The IB + Engineer in Ecuador

Awesome news! The Montreal Gazette recently published a nice article about the international cooperation project I participated some time ago: The Ecuadorian Makecation.

You can find the article here:  Students’ trip to Ecuador is no vacation.

Montreal's Gazette Article

And anther picture just for fun:

Ecuador Trip - The TV Series

BTW, lately, I have been overusing the word “awesome”.

Around five years ago when I was in CEGEP (an obscure educational stage in Quebec between high school and university), I went on an international cooperation trip to Ecuador with my fellow students, in which we mainly planted medicinal plants, trees, taught English and built two bunk beds from scratch. It lasted roughly two weeks and left me with the desire to go back and do more

Now, five years later, I went back. The only difference is that now I am an Engineer (Jr. I might add) and was responsible for two overly awesome projects: the construction of a wind turbine, and of a solar water heater.

I travelled with this year’s International Baccalaureate group. Needless to say, being back with an IB group brought me very good memories of my previous trip and all things IB (such as Cardiff).

As expected, the construction of both projects did not go without troubles. There where many unexpected obstacles and a lot of hard work was required in order to complete them. Luckily, I was with the best group of young people I have ever encountered. The Andre Laurendeau IB group of 2010 that travelled with me is remarkable and exceptional. Composed of Nicolas, Roxane, Tiberiu, Angelo, Geneviève and Adam, this is a hard-working, tenacious and dutiful team.

IB 2010 group in Quito, Ecuador

As you imagine, I loved this experience. So much so to be already toying with ideas on what to do next year  for the next trip. Although I might not be able to go, I might still be able to imagine some cool projects.

Acknowledgements

I would like to thank the IB in entirety for all the hard work they put in this project, not only thinking about the design and building it, but also for all the fund raising activities that made it possible.

Also, I would like to thank Stephanie who donated many of the awesome tools I brought to Ecuador and that were crucial in making the projects. The single most important tool she gave me being the Leatherman Wave (for Christmas) which is responsible for achieving almost everything during the trip.

Leatherman Wave in Action

Forced Acknowledgement

The OQAJ founded part of this project and deserves my gratitude. If I say it is a forces acknowledgement, it is precisely because I am forced to publicly thank them. I am very grateful for their support, but I do not enjoy being coerced into doing anything. It would have been much more appropriate if they would have just kindly asked for a public acknowledgement. Or, even better, if they were confident enough that they are being helpful so to think people will thank them naturally.

What about the Projects?

I know, I started with all this Maker thing and so far there are no instructions, no pictures of things being built, etc.

The projects deserve a lot of writing, so I will dedicate one, if not several, posts to each one of them: the wind turbine, and the solar water heater.

Stay put for more!

Crashed STM Ticket Vending Machine

The malfunctions were mainly caused by the poor build quality of the smartcards, the poor wireless reader range, the easy demagnetization of the paper tickets, etc.  Slowly (very slowly) these malfunctions are being addressed and (sometimes) solved. The unsolved bugs become a part of the daily routine and we Montrealers learn to accept them.

On last weekend, I witnessed a bug that I would have never thought possible in a professional widespread software application that involves so many money transactions: what I usually call a boundary condition bug.

The Bug

Quick note on the STM’s fares (for those who do not live it day-to-day):
There are mainly two transportation systems: the bus and the metro (subway) and the fare pricing is governed by a (somewhat) simple set of rules:

• There are monthly passes that allow unlimited fares within the month in both the metro and the bus. They are more expensive for adults than for students or elders.
• There are weekly cards that are similar to the monthly cards by work for a given week.
• There are tickets (either as a magnetic paper card or as some information on a smartcard) which allow for one fare that allow to take the metro, the bus, or both within a two-hour time limit. Should the time period be elapsed, or should you take the metro or the same bus more than once, you will need another ticket.

Since the Opus Card implementation, one of the first things that came into my mind was the attention required to compute the fare price when a month ends (i.e. at 12:00 on the 31th, 30th, 29th, or 28th), or when a week or a day ends. Everyone who has programed at least a little bit (like me) knows that these boundary conditions are usually exceptions to the normal behaviour of a program and need to be taken into account. I, of course, assumed that such mundane exceptions would be addressed by professional programmers swiftly and painlessly. I was wrong.

