Plume – The receiver PCB is about to be printed !

Finally !

After placing, routing, finding errors, fixing them, finding other ones, fixing them as well, and other last minute changes…, we finally finished designing our PCB. With Guénolé, we spent the whole week working on it so we are happy to have finished it ! You can admire the layout in the pictures below 😉

PCB layout, top traces are in blue, bottom ones in red

PCB layout, top traces are in blue, bottom ones in red

Top layer components

Top layer components

Bottom layer components

Bottom layer components

Aside from that, I also worked on the software architecture to update it and make it work with the new hardware architecture we chose.

After talking with the teachers, we have decided to eliminate the various symmetry problems we had by studying the phase of the signals we sense. By doing this, we can measure the three magnetic fields exactly instead of having 8 possibilities on each measure (±Bx,±By,±Bz). And there is only one symmetry left that we won’t be able to suppress : the three magnetic fields have a central symmetry so they are exactly the same in (x, y, z) and in (-x, -y, -z). Therefore, with this new method, I need to change the software architecture to have a good phase detection, and I think I found a good way of doing it !

And by the way, if you have nothing to do tonight at 9:30, you can come to my improvisation show : !

See you next time !

RoseOnRails – PCBs


Today was mainly lab session work on the STM32 board. I finally understood the concept of Chibi shell!! I even added some “personnalized” commands to expand the shell 🙂 Funny!

Then it was the “beep” step: run the buzzer of the STM32. It was rather straight forward now that we have well understood the concept of PWM, duty cycle, and so on.

After that, I started to think about how to use lwIP to carry out an HTTP request through the web. Though I haven’t had enough time to go too far in it yet…

This morning I also “played” with my BLE Evaluation Kit. I managed to flash some example programs and test them.

Tonight, Gleison and I started the conception of the PCB of the locomotive. But as Blyste mentioned, we’re still missign a lot of things. So, we rather started by creating issues on Bitbucket, mentioning the components we need for our circuit… impatient to have them and start this famous PCB conception at last 😉 !!!

Next step for Drops N’ Roses : PCB !


In contemplation of the PCB we’ll have to make next week, I decide to prepare all components we will need on PCB Expedition. I have to choose the package for our main components, but it is quite difficult because I do not know what constraints we have neither where we will weld our components on our PCB. I just know that UBGA (Ultra-thin Ball Grid Array) is not the easiest way, and so we have to avoid it. That is why I choose :
– for our processor : nRF51822-QFAA.
– for our memory AT45DB321E : I can’t choose between 8S2 (8-lead, Plastic Gull Wing Small Outline), 8MA1 (8-pad (5 x 6 x 0.6mm body) Thermally Enhanced Plastic Ultra Thin Dual Flat No-lead) and 8MW1 (8-contact, 6 x 8mm, Very Thin Dual Flat No Lead Package).

Then, I check with Matthieu that some other components (LED, resistors, …) were on PCB Expedition, and if not, I create an issue on bitbucket to signal it to Alexis. He will then prepare for us the components on that software.

At last, I discuss with Matthieu and Lauriane about BLE Architecture. But Lauriane will sum up this for you.

That is all for our project. I remind you that you can follow our progress here.


About STM32 lab, I helped other students debuging their program. But I did not change my own program.
Later on !

PLUME – Making some PCB and playing with coils


We finally have an emitter coil. 🙂 It is not finished yet but we are already used it to make tests. (Pictures are coming as fast as possible).

We try to see the intensity of the fields generated by our system but it is very low at the moment. It remains for us to be able to generate the signal that we will use with the coil. Indeed a “classic” frequency generator is current limited and this poses us problem. We will use an H-bridge to power the coil during our tests.

On the other hand, we continue our inquiries on the shape of the coil and the number of laps that we need. Indeed, despite our 200 turns of wire, our system is less efficient than the Razer Hydra.

Moreover, these last two days, I worked a litlle bit on my lab work with my STM32. Now, I am able to use the potentiometer and the adc to have some fun with the led intensity. Actually, I’m working to implement a serial port to USB. My purpose is to use the correct driver of ChibiOS to  configure the USB port used to power the card so that it is seen as a serial port. I hope to finish that as soon as possible.

 Finally, today I re-started to use Expedition PCB. I missed it a bit. 🙂 I look forward to be able to map our project because I really like the idea that the placement / routing of components that does not happen randomly but must comply with all rules of signal integrity and electronics.
To be continued…
Capture IMG_20140307_212525