new pcb ready

Finally the pcb with mirrored oled display is finished !

Meanwhile I did a lot of try outs with proof of concepts. As you can see one of them was the oled display. But I will treat a few proof of concepts in the next blogs.

TRIOS ROCKS !

development kit ready

It's just a pitty that the connector of the display is mirrored. That's the only thing that went wrong in the process. But we corrected this awful bug and expect a new PCB ready for end of januari 2009.

As you can see the board is totally supplied by the USB cable.

Meanwhile I can continue on this board with further proof of concepts needed for the project.

Things todo:

1. Writing a serial communication driver for the master
2. Writing a timer (PWM) based module for the triac
3. Introduce a button to change the triac state
4. Set up a parameter structure
5. Analyse the possibility to filter the CAB signal by software
6. Expand the one line protocol to the slave state machine
7. Convert the assembler OLED init code to a C module
8. Writing a UI on the OLED
9. Several tests on each of the above

Plenty things still to do as you can see but I love to dive into of course. 

Stay tuned for more news...

Development kit drawings

Schematics

PCB layout

Components:

1 X ATMEGA644

2 X ATTINY861

1 X 36p display connector 

1 X OLED display

1 X FTDI232 USB2SERIAL

3 X ISP connector

1 X Crystal oscillator 10MHz

1 X Zero crossing input connector

2 X Triac gate output connector

1 X Joystock mini

1 X DC convertor LT1303

1 X Supply stabiliser LD1086 3.3

4 X Led

The PCB and components are ready and ordered. Probably in about 2 weeks the kit is ready for use. Meanly the kit will be used to experiment and prepare the final project. It contains already 2 slaves and 1 master. The final project will have 30 slaves and one master based on the used components on this development kit. So a pretty long way to go. But my intention is that the final project must be ready for 1 july 2009. Stay tuned on this blog where you can follow up the progress and test results of the building blocks.

One wire protocol an exercise

Below the test setup with my new digimess 25Mhz scoop:

After exploring the the basic functionality I wrote my first program. It is an implementation of the Dallas one wire protocol without the possibility to read from the slave. On the picture below you can see a scoop capture. I will need the protocol in my final project so this exercise was worth to do. This protocol is defined in one of my publications and can be found on this link:

http://sites.google.com/site/guyvo67/

(LOWBP.PDF)

We also changed the type of IC for master into ATMEGA644 because this one can be mounted on a DIL feet and is easy to replace instead of using a SMD like one. The development kit drawings are almost finished so I expect it ready for half of December. In the next blog I present the development kit layout.

Exploring the world of AVR.

Exploring the world of AVR.

This weekend I tested all the functionalities of the AVR atmega8. On the pictures you can see my working environment.The intention was to learn the basics sofware tools like AVR studio 4 and hardware itself. The advantage is that when you learn thison a atmega8 you can easily port your knowledge to the complete series of AVRs.

This weekend I learned:

- INT0/INT1 edge trigger interrupt
- 8 bit timer0 basic skills with overflow interrupt
- 16 bit timer1 with CTC match
- 16 bit timer1 and fast PWM
- 16 bit timer1 PWM phase/frequency correct mode
- ICP mode on pin capturing and counting frequency
- USART streaming from avr GCC
- linking with floating point libs for printf
- sleeping modes

So pretty much everything I guess , I left timer2 because it has the same operational modes.

This was meant to be as a full preparation to the real project. Now it's much easier to give my input for the hardware design of the dev kit.The development kit will be ready at the end of December. Meanwhile I have this great little board already to experiment.

Special thanks to Ronny for that !

TRIOS overview

TRIOS high level architectural overview on whiteboard:

Three big PCBs will be made each containing: - 10 triacs 8A max - 10 AVRtiny low power control MCUs - optionally master controller with lcd Which gives a total of 30 ouputs fully dimmer controlled. One PCB will have a master unit companied with a little OLED lcd module to parametrize the slave modules. The master and the slaves will be connected by a one wire bus. At this stage it's to early to reveal to much details as we are just beginning to explore what target devices could best do the job. However a first proposition is already chosen: Slave : ATTiny861 Master : ATMEGA2560

The hardware will be done by my friend and can be contacted at manes6969atgmail.com
I will write the software using the AVRstudio 4 with gcc plugin.

The first thing we work on now is a development kit that makes live a bit easier and make things more concrete.

Stay tuned there will be much more in the coming weeks.

First prototype

First prototype dimmer based on 8A triac and pulse controlled by an AVRtiny based on zero point detection. Control pcb is totally separated from the 220V using optoelectronics.