On last Saturday, my girlfriend and I took the metro after going to the movies and she swept her card (loaded with tickets) at exactly midnight (00:00 as reported by the STM records) at the metro reader. Around 20 minutes later, we walked into a bus and surprise! another ticket was charged instead of using a transfer from the previous ticket.

The conclusion to this is that when a day ends, there STM fare algorithm fails and defaults to charge you an extra ticket. Unfortunately, It is unlikely that this bug is noticed by the STM any time soon since they are not loosing any profit from it and they do not accept bug reports. Even the lady at the reclamations booth had a hard time to understand what the problem was.

In the end, instead of a full refund she got a “courtoisie” ticket which expires on next Sunday (instead of a regular ticket that which expires in around two years). So much for costumer service.

The poor programming quality is also reflected by the poor choice of OS for their vending machines. Let’s just say that using Windows Embedded is the equivalent to eating a faulty grenade: it will rather sooner than later get ugly.

Workshop wall

This is a very simple and quick hack that anybody can do. The objective is to build a holder that will keep all the precision screwdrivers in one place and facilitate the access to them (i.e. it should be easy to see which one is which and they should be easy to take).

Materials

• A plastic container cap (I love caps)
• That’s it, you do not need anything else

Tools

• Drill or strong scissors
• Precision screwdrivers
• A nail (maybe)

Doing it

Screwdriver Holder

1. Drill or cut a big hole on the top of the cap. This hole will be used to suspend the screwdriver holder.
2. Punch in the Screwdrivers in  the rim of the cap. If you do not trust your screwdrivers to pierce your cap, you may use a nail. Keep in mind that the screwdrivers have to be very snug in the holes since they are only being held-in by friction.
3. Celebrate! you are done.

Logic Unboxing

Logic Analyzer?

(skip this if you already know what a logic analyzer is and do not want to be bored to death)

A logic analyzer is a measurement instrument that allows you to record the logic level (e.g. high or low) of a digital signal throughout time. This means that if you are working with a micro-controller and want to know what is actually happening at the pins, you can use a logic analyzer to record the signals of interest and display them on a time-line on a computer screen. This is very convenient considering that debugging microcontrollers often implies guessing or using lame techniques such as slowing down the execution time and lighting up LEDs in order to see what the program is doing.

In short, a logic analyzer allows you to view digital signals and plot them with respect to time.

Casing

The Logic, by Saleae

The Logic has a nice square anodized aluminum body which makes it very sturdy and looks awesome. It is very small (~ 4.5 cm on the side and ~ 0.7 cm high) and features some male headers on the front side and a mini USB port type B on the back. For some reason, I really like the four tiny hex screws that hold together the two aluminum plates that make the body.

Accessories

Cabels & Hooks

The Logic comes in a nice Logic Case, usually intended for some other electronics such as external hard drives I suppose, that can accommodate itself  and its USB cable, its set of wires and the E-Z-hook clips.

What I enjoy the most is that the wires are robust and kink-proof, and they are terminated in female headers, which makes them very easy to use in breadboards and on male pins in general. Also, the hooks are removable and they can connect to harder-to-access pins in a secure way. The wire colour coding follows the resistor colour coding (i.e. Channel 0 is black, channel 1 is brown, etc) and this might be a bit confusing at the beginning since the Ground wire is grey instead of being the usual black.

Finally, the carrying case is a very nice feature since it allows you to have the Logic and all the related items in one single place. The case is a bit too big, but I the extra space could be used to store some extra accessories, such as headers gender-changers (for plugging into female headers) or wire extensions with special ports (e.g. DB-9, RJ-45). I also enjoy the fact that all the components come in their own little zipper-bag that allows you to store the unit in a nice pristine condition (it may not seem like it, but I like to preserve things in their original state, unless they come from the garbage in which case I mod them and unscrupulously use their pieces).

Hardware

I did not want to take the Logic apart in order to see its guts (so far) so I do not have too many details about the hardware other than what is listed in Saleae’s website: (1)the inputs are protected to you don’t fry anything, (2) it can sample at 24 MHz,  and (3) it is able to store up to 500 M samples.

