Robots and Dinosaurs tour a nuclear reactor and particle accelerator.
On Saturday members of R&D headed to south western Sydney on an excursion to ANSTO, the centre of nuclear science, research and development in Australia. ANSTO hosts Australia’s only nuclear reactor OPAL, the ANTARES particle accelerator, the Neutron Guide hall and it’s suit of instruments and many other experiments and facilities.
At the Visitor Centre we were greeted by Professor Proton who gave a fantastic demonstration on the basics of nuclear science and the design of the OPAL reactor. We then travelled by bus into the secure site where we inspected a number of instruments and facilities such as: OPAL nuclear reactor:
The tour included an intimate view inside the reactor with a high resolution remotely controlled panning/zooming camera.
Not used for power generation, produces no electricty.
Used as a highly controlled neutron source for scientific and industrial applications.
Neutron beams from this reactor feed the suite of instruments housed in the Neuton Beam Guide Hall.
Irradiates large crystals of silicon for the global semi-conductor industry in a process called Neutron Transmutation Doping.
Due to security restrictions we were not able to bring any cameras beyond the visitor centre so the pictures of OPAL, ANTARES and the Neutron Guide Hall were ruthlessly stolen from the ANSTO website. Many thanks to ANSTO for providing free tours of such an amazing facility and Tristan Steele for helping organise it. For information on ANSTO tours see here: ttp://www.ansto.gov.au/discovering_ansto/visiting_ansto/ansto_tours
We had/will/did have our time travelers halloween party on Saturday night. It was simply great! Everyone had a ton of fun showing off their awesome costumes, arguing over which interpretation of primer was correct, and whether or not we could have jumpstarted civilisation armed only with the content of our pockets! In the afternoon we did workshops on essential time traveler skills, such as how to make fire without matches, uses of a towel, and must-have survival kits for chrononauts of all ages. Nat brought in something we didn’t think of, a series of antidotes for everything from Bubonic plague to leprosy. (Very handy!) We also did our traditional watermelon carving again this year, and had a ball.
This week, I’d like to pay tribute to something we’re probably all familar with. The Kludge. “An ill-assorted collection of poorly-matching parts, forming a distressing whole” Now there are few words that have such conflicting connotations. (except perhaps ‘hacker’…) To some it’s the worst insult you can imagine. To others it’s a mark of pride to have their work called a kludge. In just the last week, we’ve had 3 great examples of elegant kludges that I’d like to share.
You can never find a relay when you need one:
For the last few weeks we’ve been getting our MakerBot working again. We machined our own MK-V extruder entirely in-house, lasercut the drop in paxtruder to give it extra grip on the filament, a new heated build platform and other nifty improvements. Many thanks to Kean and others, for their tireless hours on the mill and lathe:
Sadly when we went to put the MK-V heater under control of the extruder controller, I blew the solid state control that Kean had made. (Darn ground plane differences between power supplies. Grrr…). We have an official relay control board in the post on its way to us, but that could be a week or more away. This was the only thing holding us back from printing right now. So we decided to find the quickest way to get the extruder running. Rummaging around the junk room muttering ‘Relays, relays…’ soon produced a solution:
An old kids electronic kit, with relay onboard. A bonus is that spring terminals means no soldering was required, saving valuable seconds!
Anybody got a light?:
One of our regular two wheelers, Bec, got stuck in a bind when the headlight on her scooter blew on the way to the space. It was dark, the local service station didn’t have a suitable globe. What to do?
After scrounging in the hackerspace’s ample junk room, she found a solution. A halogen bulb! Soldering didn’t have much effect on the pigtails, so it couldn’t be soldered in directly. Hence the socket to hold the bulb:
It may not be pretty (or shall we say conventionally pretty), but it worked well enough to get home safely:
Returning a lost mobile:
Jeremy had a rare find when he stumbled across a touchscreen phone and camera in some shrubbery. Sadly the battery was completely flat, and we couldn’t tell if it was working. Rather than shell out for an expensive charger, which might end up being wasted if the phone was broken, we pulled the back off and had a look at the battery contacts. It used a 3.6V battery, which my old nokia does too. I pulled the battery out of my phone, and we were about to make some jumper leads to connect it up, when we realised. “Hmmm… the pin spacing there looks about the same.. I wonder if we can just tape the battery in place?”. It just about fit, with the end sticking out over the case. We tried to power it up….
Surprise! It worked!
And of course, being the good citizens that we are, we had a look through to find the persons number and see if they wanted it back. They were so surprised to get a call from their mobile after a week of it being lost. “Oh my GAWD! Thank you so much!” And we were able to return it the same night.
Score three for three on kludges! So the next time you hear someone discounting something as a kludge, you can ask “Yes, but did it work?”
