Admitting defeat... V1 project, maybe dead... Maybe...

A project log for Digital Speedometer

A new digital dash to replace the slow inaccurate one on my cheap Chinese bike

Mangus TiranusMangus Tiranus 12/08/2017 at 12:470 Comments

Well this is sad. I really didnt want to post this but the last few weeks I've been really struggling and contemplating doing this... 

I have to admit defeat. This version of the project is now dead. Which is a painful shame given all the hard work ive put into it.

So whats happened?

The custom FTDI EVE display i designed... I just cant get it to work. I've spent the last few months going over it, I went through 3 different PCB designs on paper it should work but it just does'nt. Part of the problem is the LCD panel itself.

It uses an ILI9488 chip, and datasheets for the real nitty gritty how to get this to work is scarce. For starters, in the mian datasheet it says to get teh display running in 18bit parallel mode you only need to set the IM0,1,2 pins to 0, I've tested this both as floating/unconnected or low to ground. Neither automatically sets this to the correct mode.

With James Bowman from Gameduino we worked out that the datasheet is actually wrong, and to get the display to work You have to first set it to serial mode, then send a bunch of undocumented commands to the chip to flip it into the correct mode. This is not written in ANY datasheet. its purely worked out from random bits of code found online from various Chinese sites.

I've had a huge amount of help from James  trying to get this to work. hes even written a whole article on how to get ILI9488 displays to work on the FTDI EVE, you can read it here:

And we nearly got it working, if it wasnt for James I would have posted a post like this weeks ago...

So heres whats happening with the display...

Considering all the resoldering and 5V logic ive been putting this through, its possible Ive now damaged one or two or more parts and it would never work correctly...

Also my bike stuff has changed, I'm now in the process of getting my full bike license and I have a new bike already, A Kawasaki ER6f... Given that I have a nice new bike, and this isnt working. I dont want to throw any more money at this project :/

I wasnt aware before about the bridgetek display James mentions in the article beforehand. I could buy one of them and that would solve this and I could get on with programming and finishing the project, but its more money than I'm willing to spend right now for this. ($49, roughly £35)

By all means if anyone is willing to donate me a bridgetek display, ill carry on the project with that :)

The main board works, the LEDs work... and looks great. The power requirements for them is a little high and so I might replace the vRegs with buck converters if I continued. But thats only when it runs at full white bright that they get HOT! i normally only run them at 50%

So where does this leave this project? No I am not killing the project as a whole. just V1, I had always envisioned a more advanced V2, only that originally it was going to look some what the same with analogue dials and a screen in the centre...

This would now likely be using a prebuilt 5 inch display from HOTMCU, so I wont have to worry about messing up its design.

One of V2's key features would be GPS tracking for theft recovery with no subscription charges, using a free mobile provider like Freedompop to handle the minuet bandwidth requirements. So you find your bike is missing, you text the bike a command like "Where are you?" And it fires up the gps and replies with a gps location. it would then ask if you want real time updates? which you reply yes to and you then get a link to your phone with map and regular updates being posted online. 

This could be linked to a google account or some other service. i've still yet to work out the details on that. i might even design an app for the phone that the bike connects to through the internet and provides the location info to. V2 is very much a product version of the digital speedo. Something I want to develop to sell.

So yeah, for now it seems, this project is kinda some what dead... Maybe CPR can save it... money could save it but with my recent costs I just cant afford to put any more into it.