Software

The software can be downloaded for free (and used in demo mode) looks really nice but unfortunately, the current stable version works only under the Redmond OS and it does not play very well with Wine in Linux or in my XP virtual machine. I would have really liked to test its Protocol Analyzers, a function that automatically sets the names of the signals and decodes the information accordingly to the protocol being analyzed (e.g. I2C, RS232, SPI, 1-Wire).

Anyways, there is an upcoming cross-platform version of the software that looks very promising and is currently under private beta at the moment. Of course, I could not resist the urge to join the beta testing and I can say that the new software is working very nicely (so far, I have not tested it extensively yet) under my 64-bit Kubuntu setup.

Needless to say, as soon as the software goes public, I will post a more thorough review, so keep posted.

Conclusion

The Logic is a very nice piece of equipment for any hobbyist or professional (provided you do not require to read signals faster than 24 MHz, which is rare specially while debugging). It is built to last fits very nicely in a hacker’s toolbox (or even pocket). I would recommend the Logic to anyone needing a sturdy and easy-to-use Logic analyzer, and I will be using it in my upcoming projects.

Google Wave

I finally got a Google Wave account and I decided to do something for the community. More precisely, the hacker/tinkerer/DIY community.

I think it would be rather interesting to have a place where to share, showcase and discuss projects and hacks with local people (from Montreal).

If you want to join the wave you can find it here.

I hope this will result in a stronger more connected Montreal Hacker community.

Cheers, and see you on the wave!

I have a few Wave invites left so if any geeky Montrealer would like one, please leave a comment and I’ll see what I can do.

Xilinx Spartan

I spent countless hours trying to install the free version of the Xilinx software on may windoze Virtual machine and I have finally succeed.

I must say that although I have never used Xilinx ISE Webpack, I already hate it. It come in a huge installation archive (2.7 GB) and an equally huge update archive (2.4 GB). The option to use the “slim” installer (88 MB) is worthless since there is no ways of saving the required files for an eventual reinstallation (which was needed twice in my case).

Anyway, after installing/uninstalling/installing many times I finally got it to work. And this very lengthy and painful process reminded me why the Open Source world is so much more convenient. Installing a full linux distribution with tons of extra programs takes half the required storage and a third of the time. Too bad FPGA programming tools are ruled by the chip manufacturers.

All this painful process was required in order to get my new NanoBoard 3000 running. Hopefully I wont have to use Xilinx ISE anytime soon. Altium Designer was much easier to install and requires less storage.

Ever since I got my EeePC, I got obsessed by the amount of screen real-state Firefox takes for the top menu, browsing control buttons, address and search bar, etc.

The quickest fix I always apply every time I am in front of a Firefox windows that is not mine is to move the Bookmarks toolbar content right next to the File menu and to disable the Bookmarks toolbar.  This removes one full toolbar, and unless you have a really heavily populated Bookmarks toolbar, should function very well.

But I wanted to go further (especially since I saw that Chrome uses less screen real-state). So I installed a few add-ons that make a significant difference:

• AHS. autoHideStatusbar does precisely that. It hides the status bar unless it is required (i.e. a page is loading or you hover a link) or you go near it with the mouse pointer.
• Smart Stop/Reload. It combines the Stop and Reload buttons (since they are never used at he same time).
• Tiny Menu. It transforms the File menu into a single item. This extension saves a lot of real-state and the menu remains nice ans usable.

By using these extensions and moving things around in the toolbars, you can achieve a very tiny navigation interface that is perfectly usable.

Tiny Firefox

Note the Tux theme (I a’m using Personas)

The Altium NanoBoard 3000

Now that I got it, I’m dying to have some spear time in order to try out their awesome board with their quite intriguing IDE.

About the Hardware

The NanoBoard 3000 is a beautiful piece of equipment. It boasts lots of cool peripherals and is built and packed as a luxury electronics product. It is a very good looking piece of equipment while remaining perfectly functional, it beats by far all the other development boards I am familiar with (i.e. the Altera DE2 board and the Lattice Mico32 development board) while remaining much cheaper (around 50% of their price).