Cool Thing of the Week -- AGM & PCR Machine Hacking
OK, a quick and hurried update, since we’ve been a bit rushed in the last week with our second Annual General Meeting.
The results of the Board of Directors were announced. The 2010/11 Board of Directors: President: Gavin Smith Vice-President: Terry Dawson Treasurer: Jeremy Apthorp Secretary: Kathryn Small Minister for the Interior: Chris Beckett Ordinary members: Bec Owen, Max Nippard and Kean Maziels Big thanks to our previous board members Greg and Marcus, and we appreciate all your support over the previous year. Lots of cool discussions were had, and we have many ideas for upcoming workshops and funding, etc.
Also, cookies were had!
PCR Machine Hacking:
We finally hauled our new PCR machine into the hackerspace and started tinkering with it.
It’s rather large. It took 4 people just to carry it into the space!
We’re still waiting for an exceedingly rare Mac compatible serial cable before we can use it, but we’ve powered it up and it seems to be working. I successfully burned my hand on the thermowell, so we know that’s OK!
Checking if the temp regulation is occurring More updates to come!
Ok, so we were sitting around thinking about our microwave. You’ve probably come across hot and cold spots in your microwave before. But where are they located, and how can we figure it out? Well, we figured that a good way to do it would be to use the microwave to heat something, and then image it with a thermal camera.
The photo on the right shows the microwave, with the target removed to be run under cold water between runs.
1x Microwave Oven (duh)
1x Ordinary HDPE breadboard, that you don’t mind cutting up a little
1x Ordinary A4 piece of paper
1x Ridiculously expensive thermographic imaging camera, borrowed (Seriously. This cost about twice as much as my car did.)
Ok, so the first rule to bear in mind is this. Thermal cameras will lie to you. I’ll say it again. Thermal cameras will lie to you. There are two ways in particular that I’ll cover, although there are a few more. Professional thermographers have an accreditation process and lots of training before they sell their services. Let’s take the photo below. There are two ways in which the image of yours truly enjoying a frosty beverage is inaccurate:
The first is that the scale of the image is only set from 12.1*C to 35.8*C. So Anything outside that region will not be displayed correctly. This isn’t too bad, and anyone who works with graphs or scientific images will expect that. The second way that the image will lie to us is that the measurement of the object depends on the emissivity of the material it’s made from. If you try and image shiny metal objects it’ll mostly capture the infrared light that’s been reflected off it, instead of the light it gives out iself. The emissivity of aluminium is about 0.09. So that means it’s mostly reflecting the temperature of the room behind the camera, instead of the can itself. The emissivity of paper is 0.93. So that’s pretty good as a target for our imaging.
We used a layer of wet paper to absorb the microwaves and get hot for us to image. In order to hold it upright we made a stand out of an old breadboard. According to this, HDPE is listed as ‘may be’ microwave safe (I’m not entirely reassured…). So we’re assuming it’s not absorbing the microwaves too much.
Bec holding the HDPE & paper target
Before each run we carefully ran the target under cold water for 30sec to a minute. This was necessary to make sure it was a uniform temperature across the whole plate. We then microwaved the plate for 10 seconds, and observed the temperature through the camera.
Run 1: 2cm from the back Run 2: 7cm from the back Run 3: 12cm from the back Run 4: 16cm from the back Run 5: 16cm from the back Run 6: 20cm from the back Run 7: 23cm from the back Run 8: 27cm from the back Run 9: diagonal Run 10: diagonal
So the question is naturally raised, how repeatable is it? Are the patterns random each time, or is it determined by the position in the oven? So we ran the same position twice. And between each run we fully removed the target and ran it under water to cool it down to a uniform temperature, replacing it back in the same measured position in the oven.
So, we’ve got a really nice, repeatable result for where the hot spots are in the microwave. Does this mean that we can now put in our hot pockets in the optimum spot, safe in the knowledge that it’s perfectly scientifically validated?
Well, sadly not quite. Putting an object in the microwave will actually change the standing wave patterns, since it’ll absorb and reflect microwaves. Still, it was interesting to gain an insight into what’s normally an automatic process.
Bonus: Makerbot Thermographic Porn!
At the same time we were doing the microwave experiment, Chris & Max were assembling our new heated build platform for the makerbot. So we thought we’d get some shots of it in action…
I was quite suprised how hot the extruder controller got. You can see the max temp is 68*C. (Sorry for the shoddy scaling in the image, anything over 35.8*C is saturated. The maximum is reliable, though)
So, here’s our idea for a regular segment at RnD. CTOTW. This is intended to be a place for use to visit the nagging questions that bug you that you’d otherwise ignore. Like ‘What is the density of a cow’, or ‘where exactly should I put stuff in the microwave to heat it up quickest?’.
