10.1" display and a dead Arduino Due.

This may be the wrong place for this. Perhaps Displays section would be better but instead of showing my profile I am being asked to sign in (no response on clicking) and cannot start new topic. I am new to this forum. Anyone's advice on this would be appreciated.

Anyway here is my first post.

For the last two years I have used a 3.5" display in an Arduino Due based GPS/depth plotter which I use on the sea near my house.

The 3.5" display is rather slow and limited in size so I decide to try a 10.1" display with custom shield from BuyDisplay in China. Having done this I thought I would share my observations.

The display arrived promptly, in good condition and with no import duties levied.

After assembling the display/shield and Due I connected to the programming port. The board programmed OK but the display was not recognised. The documentation suggests the use of an external 5V supply for the shield so after removing the USB lead I plugged in a 5V supply and the display came to life running the text demo program which I had previously downloaded. I then plugged in the USB lead. Shortly after that all went dead.

The Due was no longer recognised by the PC. It turned out that the 5V rail was shorted out by a dead voltage regulator. I removed the regulator but did not bother to replace it as it is only relevant when using the 9V input socket. It appear that, because there is a 5V connection between the shield and the Due, there was a fight over control of the 5V rail. I conclude that the 5V pin from the shield should be removed.

Now running with 5V to the shield and USB connection as normal I was able to run the examples. It appears that the examples were written for a different product and not all functions would run. In particular the fonts are very limited. There is space for a font chip on the display board but I do not know whether another vacant space near it also needs populating.

The inbuilt fonts are fixed width, very blocky and with little choice of size but just about useable for my purposes.

Constructing a triangular direction pointer was easy and the speed of writing on the screen was good enough for it to be unnecessary to clear local areas of the screen before re-writing. Clearing and re-writing the whole screen was acceptable for my purposes. The examples include display of small images but these were of unusual format. JPG files were not accepted. The video processor can handle some interesting effects but this is of of use to myself so I did not experiment with this.

There is an SD card reader on the shield but it is only accessible by removing the display. As I read the card after each trip that was a nuisance. I tried an external reader but those I tried (standard eBay blue board and Adafruit board) had permanently enabled MISO which clashed with the the display. I tried using an HC244 buffer (enabled by CS) on the MISO line. This worked but caused the display to progressively slow down. I assume this was caused by errors leading to a slow down of the SPI clock.

A Lanmu SD card extension worked but also caused slowing down of the display. On the extension is a 330R pull-up resistor on the clock line. After removing the resistor the extension became useable.

The plotter uses 12V DC from the boat so I used two 7805 5V linear regulators. One for the shield and one for the Due. They each have finned heatsinks and their inputs are each via a 1N4001 diode to separate the shield supply from the Due during programming. When programming, the shield is powered by a separate 5V supply while the Due gets it's 5V from the USB port as usual.

There was a problem with start-up and several attempts at re-powering were sometimes necessay. I believe that unreliable start-up is not unusual on the Due. My solution was to put a MAX309 voltage supervisor on the reset line. This holds down the reset line for a short while after the 3V3 supply has reached 2.9V.

On rare occasions there are errors on the screen with additional characters added to the text particularly when the function putFloat was used for showing decimal numbers. Converting the numbers to strings stopped the errors. I cannot see how this a library issue. I suspect errors on SPI bus. The problem persists even if the SD card extension is removed.

In conclusion, there are a few minor issues which can be worked around but if a large display is required then this one works well and is good value.

I hope that this information will be of some use to someone. I welcome any comments.

This topic was automatically closed 120 days after the last reply. New replies are no longer allowed.