NanoBoard 3000 Unboxing

Features

Here is a selection of the NanoBoard’s features from Altium’s Wiki I find most prominent:

• A Xilinx Spartan-3AN device (XC3S1400AN-4FGG676C)
  

  Nanoboard 3000 Front

• 4 Serial SPI Flash memory devices
• Programmable clock 6 to 200 MHz, accessible by Altium Designer or by an FPGA design
• SPI Real-Time Clock with 3V battery backup
• Adjustable voltage regulators set to generate 1.2V, 1.8V, 2.5V and 3.3V power
• 256K x 32-bit common-bus SRAM (1MB)
• 16M x 32-bit common-bus SDRAM (64MB)
• 8M x 16-bit common-bus 3.0V Page Mode Flash memory (16MB)
• Dual 256K x 16-bit independent SRAM (512KB each)
• 256K x 16-bit independent SRAM (512KB)
• 8 RGB LEDS
• 5 generic push-button switches
• 4-channel 8-bit ADC, SPI-compatible
• 4-channel 8-bit DAC, SPI-compatible
• 4x isolated IM Relay channels
• 4x PWM power drivers
• Screw terminal headers for ADC/DAC/Relay/PWM interfaces
• SD (Secure Digital) card readers:
  • One for use by the Host Controller FPGA
  • One for use by the User FPGA
• SVGA interface (24-bit, 80MHz)
  

  NanoBoard 3000 Back

• 10/100 Fast Ethernet interface
• USB 2.0 High-Speed interface
• RS-232 Serial Port – DB9M
• RS-485 Serial Port – ‘RJ45′
• 240 x 320 TFT LCD with touch screen
• 8-way DIP-switch
• Stereo 2W audio power amplifier with 3.5mm test input jack and DC volume control
• 24-bit Stereo Audio CODEC with I2S-compatible interface
• Stereo audio jacks (3.5mm):
  • Line In / Line Out
  • Headphones
• Speakers on a separate (attached) board
• MIDI interface
• Diagnostics interface – PCI Express (PCIe) edge connector for connection of automated test equipment (ATE)
• 1.8″ ATA/IDE connector providing access to user LED and generic switch I/O
• Remote Control and IR interface.

I should post some further details and perhaps even a simple test project soon (as soon as I get Altium Designer and Xilinx ISE installed and running)

Kubuntu Karmic Koala

Kubuntu Karmic Koala is finally out! I use it since the Release Candidate came  Oct. 22 nd, and it is absolutely awesomely mind-blowingly fabulous.  All of the kinks in Jaunty have been fixed and a lot of new features have been added.

Kubuntu?

Why am I talking about Kubuntu and not about its more popular sibling Ubuntu? Well, very simply because KDE kicks Gnome’s ass any day (while blindfolded and with all of its finger stuck in its nose). I know that seems like a very bold and unjustified statement, well it is indeed very bold but totally justified.

The main difference about KDE and Gnome, besides the fact that the KDE foundation is much more solid, flexible and portable, is the mindset. In KDE you can configure (trough a nice GUI) pretty much everything, whereas in Gnome, you get a bunch of very comfortable defaults that (although they can be modified) are not intended to be fiddled with too much.

Also, KDE is much more than a desktop environment and provides a full suite of programs that do almost everything you could want to do. These programs also integrate very well together and provide as many more features and options than any sane person would need or be able to use (but who likes sane people anyway?).

Quick Review

My Desktops (Grid View)

I am currently using the 64-bit version of Kubuntu and it is performing incredibly well. The system (my laptop) boots in around 40 seconds and turns off in less than 15 seconds. The graphical performance is flawless and I can benefit from smooth performance even when doing very processor intensive tasks (such as stitching photos together).

Also, It comes with Ubutu One (a remote storage service) which is pretty convenient for sharing and backing up files.

I’ll try to do a screencast and post it in order to show off the Koala.

I was very lucky to work with Panaxion, a start-up company based here in Montreal, on the manufacturing-prototype of their invention, the Axion. It consists of an automated on-the-fly paint-blender and paint-applicator.

The Axion?

The Axion is a paint mixing device composed of five paint containers, a paint flow control system and a paint blending apparatus. The whole system is controlled by a PC. As you select a colour from a user-defined colour palette in the Coloraxion software, the machine dispenses the appropriate amounts of different colour paints in order to produce the desired blend.

