the carrier board firmly and see if things come right. Use your middle and pointer fingers on both hands, to push all four corners of the LCD down firmly at the same time. You can also try pressing it in the middle.
Firmly, but gently. Do NOT get carried away with force, or you can break the glass. I did have one LCD which 'Delaminated' from the backlight, and I ended up having to throw it out, cos one end looked good, but the other side faded away to nothing. Pressing the LCD at the faded end to put it back where it should be restored the image.
I tried using wide sellotape to hold the LCD down, and that would work for a day or so, but then it would still manage to lift itself up again!
I binned that display. They are cheap, so at least this is not happening to you on a 7" SSD1963 parallel LCD display - they are quite a bit more expensive.Smoke makes things work. When the smoke gets out, it stops!
Thanks Grogster. Pushing and prodding has not helped. Tried to read through the spec for the chip and whilst there are many configs settings available, I am pretty sure most things get hard reset on power up. Was hoping to find some NV RAM config that may have been messed up when I flashed the backpack.
Lou
Azure Guru Joined: 09/11/2017 Location: AustraliaPosts: 446
Posted: 12:08am 23 Feb 2018
Copy link to clipboard
Print this post
Flashing the MM Firmware will not effect the LCD Panel.
I think we are down to a replacement LCD as the cost of the unit is not worth repairing.
Still curious as to what the actual faulty component(s) is and what the magic smoke smell came from, but then we may never know.
Let us know your progress when your new displays arrive. I have found SC to be fairly quick at shipping if items were in stock.
Grogster Admin Group Joined: 31/12/2012 Location: New ZealandPosts: 9610
Posted: 12:10am 23 Feb 2018
Copy link to clipboard
Print this post
Damn....
Well, chums, I think we have just about run out of options have we not?
We can assume that this LCD is dead, but the fact it died on you during a firmware update is MOST odd, and all I can think of to say at this point, is that it was going to fail on you anyway, but it just happened to do it while you were reflashing the firmware. EXTREMELY unlikely, and I don't like unsolved mysteries, but that is about all I can think of now.
The LCD itself obviously IS working, because you can see it - albeit with difficultly. The backlight is a very simple device, so by twiddling with it, and putting in a series resistor as you have also now done, should tame it.
Bloody strange.....
I don't suppose you would be willing to send me this dead one to play with would you? I would compensate you for the postage. Smoke makes things work. When the smoke gets out, it stops!
Azure Guru Joined: 09/11/2017 Location: AustraliaPosts: 446
Posted: 12:19am 23 Feb 2018
Copy link to clipboard
Print this post
@Grogster
I was thinking the same thing. It would be good to get a look at this display to figure out what has happened.
Grogster Admin Group Joined: 31/12/2012 Location: New ZealandPosts: 9610
Posted: 12:24am 23 Feb 2018
Copy link to clipboard
Print this post
By all means, you can do it if you like.
It would be cheaper for loup to send to you, rather then Internationally to me. I just wanted to find out what the hell was going on with it! Smoke makes things work. When the smoke gets out, it stops!
Azure Guru Joined: 09/11/2017 Location: AustraliaPosts: 446
Posted: 12:27am 23 Feb 2018
Copy link to clipboard
Print this post
I have sent him a PM. I'll update the post with what I find if loup is willing to send it to me.
Grogster Admin Group Joined: 31/12/2012 Location: New ZealandPosts: 9610
Posted: 12:37am 23 Feb 2018
Copy link to clipboard
Print this post
Cool. Please do that! Smoke makes things work. When the smoke gets out, it stops!
Azure Guru Joined: 09/11/2017 Location: AustraliaPosts: 446
Posted: 01:18am 23 Feb 2018
Copy link to clipboard
Print this post
Also happy to send it to you if I can't find the problem and there is still something worth sending
WhiteWizzard Guru Joined: 05/04/2013 Location: United KingdomPosts: 2944
Posted: 06:34am 23 Feb 2018
Copy link to clipboard
Print this post
@Azure, @Grogster
If possible, can you replace the tiny Transistor above the Touch controller chip. You may have other 'broken' TFT modules there with you from which you can remove a known working transistor.
Azure Guru Joined: 09/11/2017 Location: AustraliaPosts: 446
Posted: 06:52am 23 Feb 2018
Copy link to clipboard
Print this post
@WW
Will do.
I have a plethora (well a good selection anyway) of SMD transistors and E24 range of SMD resistors and caps from 0402 to 1206 sizes plus odd special values and sizes. So I should be able to find suitable replacement parts to try if I can get my hands on it.
Quazee137 Guru Joined: 07/08/2016 Location: United StatesPosts: 593
Posted: 07:04am 23 Feb 2018
Copy link to clipboard
Print this post
I have 5 of the old backpacks and 5 of the new ones as well as 4 of Mick's, 20 of my boards with mix of both type of displays and all have been flashed with the LCD in place no problems. Even two on bread boards.
Just a thought check the voltages on the bridge. Edited by Quazee137 2018-02-24
WhiteWizzard Guru Joined: 05/04/2013 Location: United KingdomPosts: 2944
Posted: 07:23am 23 Feb 2018
Copy link to clipboard
Print this post
Best to simply swap for a known working transistor (and exact type) from an identical (layout) TFT module - i.e. one where the screen its cracked but the components are ok.
This is unless someone knows that exact Part No. My worry is that using an 'almost similar' type may actually cause another issue wth other components on the TFT then working outside tolerance.
Page 3 of 4
Print this page
The Back Shed's forum code is written, and hosted, in Australia.