Each week we’ll build a project, tweak something or set something on fire and report on it to you. And as Adam Savage says “Failure is always an option”. We’re not guarranteed to figure it out, or even get it working. We’re just going to hack stuff together (or apart) and have fun!
If you’ve got any ideas you’d like to see as a CTOTW, please send it to us!
The space is now equipped with (sort of) high-speed broadband, so we’re no longer breathing through the drinking straw of a dialup-speed Unwired modem.
Installing the ADSL was an Adventure™. For starters, I had to wake up at 6.30am to get to the space by 8am, which was when TPG said was the earliest time the engineer might reach the house. I arrived with coffee and Subway cookies, and looked around for the phone socket.
Um. There wasn’t one.
I peered around the front of the house, and saw aerial cables from the nearby telephone pole (which I later found out were completely irrelevant — Foxtel or Optus cables). 15 minutes of searching for a torch later, I ventured under the house with a floodlight (overkill is what we do!). I traced the aerial cables to cables under the house (my hair collected about 40% of the cobwebs off the beams). There’s a 4-wire cable, but it’s been cut. Ugh.
Half an hour later, the TPG guy shows up and crawls under the house with me to look for the cable. Eventually we found it, poking up into the board room with no socket on the end. He tests it, and to his great surprise (and my relief) it works. Clearly the line hadn’t been used for years, and for it to still be connected and working all the way through to the exchange is a small miracle.
Somewhere in the junk room a Netgear modem was found, and a 12V/1A power brick stolen from the Linksys router. The modem was configured, and Internet was had.
And there were megabytes.
No longer will hackerspace-goers be plagued by the lack of data or the irritatingly long wait for Youtube videos! No longer shall we resort to our variously crappy 3G connections! At last, we are free!
For those of you that haven’t been following, we attended Rocket Car Day 2010 on the weekend. This is an event where everyone cobbles/crafts/hacks together the most baddass vehicle they can, then crams a rocket engine into the back and competes for honor and glory!
Some people spent a lot of time on their cars:
Others less so:
Unsurprisingly at such a geeky event, RnD members were there in full force. We even had our own race!
(Note the swanky new lab coats, with RnD embroidery )
Beco Blocks, the 3d printable, snap together, ball and socket based building blocks.
Initially created as a prototype for another ball and socket based project, these blocks functioned so well I decided to make a number of variations to allow the creation of more complex objects.
Beco blocks are great as toys and i suspect might also find practical applications for things like:
Light duty hinges.
Printable helping hands
Internal skeletons for soft toys.
Snap together bracelets.
Mini pose-able art mannequin
Beco blocks thus far have been printed on our Makerbot using ABS plastic. The first Beco block printed was very lonely, being the only Beco block in existence with no friends to snap together with. Some time later when I finally learnt how to use the Makerbot myself I started printing more, and was surprised to discover how effectively they clicked together. There were practically no design iterations in making the original male-female Beco block, i was just very lucky with a ball and socket size ratio which seems ideal. The 3d models for Beco blocks include struts which allow the printing of the ball in a horizontal orientation. The struts can quickly and easily be clipped off using light wire cutters. There is plenty of room for improvement in the design and additional brick variations. I have only designed enough types of bricks to build a basic figure so i hope others add to the collection. The STL files are available at Thingiverse here: http://www.thingiverse.com/thing:1983
I came across the Super Probe two weeks ago, and just had to make one. I had a bunch of the PIC 16F870 microcontrollers left over from an earlier project, and the rest of the parts were all easily found in my parts collection.
It is a really simple circuit, with so many useful features:
- Logic Probe (L 3.7 V, P > 0.5us pulse)
- Logic Pulser (0.5 us pulses at 5, 50, 500, 5k Hz)
- Frequency Counter (8 digits)
- Pulse Counter (8 digits)
- Voltmeter (max 5 V, readings are a little high)
- Diode measurement (measures fwd voltage using 5V supply via 10k resistor)
- Capacitance measurement (1nF to 500uF, approx 100pF resolution)
- Inductance measurement (100uH to 999.99mH, but not very accurate)
- Signal generator (0.5 V square ware @ 500 Hz)
- NTSC video generator (white dot pattern)
- ASCII test pattern (A-Z then CR/LF at 1200, 2400, 4800, or 9600 baud)
- Midi test output (plays middle C on selected midi channel)
- R/C servo test output (approx 770 to 2300 us pulses @ ~28 Hz for servo control)
- Square wave output (1 to 9999 Hz 5V square wave)
- Pseudo random output (10 kHz PRNG)
- IR test output (38 kHz 50 % duty cycle square wave for IR receiver testing)
- PWM test output (6 kHz square wave with 3 % to 97 % adjustable duty cycle)
In addition to the PIC, you just need 4 common anode 7-seg LED displays, a 20 MHz crystal, and a few resistors, caps, etc.