The machine is rather ingenious and is going to make its debut in the professional decorative painting realm very soon (I’ll keep you posted on that).

My Involvement

I was involved in the mechanical and electrical design of the first manufacturing prototype. I was lucky enough to learn SolidWorks and draw the entire machine (up to the very last detail). The resulting 3D model was then used to build the actual prototype and to produce some multimedia material. Needless to say that I was very happy to participate in the construction.

Title:	The Axion
Description:	Concept video based on the SolidWorks CAD model.

Did you say start-up?

Panaxion is a start-up company and is soon going to manufacture Axions. Their website is under heavy construction at the moment but it should become stable (and beautiful) soon. If you are interested in decorative painting (in the non effeminate sense), I suggest you keep an eye on this project.

After having the rig working, I went out on my bike with my tripod, my laptop, my digital camera and the panoramic photography ri. I decided I wanted to capture the sunset at Mount Royal but when I was halfway there, I noticed that the sun became completely covered by clouds.  So I decided to shoot some other panoramas instead.

CCA

Here is a nice 360 deg view of the Maison Van-Horne Shaughnessi (a.k.a. The Canadian Centre for Architecture). This building is located here. Keep in mind that you can click on the title in order to see the panorama full-screen.

Title:	Maison Van-Horne Shaughnessi
Description:	A 360 deg panorama shot from the sidewalk.

When watching carefully, it is possible to notice a little parallax error near the base of the descriptive plate in front of the building. I could have corrected this but I noticed it too late and I am extremely lazy. You can also see some cars blending with their surroundings as Hugin, the panorama stitching program, tries to make sense of a changing scene (with cars passing by).

Place Ville-Marie

This is a panorama of Place Ville-Marie taken at the central fountain.

Title:	Place Ville-Marie
Description:	A 360 deg panorama shot at the Place Ville-Marie fountain.

The parallax errors in this one are much more significant, but then again, laziness got the best of me. They are mainly caused by the fact that the camera was not perfectly flat on the tripod when shooting this.

If you want to know what are the tools used to create these panoramas, please read my previous posting on Making Panoramas.

This script requires my Pololu library and includes some codes from here in order to control the camera. Besides the basic requirements of lib_pololu,  the script also requires gPhoto. If you are running Linux, you most likely  already have it but in case you do not, you can install it through your favourite package manager or by using the console (e.g. for Ubuntu/Debian):

sudo apt-get install gphoto

If you are using some other OS, you can download gPhoto from here.

The Code

Again, many thanks to RobotShop who provided the hardware that made this project possible.

Read the follow up posts: Panoramaker, where I expose the software, and Automatic Panoramas in Montreal, where the final result can be seen.

I have finally completed my second project sponsored by RobotShop. I apologize for the immense delay, I really missed my promise of rolling out a new project every two weeks. Let’s just say that I had a lot going on lately and I could barely keep up with my obligations, let alone blogging or building new projects.

Place Ville-Marie Panorama

This time, I built a panoramic camera. My main objective was to have a platform that can be used with pretty much any camera and that can produce panoramas with a minimum of work. If there is enough interest from the public and if this prototype is well received by the DIY community, I’m planning to produce (and hopefully sell) kits that would include all the parts to build this device.

Materials

• Hitec HS-785HB Winch Servo Motor
• ServoCity SPT200 Direct Drive Pan & Tilt System
• Pololu Micro Serial 8 Servo Controller
• Hitec HS-425BB Servo Motor
• A digital camera
• Small plastic container
• Battery holders
• (Rechargeable) Batteries
• A piece of aluminum (I used an old heat sink)
• A long nut (of the same size as a tripod screw)
• A screw (that fits into the tripod attachment on the camera)
• A tripod
• Cable ties
• A small piece of neoprene (or any other mushy substance)
• A USB to Serial cable
• Some female headers and wires

Putting It Together

Panoramic Camera Mount

The first step was to put together the ServoCity Pan and Tilt system. This took away much of the building work since it is really simple to put together in no time at all. Nevertheless, I applied some modifications to it: I discarded the bottom plate that should be attached to the panning servo (since I am using a larger winch servo that would not fit otherwise), and I drilled a hole on the top plate in order to be able to fasten the camera to the rig. Note that I also included a little piece of neoprene that was lying around in order to prevent the bottom of my camera from getting scratched.

