After the success of the Fakugesi PCB festival Pass, we were contacted by North West University to design an interactive badge for their GeeXpo event. We started playing with a few ideas, but as usual time caught up to us and we realized that we had 10 days left before the festival and no plan in action.
After a quick chat to our main PCB guy at Bosco (Hi Winston!), we figured out that we could probably still pull this off. A design was made and after a quick prayer to the PCB Gods we sent off the Gerber files for manufacturing.
Luckily we designed it around components that we either already had or could get in qty locally. While we waited for the PCB’s we sourced the outstanding components.
The design basically consists of my favorite STM32 micro-controller, an IR Receiver, some passives and led’s. The general concept was that there are various challenges at the GeeXpo. Complete a challenge and with an IR transmitter it would send a code to the user’s badge. This would turn on an LED on the badge to indicate that the person has completed the challenge.
It seems pretty simple, but it was very well received and everybody loved having a PCB badge that had blinky leds. Since the challenges were held at various exhibitions/stalls around GeeXpo the attendants really got involved in each exhibit.
Anyways, I digress. We had the 100 PCB’s and the components by Wednesday evening. I built a few up to test the main hardware and that we had a few prototypes so that Sebastian can start writing the firmware. Luckily no issues with the PCB’s … phew, thats half the battle won.
At our weekly thursday meetup, I lured everyone with beer and promise of new knowledge. It was a mere half-hour later and I had everyone working in a production line that would put a sweatshop to shame … 😉
Not sure if everyone had fun, but they all worked hard. (Thank You!)
Wolff monitored the re-flow ovens (and the beer …)
Sebastian/Romeo and Tom also spent the Friday fixing, soldering, building and flashing firmware.
The final badge works very good and it looked awesome!
Romeo and Sebastian put together some pretty cool base stations that transmits the challenge completed code when … you completed a challenge. Chris made some awesome laser-cut enclosures for the base station.
Saturday was GeeXpo and boy did we have fun. We had lots of interactive exhibitions for a change and everyone thoroughly enjoyed it.
The challenge at our table was a robot arm that was controlled by a game-pad. You had to use the game-pad to pick up a trinket and drop it into a bowl.
We met a lot of new people and as usual the people enjoyed our projects.
Thanks to everyone that helped out with this badge project, its much appreciated and see everyone at the next GeeXpo!
On 21st and 22nd of October we participated in the Microsoft Open Government Data Hackathon. The theme around this hackathon is that you build/develop something over these two days that make use of the openly available government data. There has been numerous hackathons like these all over the world. The prize was R50000! That is the kind of money that could mean a lot for BinarySpace in terms of getting tools and equipment for the space.
While we were a bit uncertain about what exactly we were going to do, we decided to form a team and enter the hackathon (this required some emails and a few leave forms).
Before the hackathon we started looking at various open data sources to see what is out there and what we can do. After lots of discussions on our local telegram group we decided on an idea. The requirements for this hackathon was basically; build something over the two days that use the government data to give it a better meaning and make it available to everybody. There was also a mention to make it fun!
From the beginning we knew that most of the open data hackathons ended up in various cool websites and apps. We wanted to do something more physical and different. We got the inspiration from the Points Sign, which is essentially a smart street sign. The hack was to build something like that from scratch over a period of 2 days, and using government data in a meaningful way. From the beginning this was a very ambitious hack in the timeline we had. We settled on a more basic government data set which was the positions of every public government facility in the country. This includes Hospitals, Home Affairs offices and even schools.
So the plan was to create a smart street sign. It would have a control panel so that you can select which government facilities you are looking for and also have some emergency feature. Press the button for the facility you wanted and the signs would point to the 3 closest facilities while showing how far each one is from the sign.
We are calling it Pointr! The smart logo you see above was designed by our Isobar assigned team member and new friend of BinarySpace Rico Smith.
The biggest challenge in this project is the mechanical side of things. Making 1 sign rotate is easy, add 2 more on top of that and it becomes more difficult. Each sign needs a motor, power and communication lines for the sign. That is a lot of wires that need to go through the same pipe with the rotation mechanism. Granted we could have left the wires hanging on the outside (it’s a hack after all), but we wanted it to look nice.
