Last night and preparation to our first presentation

Good morning, today is the day of our first presentation! This night we started again to work on the Kinect, and created our structure for the sensors.

Now we have even more PCB! Because our sensors on the cube have each one their own PCB. One sensor alone works well, but when we linked all of them together we can’t control them, we think it’s a hardware problem, it won’t be ready for today. For the Kinect we can detect the movement of a hand and change the luminosity of the cube according to the distance between the hand and the cube. Now let’ convince the public our project is worth living 🙂

Now it’s beagleBoard time

We finished the work on the mainboard (at least we hope so), the frames are managed with double buffering, and the images come by USB. We can display images that we have created or .gif on our cube. Now we have to make the detectors talk with our beagleBoard.

On Friday evening (or more likely: night) we did our test of sensors on our big cube to find the proper values of resistors, our best results are with 40Mohm. Since Yesterday a tinyAVR can communicate with the stm32 and give him his values.

We made game to play with our cube.

Here it comes

Lately we finished soldering the cube ( also drill a new time the six pcb) and prepare all the software to welcome it. We spent the last two days trying to light on properly our cube, we had trouble with the timings, but now it’s ready.

Our cube is now all soldered, pcb are placed and we can display on it!  To prove it here is a little video, be patient for the images. Like other screens the images of our cube doesn’t look like what we see.

The cube is coming

After blood, toil, tears and sweat we have our 12x12x12 cube! We’ll post the photographs tomorrow, with some animations we hope.
We will be able to show it into 11days, on our presentation!

The base of the cube

Yesterday we bought wood for the base of our cube, it’s now all painted, we are ready to drill it (just wait the last paint go dry). Yesterday Alexis soldered our PCB  and we tested them this morning, there are a few mistakes to correct. It’s quite hard to solder a PCB done at school…

Our software is ready to display on the cube, we just need the screen!

Two LEDs are flashing!

Today was a big day: we planted two LEDs in our big PCB and made them flash! It’s not that easy because to do that we have to control the stm32, make him communicate with the driver, configure the driver, give him the good values and resolve problems of ground (it’s easy to resolve but you have to think about checking them).We can also control the future supply of the floors.

Finally, with Thibaut, we tried to communicate with the gyroscope through I2C, but we discovered that we had some troubles with the streams… Let’s continue tomorrow!

Where I am in Roled

It’s been a long time since I haven’t post, but I mostly soldered LEDs together, and set an environment for our card to work, so it wasn’t really exciting for you.

For our cube, we soldered the 144 columns of 12 LEDs, and the 12 floors of 12 columns. Now we only wait for 5 big PCB to make our cube (we already have one 😀 ) and show you beautiful animations. We have the PCB of our mainbard, it’s all soldered, and for a few hours I can flash on our processor and even communicate through SPI! To see what I send through SPI I use a logic analyser (Saleae). Our cube will be sticked in a wood board and, the 6 big boards will be screwed on under the wood board.Tomorrow, thanks to our big PCB, I’ll (try to) light on a LED soldered in the PCB using the driver which communicates with our board in SPI.Have a good night!

Never been so close

I haven’t post since a long time but this doesn’t mean I haven’t work!

I spent the last days making PCB, and doing it again, and doing it again. Not because I did it bad, I don’t know it yet, but because each time we changed something… This time I think it’s over, so we may start to solder our LEDs this weekend.

The new changes are in majority about sensors. At a moment we thought we will link all our captors directly on the stm32, then we thought we were going to put a stm32 on each card (we have six big cards under our cube with the drivers), then we changed those stm32 to  AVR, and yesterday we chose to put tinyAVR directly on the cube, one AVR per detector all linked by a I2C bus. What’s more we added a Beagleboard. That let us with : one Beagleboard, one stm32 and 45 AVR.

Conclusion : we have a Beagleboard which makes all the treatment, a STM32 in charge of the display and collecting datas from the captors: an accelerometer and 5×9 presence sensors (the tiny AVR).

ROLED’s Cube

Our project is to make an interactive cube of LED.

We have a cube of 12x12x12 LED (45cm x 45cm), commanded through a stm32 which is in charge of the display. We add in the cube some sensors, we will detect the presence of the hands nearby wires added in the cube and detect the rotation of the cube thanks to a gyroscope. The stm32 is also in charge of collecting information on the sensors. We can communicate with this stm32 thanks to USB.

We have a beagleboard at the base of the cube, linked with the stm32. Here we can create animations or games using the two types of sensors and display it on the cube. What’s more we will add a Kinect connected on the beagleboard, we can reproduce an object in the cube, we can play games such as Rubik’s cube in the cube of LED through the Kinect.

We have chosen USB because if we want we can connect the stm32 directly to a computer or a mobile phone, and create a program on it to play with the cube.

Let’s concentrate a bit on our LED. We will have 1728 RGB LED, so 6912 wires to connect. It seems quite a lot to connect on a PCB, so we will use multiplexing. Our LED are with a common anode, on each floor we link all anodes, and on each columns we link all “red cathode” together, all “blue cathode” together and all “green cathode” together. So we will light one by one each 12 floors, quickly enough to see the whole image.

Talking about our PCB, we don’t want 444 wires to connect on the PCB so we want a huge PCB as big as the base of our cube and plant our cube in it. On it we only will have the drivers and the links between the base of each column and the driver, so we will make this PCB in the school with only one side (in reality we will make six more little PCB side by side). And for the rest of the electronic we will have a little, normal, PCB.

Here is our architecture:


Days 28&29: LED are with us

We received our 2000 LEDs yesterday! They are really bright.
On Sunday I continued my research on the Theremin and the capacitance.
On Monday, we decided of the size of our cube. Between two LEDs we will have 4cm, so we’ll have a cube of 45cm. With this size it’s big enough so we can see all the LED and it’s small enough so we have a good resolution and we can play with it. We also did some tests on the LEDs. We have our wood to make our “mould”. I met Hugo, a member of the robotics and he pierced our wood. (Thank you!) So today we will be able to do our base to make our cube.