The mounting hole for the camera must be placed so that the lens’ pupil is at the centre of rotation. This way, the horizontal rotation axis will be close to the no-parallax-error point  (or whatever it is called) of the camera and will minimize the parallax errors.

Then, I used an old heat sink as the main structure since it is sturdy and basically free. I used the trusty Dremel to adapt it and cut the proper holes and slots in order to mount all the remaining pieces. The pieces to be mounted on the aluminum plate are the battery holders, the Pololu servo controller, and the winch servo motor. (or whatever it is called

I encapsulated the Pololu servo controller in a small plastic container I got from for free while on a trip with my girlfriend to the beauty/ soap/cream shop. I also used two 2-AA battery holders in order to provide power for the servo motors. I used 29000 mAh NiMH rechargeable batteries that gave me several hours of autonomy. In order to connect the battery holders to the controller, I soldered a two-position female header and insulated the leads with heat-shrink tubing.

I used almost exclusively cable ties to tie everything on the aluminum plate except for the winch servo motor that I screwed in and the long nut that was also screwed in place (after being drilled sideways).  I also had to drill the bottom aluminum face in order to allow for the tripod screw to be inserted into the nut.

Operating it

Panoramic camera in action

This first prototype requires a laptop to be operated, which can be a little annoying.  I plan to use my EeePC in the immediate future and an embedded computer for an eventual commercial kit. It basically works as follows:

1. The camera is set on the panoramic mount, which is fastened to the tripod.
2. The servo controller and the camera are connected to the computer trough their respective USB cables.
3. The controlling program is run.
4. The user waits in awe while the camera takes pictures by itself.

In order to control the hardware, I use a python script that uses my Pololu library and gPhoto in order to operate the servos and the camera respectively. I chose gPhoto since it supports a very wide range of cameras and it is very easy to use.

For now, taking a full 360 panorama takes about 15 minutes. This is a very long time and is mostly due to the fact that my script was hastily put together without care about the performance and in very little time. I will, very soon, post a cleaner version of the code, as well as all the panoramas I took properly processes and in full format, similarly to what I did with my San Francisco panoramas.

Acknowledgements

RobotShop.com

I would like to thank the great people at RobotShop for providing the Pololu Micro Serial Servo Controller, the  ServoCity SPT200 Direct Drive Pan & Tilt System, and the Hitec HS-785HB Winch Servo Motor. This is the second (and hopefully not the last) project they sponsor here at Carlitos’ Contraptions. Without their help, I would have never been able to afford any of the materials (except for those that come straight from the garbage as usual).

They have also being very patient and understanding about my unexpected delay in rolling out this project.

I had an old blazer from my girlfriend’s father that was too oddly shaped to fit anybody I know, but I liked the fabric and I thought that it could become a very nice laptop sleeve. The only problem about that idea is that I do not know how to sew. Of course, not knowing how to do something, has never stopped me before and is not likely to stop me any time soon.

Tux laptop Sleeve

Materials

• Old blazer
• Zipper
• Fabric Marker (not required but rather cool)

Doing it

Making this was surprisingly simple since I expected great difficulties coming from the sewing part. I decided to go with a design as simple as possible and to minimize the number of stitches, this resulted in making something extraordinary similar to a simple cousin.

I simply cut a rectangle big enough so that, when folded in half, it could contain my laptop from the blazer and then stitched the bottom and the side of the resulting pouch.  Then, I added a zipper to the top. I used zigzag stitches all the time and, of course, I sewed the sleeve from the inside.

Also, I chose a section of the blazed which had an internal and external pocked so I could use them to carry some extra stuff. Furthermore, I kept the (synthetic?) silk interior lining that looks better than the bare exterior wool.

Finally, I wanted to ad some decorations to the pouch so I decided to use my brother’s fabric markers (he is using for making pretty cool disguises) to ad the word “Linux” to one side of the sleeve. Then I asked him (since he is much more talented than me at drawing (and since I am very lazy) to draw Tux on the back. To achieve that he used a stencil made from the image below that I got from the internet.