Andre and Chris handled the mechanical side in terms of design and printing. Andre made a very cool design that involved a pvc pipe, a steel pipe welded onto a base and some 3d printed parts
The motor is essentially offset on the inner side of the pipe, a belt/pully mechanism used in 3d printers is used to do the rotation. This leaves space to bring through the wiring for the next rotation mechanism. Our sign can do a full 360 degrees, but not continuously. This means that if the sign is pointing at 340 degrees and needed to go to 10 degrees, it would basically move counter clockwise to 10 degrees instead of just moving forward over 360 to 10 degrees. For our purpose this wasn’t an issue at all.
The sign was cut from 6mm acrylic, and the plastic parts were printed with PLA.
On day 1 we had the first movement up and running, and we were very happy with how smoothly it went
Unfortunately on day 2 we decided to spray paint the pole. The not-fully-cured paint messed that up and we spent a while scratching off paint where the rotational parts go. Lesson learned 😉
For the main control panel, we had originally planned on just drilling a few holes in the pipe and mounting buttons with a printed piece of paper indicating which buttons to press. We are however suckers for a challenge so we decided to 3d print panels, and while we were it 3d print the icons as well
Rico designed the icons, which Andre converted into something printable, which Chris then printed … good teamwork 😉
Since there are lots of various government departments we decided to only take a few for our prototype. We decided on 4 emergency buttons; panic button, police, hospital and fire department. For the information side we chose Home Affairs, Libraries, Child Services and Courts. Press the panic button and police would be notified that someone needs help close to the sign. Press any of the other buttons and it would point the 3 signs to where you want to go.
All in all our mechanical side worked and looked beautifully
For this hack the electronics were relatively simple. Michael and myself (Tom) tackled that part of the challenge.
The electronics side basically have to decode the location data and then depending on which button was pressed it needs to to calculate the rotation of the motors (ie. the location of the destination) and send a message to the led displays.
Our biggest problem was the led displays. For our size we needed a lot and the few places that had them at an affordable price didn’t have enough. The suppliers that had enough were not cheap.
After some searching Andre came across a place that sold these car displays. They were on special (leftover stock from a previous bigger order). At R200 these were a bargain so we cleared out their stock.
Upon arrival I opened one up to see how easy it would be to hack it to our purpose. Luckily the electronics were pretty simple. Basically a micro controller with some shift registers. We desoldered the micro, soldered some wires onto the pads to connect it to an arduino. A bit of code later the led display was working.
We did however have some issues with getting the scrolling and font characters to work properly so for our demo we pre-programmed a few messages for the demo.
For driving the stepper motors we settled on a ramps 1.4 shield with an arduino mega (the exact same electronics you find in most 3d printers). Our original idea was to use a raspberry pi that would decode the data and control the ramps board. Michael wrote python code that would handle all that. Again, we ran out of time and decided to hard-code the data and process directly on the arduino mega.
Most of the other teams did a mostly software project (Except the Bushveld Labs/Isobar Team) and I think we scared them with our impromptu makerspace table.
I’m very proud of what we as a team managed to accomplish in those two days. And even though we didn’t win the prize money we had loads of fun, learned a lot and made some new friends. We have a nice project and we are taking it back to the space to make improvements and finish the parts we couldn’t at the hackathon. It will be a nice project to show off at the space and future events
Congratulations to #TeamBaby who won the Hackathon. We’ll see you guys next time!
Thanks to Kerry and the other peeps from Isobar and to Microsoft for the grand prize sponsor. Hope to see more hackathons with this kind of prize money 😉
The hackathon was held at the same building where the Diz MakerSpace is located. This place is awesome! If you in the area go say hi to Rick, Jarred, Daniel and all the others at the Space (111 Smit Street, Johannesburg)
A few days before the hackathon I put together a little event cam that would allow the other BinarySpace members to join in the fun. It’s basically a raspberry pi with webcam. On it was a modified version of our BinaryBot. All the members in our Telegram group could request a picture from it at anytime. There are some improvements to be made but this was a good test for it and you can expect to see our event cam at future hackathons and events.
