Rastello Club

A wireless glowing programmable juggling props, with 9-DOF IMU

Similar projects worth following
The aim is to make an open source connected glowing juggling club, equipped with 9 Axis (Gyro + Accelerometer + Compass) MEMS, hoping to create a small community .
I would like to create a solid basis and explore some part of the endless possibilities.
Many uses are possible, for many users profiles.

With this system, I would like to :
- Juggle by night.
- Program a sequence of colors on music.
- Interact with other devices. ( as an example : stage light, synthesizer... )
- Create music or sound, by connecting your club to Pure Data or Supercollider for example.
- Use it to create teaching tool to learn juggling.
- Create a detection algorithm of juggling pattern ( siteswap).
- Make fun juggling games
- Use it for different purpose ( baby sleep, Mailbox alert, mobile art installation... ).

( This project is not dead but bien raides since the move. But I do not despair of taking it back one of these days. )

At the beginning

As a juggler, I first used commercial programmable glowing club. But I encountered some limits : no multi platform, no radio communication, no flexibility, no 9-Axis, but above all no possible communication with developers to improve clubs.

Also I realized that it would be more fun and effective to make my own hardware : the clubs of my dreams ! or almost :),

I called them Rastello Club ( a glance at the famous Enrico Rastelli ).

  • It should be open source, multi platform and easy to contribute.
  • It should meet the requirement of a professional juggler. and at least mine :).
  • it could be used by a community of users (all level of programmers), by flashing the adapted firmware.

To begin I will focus on 4 firmwares ( maybe the 1 & 2 can be merged) :

  1. You can select pre programmed sequences.
  2. The use of a simple grammar to allow you make your own program.
  3. Implement micro python, and let the user be free to deal with Python power.
  4. It is not really a firmware but a C-Api with drivers already implemented. Demanding developers would feel totally free to use them or bare metal program.

Communication overview

You maybe wonder :

Why did I choose a RFM75 2.4GHz radio module and not Bluetooth ?
Because most of the time Bluetooth is used in a piconet and a master can communicate with a maximum of seven devices. The other solution ( scatternets) is less documented for a juggler who doesn't know very well Bluetooth. Also the data rate of Bluetooth low energy is 1 Mbit/s ( according to wikipedia ) while my little, low-cost radio module is supposed to go up to 2 Mbit/s.
So why not Zigbee solution ?

I like XBee module and I often use it when I need to implement quickly a wireless communication. But finding a very small module that could hold on my thin pcb was difficult, and all the more as I looking for a cheap one ! So quickly, I abandoned this idea.
And a WiFi module ?

I think this could be a good idea, mainly because of the famous ESP8266. It's really cheap and exists in versions that were small enough. Moreover, it seemed that the community had done a remarkable job. In terms of data rate, I couldn't find any information. According to a guy on IRC, it would be better than radio modules he used. The wrong side for our application is that it uses much more current. Finally, I preferred keeping the RFM75, but kept the ESP8266 in the back of my mind, in case there would be no more solution. It would be also possible to replace the ARM cortex M by the ESP8266, but for this project, I prefer a mature, well documented, and comfortable microcontroller.

A note on the Raspberry pi bridge and softwares

This page focuses on the hardware, so I'll not go into details, but I must talk a little about the bridge and soft. As we use a radio module in the club, we can't communicate directly with a smartphone, a tablet or a computer without adding a radio transceiver too. But smartphones, tablets and computers can easily use WIFI or Ethernet. For this task, I used a Raspberry Pi, with Ethernet and a USB WIFI dongle ( for now it suits me), who acts as a bridge between WIFI / Ethernet and 2.4GHz Radio. Thanks to freakone, I could quickly have communications with a smartphone, and have a minimal web app in Node.js ( also a little of python for the UDP/TCP server) to change club's colors.

Nodes.js, HTML5... allows having easily a multi platform solution accessible in a simple browser. I think this is a perfect solution to make an app with configuration tools, survival tools and daily tools.

However, I don't intend to make "the big one" app ( with GUI, multitrack, audio... ) as a web app, for 2 reasons :

  • I was not very much experienced in HTML5 and would be more efficient with QT / C++.
  • the aim of the application is not to be used on a smartphone.

I have already started a preparatory work with QT5 and C++, but I remain open in case a HTML5 ninja would join the project. But it would be also cool, to have other applications dedicated to new tasks.

Early link to Rastello...

Read more »


Rastello Club 0.1 alias Bob Schematic

Portable Network Graphics (PNG) - 159.83 kB - 05/02/2016 at 02:37



Rastello Ring (3 LEDs) 0.0 Schematic

download - 21.89 kB - 04/27/2016 at 01:07


  • 1 × STM32F411CE ( ARM Cortex M4 ) QFN48
  • 1 × RFM75 ( GFSK transceiver ) SMD PACKAGE
  • 1 × MPU-9250 ( 9 Axis MEMS ) QFN (3x3x1) mm
  • 1 × N25Q032 up to N25Q128 (Memory ICs / FLASH Memory) SO8 Wide
  • 1 × MCP73832 ( charge management controller) 5-pin SOT-23

View all 6 components

  • Early demo with new arrivals PCBs

    Laurent05/29/2016 at 20:11 0 comments

    New PCBs ready to be tested :

    Very first video :

    Hardware seems to be ok, it only remains to test flash. And put a new video here ( sorry for the quality ).

  • Rastello Ring version 0 ( PCB with 3 WS2812B )

    Laurent04/27/2016 at 22:36 0 comments

    Rastello Ring contains 3 LEDs RVB or RGBW ( WS2812B ) which glow the handle.

