6 years, 8 months ago.

white canvas - no graphics - no text - ILI9341 - LPC1768

Hello Peter, gents - I have carefully rebuilt your setup using a TFT ILI9341 and mbed LPC1768 - the code compiles but canvas remains white. I have uploaded my schematics - checked wiring carefully - and also tried to compile and run fractions of the main.cpp in order to view at least a string (result is NOK) - there must be something trivial I am overlooking ... I am using the basic code without use of the extra SD Card ...

Would appreciate your feedback ! Best regards, Michel /media/uploads/gazelle/ili9341.docx

Question relating to:

test vor the new lcd with a ILI9341 controller

For the seeeds 2.8 touch screen .. don't forget to pot 100Kohm betwenn reset pin off the tft and power

posted by dujardin stephane 26 Nov 2015

4 Answers

6 years, 8 months ago.

Have you tried to switch miso and mosi ?

Accepted Answer


Hello Gentlemen, I added a debug LED1 function and observe that LPC1768 is executing its code ... - wiring has been checked again, is according schematics in MS WORD doc and Vusb (p39) = 4V8. I have attached update of document - SRC code mbed and binary for LPC1768 - it contains schematics - wiring scheme - data capture with protocol analyzer

I appreciate your feedback !

Thank You Michel Heuts

posted by Michel Heuts 01 Jul 2014


posted by Michel Heuts 02 Jul 2014
6 years, 8 months ago.

That is the default initialization of those displays. I would expect really a connection is either missing, or a pinout wrong. You can try the Read_ID function, assuming you have connected the MISO line too.

Hello Erik - I have checked my wiring over and over - it is documented in the Word Document I have posted above. The only other option is that the TFT display is broken - thanks for assisting ! Thanks - Michel

posted by Michel Heuts 30 Jun 2014
6 years, 8 months ago.


While I'm using the FRDM-K64F platform I have been unable to get anything displayed on my screen. I know its working as I've been using it on an Arduino board using the AdaFruit library.

I noticed looking at the code that it was only sending 8bits at a time for the FRDM-K25 platform so made it do this for the K64F as well but this didn't help.

Any feedback on progress on the original query would be helpful.

6 years, 8 months ago.


I had a quick look at your document and have the following suggestions: - force the lib to use the 8 bit mode for SPI as initial test. - use spi.format(8,0); that is the required mode for the ILS9341 - make sure your logic analyser is also decoding for SPI(8,0). According to the screenshot is it currently not setup for that mode. This means it will not properly display the values you transmit -make sure the logic analyser is set at high enough sample rate. Otherwise you will miss clocks or the timing seems to be off since every transition will be at multiples of your samplerate. - with these settings check the CS again. In your screenshots it looks like CS is switching between high and low at irregular moments (ie in the middle of a communication). That will reset all SPI communications and cause a failure. Is the wiring stable, is some other pin or signal shorting or interfering with CS?

I have Peter's code working - great news - all graphics visible on canvas !

The wiring and hardware were OK from start - the project always compiled correctly but this time I forced the mbed cloud compiler to update the mbed libraries during import of Peter's program and that made the fix ! There was no need to change SPI format in 8-bit mode - Peter's code works fine on LPC1768. So most likely there is some inconsistency with (legacy) libraries.

This is good news to start the weekend with ...

posted by Michel Heuts 03 Jul 2014

Which version were you using? I can't get mine to work for some reason, and I'm reasonably sure its a software problem.

posted by Caleb Washburn 16 Jun 2015