It’s been three months since we originally decided to take part in the Global Space Balloon Challenge 2015. I got up at 4am on Saturday morning… even with a 3 year old I’m still not used to getting up that early. Took a quick shower and headed over to BinarySpace hoping … nay praying that I loaded everything we would need today. Today was the coming together of our HAB launch. How hard can it be? You take a balloon, fill it with helium, add a camera and let it go somewhere…
In the past months we learned about ‘Flexible use of airspace’ approval, helium calculations, ham radio, APRS trackers, flight predictions, etc. Launching a high altitude balloon is by no means an original idea so we spent quite a few hours researching the internet, on articles documenting other successful and failed launches. We also got advice from various groups in South Africa, that do this on a regular basis (HABEX, Habspace).
We were at our launch date. The goal: to launch a helium filled balloon, that should reach an altitude of 30000m and to then successfully recover it.
We used a 600g weather balloon, a parachute and a polystyrene cooler box for our payload. Inside the payload we had a GoPro camera, donated by Wayne Gemmel, a Trackuino aprs tracker that we built at the space, some batteries and a mintyboost kit to supply some extra power to the GoPro. We also threw in a hand warmer to keep the electronics functioning at the very low temperatures, found at high altitudes.
All the guys from Vanderbijlpark met together at BinarySpace and from there we left for our destination (which is about a 2 hours drive). Other members from the team came from Johannesburg and Witbank and we met them on site. We arrived at 7:15am and there was a group of people already waiting for us. Originally we were under the impression that the airfield was abandoned, but we found it closed off with security. A quick chat with them, explaining what we wanted to do, and we got in.
Since we wanted to launch as early as possible we immediately setup a spot and started getting everything ready. Wolff came prepared with a nice ground sheet for inflating the balloon and a table for preparing the payload. I started finishing up the payload while Wolff and Hanno took charge of inflating the balloon (with a few helping hands to make sure the balloon was handled carefully).
I started finishing up the payload while Wolff and Hanno took charge of inflating the balloon (with a few helping hands to make sure the balloon was handled carefully.
Preparing the payload involved, sticking down all our electronics with duct-tape, connecting the lithium batteries and starting the recording on the GoPro camera.
We also had to make sure that our space commander is secured on the payload.
Half an hour later Wolff and Hanno were ready with the balloon and we attached the payload to the balloon. At this point we were ready to launch, but had to wait while Michael phoned Air Traffic Control to get final permission to launch.
Once we received permission we did a quick countdown and let the balloon go. It was almost anti-climatic after all the work. We all watched it go up and disappear into the skies…
Minutes later someone told us that the signal is coming through properly and that even the guys in Johannesburg were getting the signal on their radio’s. We packed up and started the chase
Every minute a new APRS packet was received telling us that our balloon is going higher and higher but also heading into its own direction. It was moving fast.
40 minutes after our launch the signal fell silent. No pings received by any of the digi-repeaters. We kept driving in the same direction for a while waiting for the next position, but nothing came. We decided to stop for a bit, contact the other HAM guys and hear if anybody was receiving something. You could see on the faces of everybody that they thought this was the end. At least we launched it successfully…
15 minutes later we suddenly got a new position message. The balloon was still alive, but has flown quite the distance and it also went up another 10km in the time we didn’t receive anything. Time to move! We jumped into the cars and headed to the current position received. Unfortunately, our happiness was short lived and the signal fell silent again. We decided to head to the last position and then again would decide what to do once we get there.
This time it took about 30 minutes before we received a new message, but it also changed direction. We joked that it was probably heading back to the launch site.
The updates came more consistent and we watched our screens as the balloon went higher and higher. At around an altitude of about 30000m we were waiting for an indication that it had burst and was coming down, yet it still went up to 31338m and then to 32507m and then it started dropping … fast. It had burst and it was time to move again. The next 40 minutes consisted of driving fast over dirt roads, changing direction a few times and some cursing (mostly because balloons don’t have to follow the roads).
Eventually we ended up outside a farm. From the data it seemed that the balloon stopped descending. We stopped at an intersection, pulled out a different antenna and after a few received signals, we knew it was on the ground and somewhere on that farm.