Tux Stencil

I think the final result is pretty cool and very useful. The pockets are great, I can put all of the things that I usually need to carry with my laptop (e.g. mouse, headphones, USB keys) without any troubles. The only thing missing is perhaps a handle that I may add in the future.

My Brother Painting Tux

I opted for a Dell Vostro 1320 and must say it is a beautiful machine. It is small, powerful, light, and most importantly shiny.

Vostro 1320

Some Specs:

• Intel Core 2 Duo P8700 @ 2.53GHz
• nVidia Corporation GeForce 9300M GS
• 4GB RAM
• 320 GB HDD
• 13.3″ screen matte (I can’t stand glossy displays)
• Intel wireless adaptor draft n
• 12 cell battery
• Mono integrated speaker (this is its only weakness)

I am running Kubuntu 9.04 with KDE 4.3.1 and it is absolutely marvelous. Everything works perfectly (except for the microphone). As a matter of fact, KDE is so absolutely mind blowing I’ll make a post exclusively about it soon (hopefully).

I know I am running late with my posts and answering messages but I’ll post some new projects very soon.

My post on the Xbox Media Center has been getting thousands of hits of  “people” trying to posts comments. In order to stop this, I disabled the comment feature in this article and I changed its address. Eversince, the site has been doing fine without any errors.

And now, for something completely different:

Mysterious Safe Box

I found a little safe box in the garbage some time ago and I would really like to open it. The box features some dents and scratches that show some people tried to open it in the past but did not succeed. Also, when shaken, it produces some interesting noises that indicate it is not empty and may contain a valuable treasure some cool junk.

The Plan:

Of course, there are many ways of achieving this (e.g. cutting holes into its walls, removing the hinges, hiring a locksmith) but I do not want to break the safe, nor do I want to hire anyone to do it in my place. Inspired by so many movies, the first thing I tried was to listen to the (nonexistent) clicking of the dial as it turns: it did not work at all.

Very quickly, I decided that the appropriate way of doing it (given my set of skills) was to have a machine do it in my place. So, I decided to build a little manipulator that will test all possible combinations of the safe until it opens up.

Since this requires precise positioning, I thought a servo motor would be the best choice of actuator (and also because I have some other project ideas involving servos).

The Materials:

• Pololu Micro Serial Servo Controller
• Hitec HS-425BB Servo Motor
• Male and female headers
• Power adaptor (4 to 6 VDC)
• Wire
• Heat shrink tubing
• USB to Serial cable
• 2 gears and a spindle (one four times larger than the other, I got mine from an old photocopier)
• A project box (in my case, an old computer power supply case)
• A bottle cap and an old heat sink
• Steel wire (e.g. form an old coat hanger)
• Magnets (the stronger the better)
• Various screws

Essential Tools

• Screwdrivers
• Rotary tool (Dremel)
• File

The How and the Why:

Electronic Parts

I chose the Pololu servo controller since it is easy to interface by either using a computer or a simple microcontroller. Also, I already own a USB-to-serial cable (that I normally use to program the Arduino) that I can use to send commands to the controller.

The choice of the servo motor was based in getting the maximum torque at a reasonable price.

In order to control the servo motor, I devised a simple python module as seen in my previous post.

Since, usually, servo motors have a motion range slightly greater than 180 deg, I decided to use gears to be able to produce a motion range large enough to operate the safe (at least two full turns).

Mechanical Parts

I used an old heat sink and cut it with the Dremel in order to produce a bracket for the servo and a mounting hole for the secondary (smallest) gear axle.

I attached the larger gear to the servo directly using the brackets and screws that came bundled with the motor. Also, in order to hold the shaft in place, I used e-style retaining rings and spacers (other dummy gears and shoulder washers) in order to match the grooves already present in the shaft. Of course, I got all these handy mechanical parts from an old photocopier I found in the garbage.

Dial coupling attached to the small gear

In order to couple the small gear to the safe dial, I used a bottle cap which fitted perfectly over it. The cap has some child proofing which provided a firm grip for the dial.