    You can take a look at the simple schematic.

  • Proto 0.1 Alias Bob

    Laurent04/24/2016 at 02:43 0 comments

    Alias Bob, ready to be sent !

    4 Layer PCB ( 1 layer for GND, and another for 3.3V ) :

    In theory :

    • Correction of MOSFETs errors.
    • Adding a reset supervisor to cut the buck and protect battery.
    • Improved routing and planes.
    • Improved Footprints and mask ( IMU, Buck Boost, USB, Microcontroller).
    • Removal of unused I2C pins.

    I know that I not really follow all High Speed USB design Guidelines, but I think I'll still keep this for now, because :

    1. Traces and planes are better than on my previous PCB. And it worked.
    2. USB DFU is 12mbit/s and i don't need to use USB CDC ACM at his max speed.
    3. In case of problems with USB Signal Integrity, I have some good musicians Idea that starts with the letter C for the next PCB. :-)
    4. And Finally, libopencm3 Ninjas confirm what I thought !

    You can look at the Rastello 0.1 alias Bob schematic.

  • Proto Zero Alias AC/DC

    Laurent04/17/2016 at 01:41 0 comments

    A log with a little delay, so that the project is more up to date !
    This is Proto Zero, there were some small mistakes, but fortunately apart the MPU 9250 footprint, nothing serious. It allowed me to move forward on drivers.

    Indeed, I advise you not to follow the recommendation of the datasheet, if you plan to solder it yourself ! I tried to extend pad on the new one.
    There are also others improvements
    . I will come back over, but first I would like to finish update the project details.

    Above Proto Zero, but I expect now Proto 0.1 Alias Bob...

View all 4 project logs

Enjoy this project?



unigamer wrote 04/23/2016 at 17:08 point

Hello, I'm always interested in LED props. I had a crack at making some of my own a couple of years ago (see video below) and have followed the developments of other people's attempts. So far I think the only properly smart props commercially available are but I don't that's open source. Another mention would be I don't have time to work on this kind of project just now but hope to in the future. Best of luck!

  Are you sure? yes | no

Laurent wrote 04/23/2016 at 22:36 point

Thanks, for the video and your github.  I'll add your link. This is the beginning of the project, but if you find it interesting, ( now or later ) ,  you are welcome. 

  Are you sure? yes | no

unigamer wrote 04/24/2016 at 07:37 point

I will do. I have a bunch of notes somewhere on things I'd like LED clubs and other props to do. From the top of my head the two most important were that the system should be designed with the idea of many types of props being capable of sharing the same system (eg hoops, poi, staffs etc). The second was the importance of making something that would actually be enjoyable to juggle (this is were my project broke down - I wasn't prepared to spend ages working on the ball casing).

I'm about to start making juggling clubs so if I get any results useful to this project I'll let you know.

  Are you sure? yes | no

Laurent wrote 04/24/2016 at 17:17 point

Indeed I agree with you about the 2 points. See you very soon. then !

  Are you sure? yes | no

duanesteel wrote 04/20/2016 at 00:01 point

Hey I made a similar project a couple years ago (based on teensy 3). I'd be curious to be involved if you're looking for contributors. I've made a set of 10 of my version already for some professional jugglers.

  Are you sure? yes | no

Laurent wrote 04/20/2016 at 00:43 point

Awesome ! 

  Are you sure? yes | no

Neon22 wrote 04/18/2016 at 11:34 point

This looks great! Are you aware of micropython on the ESP8266... ? It was KS a little while ago - met its goals and so initial builds are out for adafruit's Huzzah boards. Comms are up but not finished and dev is fully funded to MQTT integration level and python control.

Also I have had luck with this for inductive charging. Several choices...

Really hoping you go ahead with this - my dube clubs need an overhaul (Not that I could ever afford the postage these days I'm on other side of the world).

FWIW I agree on 6050/9250 - cheap and well documented thanks to OS community.

  Are you sure? yes | no

Laurent wrote 04/18/2016 at 14:41 point

Hi, yes i saw ESP8266 in the github path, and the Huzzah boards looks pretty cool. But as it is, the board is too large. Also i would have to review all my design and study in detail the ESP8266. On other side STM32F411CE ( have DMA, FPU, USB, DFU and 48 pins) and I have already started.
i'didn't know MQTT, I'll watch...
For inductive charging, i had thought it, but i don't find it necessary and adds complexity to clubs, ( my opinion might be different for balls).
I saw that you seem to know very well micropython, this isn't the first priority, but I would like to add libopencm3 in micropython path. When I'll be there, maybe I'll have some questions for  you :).

"FWIW I agree on 6050/9250 - cheap and well documented thanks to OS community".
not sure I understood.  In any case I find that there are gaps in InvenSense documentation,  particularly "Advanced Hardware Features".

Thanks for your interest

  Are you sure? yes | no

Neon22 wrote 04/19/2016 at 05:27 point

yes my information is to late. Your development is well underway. Looks very exciting. the 411 is an excellent chip I think. I'd like to help out in any way I can.

  Are you sure? yes | no

Laurent wrote 04/19/2016 at 14:03 point

Very cool, step by step, I'll push the project to github.

  Are you sure? yes | no

Yann Guidon / YGDES wrote 04/17/2016 at 02:11 point

At last ! your project finally has a page here :-)

  Are you sure? yes | no

Laurent wrote 04/17/2016 at 10:14 point

Thanks to you Yann ! Indeed, it seems that it is more practical than Facebook :)

  Are you sure? yes | no

Similar Projects

Does this project spark your interest?

Become a member to follow this project and never miss any updates