10 minutes later the payload was found…
Apart from a few stickers that started peeling off the payload was in perfect condition, even our space commander survived the trip
From left to right:
Wolff, Sebastian, Matthew, Michael, Tom, Hanno, Wayne
Final payload weight was 630g
Final Nett lift was 960g
Final Neck Lift was 1590g
Lost Signal 08:44 at 11992.05m
Single Ping at 09:01 at 16510.71m (internal temp -3)
Proper Signal again at 09:30 at 24954.59
Highest Altitude: 32507.22m
Balloon Burst at 09:51
Flight Time ~ 2 hours 32 min
Predicted Flight vs. Actual Flight:
We would like to thank everybody who was involved, including the ham radio teams who helped with tracking, the various parties we pestered with questions and many more. We couldn’t have done it without you.
Ham Radio Operators who helped tracking on the Day
Matthew – ZS6MDV
Christo – ZR6LJK
Gert – ZS6GC
Willie – ZS6WBT
Francois – ZS6COI
Chris – ZS6COG
Judy – ZS6JDY
For the more technical details please see the project page on our wiki. We are busy updating it with all the details. We also started working on our next balloon project. If you would like to join us and/or sponsor our next HAB adventure then please contact us.
Update 15/04: Our next launch is in planning stages. We are currently applying for approval for the 15 of August 2015
I’m long overdue on a meetup update so here is a nice big post with things you might have missed the past month at BinarySpace:
Gary Immelman came to the space to give as an introduction into Amateur Ham Radio. It doesn’t take long to see that Gary is a very successful and passionate Ham Radio operator. He told us some very cool stories of his electronic and radio adventures. The people he has communicated with (even the Astronaut that came to dinner after a chat they had over the airwaves).
Amateur Radio covers so many aspects and there is lots of toys to play and experiment with.
It was a very interesting talk and inspired a few of the members (including myself and Michael) to start studying for the Amateur radio license exams happening in May.
We had a 3d printer themed evening with local ‘large scale’ 3d printing guru Hans Fouche as a guest. He brought along lots of 3d printed goodies including printed shoes, a vacuum cleaner and a 3d printed lawn mower.
Hans shared a few of his successes and failures (not a lot really) in his 3d printing adventures. If you haven’t heard or seen the 3d printed lawn mower in action then surely your living underneath a rock. It’s allover the interwebs!
Also check out the 3d printer that Hans built, it’s called the Cheetah, it’s fast, large and it’s awesome!
I’ve also showed off my scaled up 3d printed 3d racers.
Between all the meetups and hacking we have also been hard at work at improving the space. It recently got painted, were adding more work space, shelves for various hacks, projects and our beer glass collection. A new keg system is up and running. It’s still a work in progress but it’s getting better and better every week. Thanks to everyone who helped out and all the donations we have received.
Between all of this we still found time to check out the new Raspberry PI 2, work on our HAB project and build some Arduino game goodness:
We also have a few new regular members. Welcome and we hope you have fun!
Lots of other projects also happening, but sometimes I’m just having so much fun that I forget to take pictures, so sorry if I missed something. But then again, if you attend the meetups you won’t miss anything 😉 Also, check out the project page for all current and future projects.
We had some awesome meetups the past two weeks. Last week we had our ‘Etch your own Arduino workshop’ which went great. We used the Nanino design for the pcb and they turned out great!
For the toner layout we used the blue Press-n-Peel, its very easy to use and you get great results. For the etching we used ammonium persulfate which is not as nasty as the usual ferric chloride (although you still shouldn’t drink it!), we used gloves when handling the etchant.
We used a regular steam iron for the toner transfer onto the blank pcb. Not sure if the guys would appreciate it if their wives saw that they can actually handle a steam iron …
After some successful etching the board just needs to be drilled, cleaned and the components can be soldered onto the board.
Just about everybody is currently at this point and we’ll finish soldering up and testing the boards next week!
Last year we started working on our own Cyclone PCB Factory. We were missing some parts back then and the project got shelved for a while. Last night Gerhard and Romeo started working on it again. They managed to get all the mechanical stuff sorted and next week we are hooking up the electronics.
Sebastian showed off his Perspex entertainment pc which is coming nicely.