The Project Box

Once the assembly was done, I mounted it into an old power supply box. Although the box required some drilling and cutting, it was very easy to adapt. I would recommend using this type of boxes for other projects since they are sturdy and easy to machine.

Electronic Assembly

So to include the electronic parts into the box, I used a small plastic capsule (that must come from the time when my young brother liked those little toys they sell in a dispensing machine). In the capsule, I placed the Pololu servo controller, the USB to serial cable and the servo motor cable, and the power supply cable.

I needed to adapt the USB to serial cable in order to match the pin-out on the controller. For this, I used a male and a female header, and some wire. I connected the GND and the V+ pins to their respective counterparts, and the TX pin to the Serial-in pin.

In order to supply the appropriate power, I used an old cellphone charger and I replaced its original connector with a female header. As always, it is good practise to isolate the electrical connections with some heat-shrink tubing.

The (disappointing) outcome

Very gentle safe cracker

Once everything was done, I fixed the new little machine to the safe and got ready to get it trying codes. I used a coat hanger and some rare earth magnets in order to hold the cracker firmly in place.

To my great disappointment, I realized that the gears I used provoked a (much expected) torque reduction. This meant that the system is not strong enough for turning the dial to a set position reliably. No matter how much lubricant I used, the dial was too stiff for the little robot.

This represents a (temporary) victory for the safe, but the war is far from being over.

Epilogue

I decided to publish these results in spite of my failure since too often we read reports of success (notably in science) and often forget that we can learn from failures as well. Never do we read about scientits proving their original hypothesis wrong, but very often, documenting those mistakes could prevent others from doing the same.

Finally, for those wondering how I was planning to pull on the safe lids so it opens, it is remarkably simple: since the safe does not have any handle to latch it closed, I merely need to hang it from its handle and try the codes until it opens and the bottom part goes down.

The safe hanging unaware of its fate

Finally (this time for real), I cannot say too much about my next plans on attacking the safe, but be sure that they involve a stepper motor.

Acknowledgements:

RobotShop.com

I would like to thank the great people at RobotShop for providing the Pololu Micro Serial Servo Controller and the Hitec HS-425BB Servo Motor. As I mentioned before, they will be sponsoring a set of projects here at Carlitos’ Contraptions in the foreseeable future.

They were also kind enough to quickly ship a replacement gear set when I ruined the original one on the servo motor by hand forcing it to turn (I know, I sound brilliant).

Pololu Micro Serial Servo Controller

I started writing a library for controlling the Pololu motor controllers with a computer trough a serial port.

I’m writing this in Python so the code can be cross-platform but I would be very glad to have some feedback about running it on other OSs than Linux. Actually, any feedback would be very welcome.

As of now it can interface with the Pololu Micro Serial Servo Controller that I got form RobotShop.  I am planning to use this code in my upcoming project RobotShop is sponsoring. I will supplement this library as I get newer hardware to work with.

I know there is already a python interface for it but I really wanted to have an object oriented way of managing motors (i.e. they can be instantiated and controlled more easily).

You can download the library here: lib_pololu.py (you will need to change the extension of the file to .py instead of txt).

In order to properly use this library you will require:

1. Python
2. Pyserial

If you use a civilized OS you may be able to get all this by typing this in a command prompt:

sudo apt-get install idle python-serial

Here is a sample script that will use the library in order to control a servo: servo_example.py (you will need to change the extension of the file to .py instead of txt).

Note for Redmond OS (aka Window$) users: you will need the Win32 Python extension for pyserial to work.

RobotShop.com

The robotics parts retailer RobotShop will be sponsoring robotics projects at Carlitos’ Contraptions. As a result I’ll be rolling out robotics projects for the next weeks/months.

I’ll try to make simple projects at the beginning and build up the difficulty level as I go. The idea is that, if you completed the the first projects, you will be able to reuse the skills/parts/code to build the future projects.

Of course, all articles are going to be published under the usual Creative Commons license and the code is going to be published under the GPL license.

I am planning to use Python for any program running on a computer and Arduino as a micro-controller platform. These choices aim at making the projects as open and as accessible as possible. Also, they ensure that they are perfectly cross-platform.

As always, I am open to any suggestions or wishes regarding the projects.

Note: the RobotShop logo above is not published under the CC license.