This is for you David you little pri.k..
I have 2.4 inch KMRTM24024-SPI
It looks as if you have 2.8 inch KMRTM28028-SPI
I am utterly gobsmacked by your video. There are no resistors.
Please can you measure VCC at the breadboard. It looks as if your Duemilanove is powered by 5V from the USB cable.
And the Breadboard supplied via the 5V pin on the 3x2 ICSP header.
Seriously. If you short out the AMS1117-3.3 chip and then connect 5V, the AMS1117 will die.
If you power ILI9341 at 5V, the ILI9341 will have a short life.
If your display is operating with VCC=5V and 5V logic I am amazed.
Yes, semiconductor chips do not fail immediately. The AVR has absolute max voltage = 6.0V
I suspect that it will survive 7.0V but I certainly would not risk 9.0V (which is the equivalent of your 5V to ILI9341)
David.
dsyleixa:
thank you, that is awesome - for the next tryCould you show please which jumper has to be soldered?
dsyleixa:
There is no jumper and neither a current path leading away from Vcc nor an imprint about 5V.
Neither of my boards have the printed message. They both have a clearly marked solder-bridge J1 and regulator U1.
I tried to use your links from #4. Neither worked for me. I suspect that you will have a board number like KMRTM24024-SPI or TJCTM24024-SPI. Please quote your board number if you can't provide a link.
koktelici clearly has a "working" display.
Datasheets clearly specify maximum voltages.
You can do whatever you like at your own risk.
David.
the only numbers I can find on my boards are
QR4 5285S01 G2/5 TP28127
but anyway, no solder bridge, so no fix, and as it could not be fixed, I meanwhile got my money back from the seller (but I didn't have to resend the TFTs though 
)
Perhaps I'll once buy one from Adafruit.
Google found this customer review https://www.amazon.co.uk/Generic-Touch-Screen-Card-Socket/dp/B00K0F2QR0
Which is a Blue 2.8 inch Mcufriend shield. Nothing to do with the Red SPI displays in koktelici's photos or video.
Nor anything that would produce my reply in #5. My reply suggests that I had located your links ok.
Seriously, you should be able to identify the colour of a pcb. And I would expect some better information or numbers printed on a pcb.
Your Original Post #0 suggests that your screen works with M0 and Due. Which library? How did you wire it?
Shields are a lot safer than hand-wired modules. There is no confusion with the connections.
David.
Seriously, you should be able to identify the colour of a pcb. And I would expect some better information or numbers printed on a pcb.
Your Original Post #0 suggests that your screen works with M0 and Due. Which library? How did you wire it?
My shield is a red one, like this one (I already linked to that above!)
but it's beyond me why you start arguing.
Of course there is no issue about the wiring, it's finally dead-simple. I'm working with Adafruit libs (GFX and ILI9341).
I have no problem with my ILI9341 on my Due or my M0, it just didn't ever work with AVRs. To me the issue is resolved, as it appears to lack from 5V compatibility.
I am not arguing with you. Your original question was answered in #1, #5.
I was very surprised by koktelici's #14, #17, #20
As far as I know, the Red SPI boards require 3.3V logic and have no level shifting circuitry on the pcb.
As a general rule, semiconductor chips are powered by 3.3V (via AMS1117) and all logic should be at 3.3V levels. If a logic input is > VCC + 0.3V the on-chip substrate diodes start to conduct. Chip behaviour is undefined. But the substrate diodes should protect the chip against serious damage (if the diode current is not too high)
It looks as if koktelici has violated the datasheet by powering the chip with 5V. The substrate diodes will not conduct until 5.3V. The undefined behaviour will not happen. On the other hand the chip is being HIGHLY stressed. After all, the chip was designed for 2.8V in the first place.
Think about it. If the 2.8V chip was designed to tolerate 5V logic, the datasheet would say so. And commercial manufacturers would not need to mount level shifters on their displays.
David.
On power supply I've measured 4.2V but on data lines 4.4v .
I found that this is working by accident. I've connected data lines using 3.1v zener diodes and it worked.
Only to find out that i didn't connect ground of the whole array of zeners so it basicaly worked directly from arduino outputs.
Then i suspected if anything is working as instructed by seller and desoldered the short but it doesn't work that way. I know it shouldn't work like this but it does.
I've been tinkering with this for hours and hours and if it was to burn out I think it would be by now.
I think that this explains everything. Your ILI9341 is powered at 4.2V. Your Arduino Duemilanove is probably running at less than 5.0V too. Not quite enough to make the substrate diodes lock up the ILI9341.
Most USB cables are quite capable of supplying up to 500mA. I am surprised that the end voltage has dropped from 5.0V down to 4.2V
No, I would not expect the chip to die within hours. It might run for months at 4.2V
On the other hand, a new PC with new USB cable and new jumper cables on your breadboard might result in higher voltages at Arduino and ILI9341.
I do not care what risks you take with your own hardware. You avoid the expense of six resistors (or a level shifter chip).
I try to write accurate replies in this Forum. If a pcb contains level shifter circuitry, I am happy to advise its use with 5V Arduinos. If the pcb is designed for 3.3V logic, I will advise using 3.3V Zero, Due, STM32, Teensy3.x, ...
In your breadboard it would be very easy to insert series resistors but obviously with associated costs.
David.