Myself and Sebastian are also currently working on a new monthly project for December which involves the new esp8266 wifi modules and some ws2811 LED’s. Sebastian gave a little demo of the code he has done so far. It’s going to be awesome … more info soon!
We had another great meetup last night, even though we were only a few. The keg system got a few bugfixes and seems to be running smoothly. The NeoPixel behind the tap now spins while pouring beer. We are currently investigating adding an RFID reader, but we’ll see how it goes. The actual top still needs attention as well and the electronics needs a cleanup. But it works great!
We recently got our hands a a numeric led display which didn’t work (or atleast we weren’t sure how it worked). We opened her up, tossed the old electronics and spent a good time reverse engineering the display (Sebastian did a good job of tracing all the signals). Once we knew how it worked it was a matter of putting a small test program on an arduino and giving it a go.
We were rewarded with an animation of all the segments being turned on one-by-one. Gys and Lionel have plans to make a wall mount bracket for it. As soon as it’s attached to the wall we just need to decided what to show on it, time? beer consumption? naughty words? Only time will tell
Lastly I powered up the pocket robot arm for a quick spin. This will be the first project in our planned monthly project evening. More info coming soon!
If you missed the meetup this week then you missed out! Felt like Christmas with all the different toys that ended up on the table.
Sebastian and Chris started working on the PowerHome project (our STM32 IOT Contest Entry). They made some good progress with the power monitoring devkit and at the end of the night they were monitoring the power usage of a heated 3d printer bed.
Lee rocked up with a bunch of new toys he received, including the new CubieTruck, an awesome usb LCD display and much more. I think we were all a little jealous, but Lee likes to share his toys so it was all good 😉
I also did a little demo of the new uMotio prototype and it turned it a little competition between Lee and Gys on who could do the shortest laps with SuperTuxKart. It was never officially declared but it looked like Gys was the winner 😉
We also have a new keg with Raspberry Delight beer so no one was thirsty.
So last night we had an extra fun meetup. We had an idea for a final xmas project before the year ends. It’s a Automated Beer Bottle Xylophone. We are planning on doing this over two evenings. Here is the result of our first test after last night:
Hi All, as promised we prepared a new post with all the juicy details of what was used to have BG send tweets all over the world. Originally when HelloComputer came to us for help, we had grand plans of creating new designs and circuit boards. Unfortunately the timelines would not allow for such a system (at least not for the original pilot). We then decided to use as much off the shelf components as possible. Great products such as the Arduino and the variety of shields available makes prototyping systems like this a lot easier. This system has some much more potential to be developed and expanded and it’s definitely something we are planning to continuously work on.
The concept behind BG the tweeting badger is basically that we added infra red motion sensors to BG’s enclosure at various places. As the sensors detect his movement it sends the data through to a base station which in turn sends it through to a server. The server handles the conversion from sensor triggers to tweets being sent. While we could tweet directly from the base station we wanted a system that could be managed easily in terms of keeping the tweet database up to date and making sure that BG doesn’t become a twitter spammer.
We tried to keep the wireless sensors as simple as possible. The main concern was that power usage since BG doesn’t have any electrical sockets in his enclosure (He is more of a reader, so he doesn’t watch TV).
Because you are planning on using electronics combined with little bits and bobs, you need to make sure that BG can’t get to it. Worst case scenario if he did manage to get his hands on it, it should be rugged enough that he can’t get inside the enclosure of the sensor and chew on any of the internal bits. The safety of BG always needed to come first.
We ended up going with a ABS molded enclosure with a clear lid. The enclosure is IP65 rated (We don’t want any rain to spoil the fun).
The wireless sensors were attached to various places in his enclosure. One of the challenges was that since BG is a pretty good climber, so we couldn’t just attached the enclosure to the side. The solution was to attach it to the poles with an aluminium arm that kept the sensor out of reach from BG. In other places, for example his bedroom, we could mount the sensors directly against the wall as long as it wasn’t within close range of any of his play shelves.
The hardware used in the sensors consist of three parts, mainly the infrared sensor, a wireless radio and a battery pack. The wireless radio is by default in sleep mode (to keep power usage low). When the infrared motion sensor detects movement it triggers a digital input on the wireless radio, which in turns wakes up and sends the signal to the base station.
