Still some problems

Yesterday we found out that the I2C expanders (PCF8574) could not source the LEDs of the LED matrix. We did some measurements and apparently the output ports of the PCF8574 IC's gave only 70µA as output current which is to low of the LEDs to light up. The LEDs need 20mA. As a solution we drew another PCB to place on top of our original PCB. This new PCB has an LED driver circuit with a transistor for each LED. Below you can see the new PCB.

We also have a problem with our program for the arduino, it seems that using the NFC card reader with SPI and the I/O expanders with I2C work perfect separately but if we combine the 2 programs, only the SPI works and the I2C doesn't work anymore then. We are still trying to find what causes this problem, maybe the arduino is not able to use SPI and I2C at the same time or maybe it's something else, we don't know but we hope we'll find it soon.

finally some progress

Our PCB is made by the LPKF machine, this week we are going to solder all the components on the PCB. The I2C finally works, I found the error, I used B01000000 (8 bit) as address for the I2C but apparently arduino uses the 7 bit address so B0100000 and adds a '0' or '1' (for write or read) by itself. I also noticed that I made a little mistake in the last schematic of the test setup (the LED was reversed). I changed it in the schematic, below you can see the right one.

here you can see the program code I used for this schematic:

Today we are going to finish the pcb gerber files so that our print is made and we can finally start with soldering. Our program in arduino is still not 100% this is probably within the i²c part of the code. To test our program we are using a test setup with 1 i²c, 2 pullup resistors of 4K7 and a led. The schematics of the test setup you can see in a former blog.

test setup

Today we tried the test setup below to see if our I2C communication in collaboration with the NFC reader works. We didn't succeed yet but we are going to try this again the next time.

We are expecting our components and PCB somewhere next week. This week we did some tests of our project, they where somewhat succesfull. We are hoping to start building our PCB next week so we can start programming the arduino in combination with our PCB.

Printed Ciruit Board design

Below you can see the result of the PCB design of our project:

Here you can see the basic code we are using for reading the NFC-card:

Bi-color LED matrix

Because we could not find a RGB LED matrix as I mentioned in my previous message we're now going to use two 5x7 bi-color LED matrices. We chose the TBC20-11EGWA , this is a 5x7 LED matrix with the colors red and green (with yellow/orange as the third color, this is the color you get when both the red and green led are emitting light).

Here you can see the schematic of the 5x7 bi-color LED matrix:

Here you can see the complete schematic of the PCB we are going to make:

As you can see there are some other changes besides another LED matrix, we decided to use three 8-bit I/O expanders (PCF8574) instead of the shiftregisters. These are I/O expanders you can control with I2C. We already used I2C with the arduino uno so that makes it easier for us to write the program later.

Below you can see the schematic for driving one segment, the other one is almost the same except for the hardware addresses of the remote 8-bit I/O expander PCF8574.