For the wireless radio we used a XBee radio. These are nifty little radios which can be setup very easily, but they are slightly expensive.
The following was set up on the XBee radio’s:
– Change the communication to API mode
– Configure the DI (digital input)
– Configure the sleep mode
The base station is a big part of the project, since without it, the sensors would be useless. We also needed something smarter so we decided to go with an Arduino, easy enough to program and with the shields available it would be minimal hardware development. The base station has three main functions. It receives the trigger information from the sensors and handles the connection to the server. Again there was no power or IT infrastructure within close range of his enclosure, so we had to make it a solar powered solution.
Since we were using wireless sensors we could place the base station outside of the enclosure but still very close to maintain connectivity. We used a similar enclosure with the clear lid, but slightly bigger to keep all the bits together. Originally the base station was mounted on top of BG’s enclosure, but 5 minutes after it was mounted he managed to climb all the way to top and loosen the nuts and bolts that kept it attached. We promptly moved it to the top of the porcupine enclosure next door (they don’t climb that well).
The hardware used for the base station consist of an Arduino, a GSM shield and a XBee shield. Other bits include the solar panel, a battery pack and the solar battery charger (The lipo rider pro is awesome and so easy to use!).
The software for the base station was designed as a basic state machine. Depending on the current state it will either listen to the sensors, check the GSM connection or send through data to the server.
The GSM shield makes it very easy to send through data to a server. Once the initial connection is created, you can access perform simple HTTP POST and GET actions with the built in AT commands. Just remember to keep on the lookout for connection state changes so that you can reconnect the GPRS data connection if it’s not running.
The XBee side of things are simplified by using this XBee library. The main reason for using the XBee shield is that the wireless radio’s run on 3.3V and you need to do level shifting to connect it to the 5V Arduino. The XBee shield handles all of that.
The Arduino IDE makes it very simple to upload a program to the Arduino. The only thing to keep in mind is that keeping the shield plugged in while trying programming the Arduino does not work.
The system is currently running very nicely, but there is lots of room for improvements. At the moment the wireless sensors are rather dumb. They currently only receive the actual trigger and we would also like to add current battery levels in the reports. Most of the stuff is also mostly being kept together in the enclosures using duct tape and double sided tape. While it does make for a neat hack it doesn’t exactly look professional.
Now that the project is up and running we are going to take out time to design some proper electronics with new pcb’s and hardware. Another reason for wanting to upgrade the electronics is that prototyping equipment like XBee shields and Arduino boards aren’t really designed with low power applications in mind. Custom PCB’s designed strictly for the application could deliver anything from a 100% – 250% increase in battery life, as well as better management of the wireless sensors.
And finally we are in the process of designing some inserts for the enclosures that would keep everything in the right place and will make it more presentable. These inserts are currently being design in OpenSCAD and will be 3d printed on our in-house 3d printer.
Thanks to HelloComputer for getting us to do this project with them. It’s been lots of fun and we’re looking forward to more projects.
Have any questions? Want to stay updated about future BinarySpace projects? Check out our Facebook page, or join our mailing list
Not too long ago HelloComputer requested help on a new project that they were working on to get more exposure for the Johannesburg Zoo. When we saw the request, Sebastian Schocke and I (Tom Van den Bon) jumped on it. It sounded like an awesome project and we wanted to be part of it. Today we are proud to announce that the system went live and the Zoo has their own social media representation called BG. BG is the world’s first LIVE tweeting badger
Obviously we didn’t just hand BG a smartphone and trained him to use it (although that should maybe be a future project!), but rather we turned his home into a more hi-tech home. This consisted of outfitting his enclosure with Infra-Red sensors that would detect him performing various activities such as playing on his Jungle Gym or when he has retired to his room.
We will go into more detailed posts about the hardware and software used in future posts, but the basic concept consists of wireless sensors using PIR sensors and xbee wireless modules which communicate with a base station.The base station consist of a xbee/arduino and gsm engine.
When a sensor gets triggered it will send through a notification to the base station which will in turn send it through to a server which handles sending of tweets based on the data. This effectively turns BG into a Twitterer 😉