I found the display to be a bit dim, but the MAN6610 7-seg displays I used were quite old and would be rated quite low in brightness. I left off the LM2931 regulator as I will just power it off 5 V, and so I actually put a PIC ICSP connector on the end of the board for re-programming, and an easy way to connect power.
We are looking to make up a PCB for this so that members can make their own. It is a bit of a pain wiring up the 7-seg LED’s (using the LTC4627 would help a lot).
Over the last couple of months, Gav and R3becca have been discussing a new project idea - the Geigerduino. The idea being to be able to build an simple device that can detect cosmic rays or other radiation using a Geiger-Müller tube and an Arduino, and then publish the details so others could do the same.
In late August, whilst having an R&D group dinner at Rockdale McDonalds, Gav quizzed me about generating high voltages, and I drew a rough schematic of a DC/DC boost converter (on a napkin of course). He told me the goal was to generate a PWM controlled voltage of up to 1000 V, using USB as a power supply. Having planted the seed, over the next two weeks I did some additional research and realised we should try a DC/DC converter using an isolated transformer with a voltage multiplier to get this kind of output.
So, two weeks ago, we began prototyping this at the space. I built up a simple circuit on a breadboard, using a small transformer, some high voltage diodes and capacitors, and a MCP14E5 mosfet H-bridge driver chip. I had acquired a small batch of these transformers from a surplus electronics place a few years ago, and they seemed appropriate with a DC resistance of 11.5 ohms on the primary, and 2,300 ohms on the secondary. I had no other specs, but I think they were intended for generating HV to drive EL or CCFL lighting. The MCP14E5 is actually intended to drive the gate of power MOSFETs, but I had it handy, it is fast and can supply significant current, and it seems to do the job.
I also needed a signal source to drive the transformer, and seeing as R&D had recently done a group purchase of Bus Pirates (preorder 2), I decided to use one of those. The Bus Pirate has a PWM generator mode that can generate 1 kHz to 4000 kHz signals with varying duty cycles.
So, what was the result - first we tried 1 low frequency, possibly 10 kHz, and got almost 500 V DC out.
A bit of tweaking, and we found that these transformers work best at around 50 kHz. This picture was taken when we were using about 8 V at 200 mA to drive the transformer.
Yes, we are generating nearly 800 V DC using USB and a simple breadboard circuit!
Gavin and I then made up these circuits on some prototype PCB’s, and found we could generate 600 V DC quite nicely from USB power. Gavin even incorporated his Arduino DangerShield for PWM control.
This post is to describe a particularly epic piece of work by one of our members, Bret. He’s built a full-sized flight simulator cockpit in his house! Here’s his description:
The simulator was inspired by a visit to a facility in the north of Sydney which maintains and operates some impressive aircraft (but it would be improper to disclose who that would be :-) )
During a flight in one of their sims, I began to abort a landing since it just wasn’t quite right, and the instructor gently told me that I’d make it in. Only afterwards did I consider what had happened. It was so realistic that my natural piloting sense told me I was at risk, and to get out of the situation as I had learnt (namely to make the engines make a lot more noise, and go around to have another go). My later realisation was that I felt so immersed in the experience that I was as good as flying a real one! So, what other response than to decide to build my own (albeit not quite so complex or expensive)?
The sim is reasonably closely based on a Boeing 777, partly because the www delivered me some sort of ok dimensioned drawings which I was able to work from. Also, the 777 is all glass cockpit, making the instrumentation easier to manage. Finally, the appearance of a 777 is quite close to a 747-400 so I can always make a change in the future if I need to! Boeing or Airbus? Well. knowing a good few Qantas engineering types, they tell me “if it’s not Boeing, I’m not going!”. Decision made!
Technically, the sim is MS FS2004 (FSv9) on one pretty grunty quad core machine, running the flight model and front window view, and the IO and displays for the various parts of the cockpit are shared between 3 other win PCs. Video is the limiting factor for a sim I have found, as long as the main flight computer can handle the “visual experience”. Various freeware applications interface the systems and the aim is to provide a close to authentic experience without ever needing to “touch” a PC, only via the normal aircraft interfaces.
The project started just after Christmas 2008, and progresses in fits and starts depending on how busy life is and what I need to buy next. It is intended to always have it flyable, so that I can’t lose interest!
Recent flights have been Sydney to London and return (flying at the same time as some friends going on holiday) and the weather that is “provided” is very realistic. I landed in thick fog in London (simland) and my mate told me a day later that the real aircraft had diverted to Rome because they wouldn’t be able to get into the real London. That made sense, because I pretty well found the sim runway by braille!
The MakerBot Assembly Day was a rip-roaring success! Thanks to all those who attended and helped our bot crawl into life.
It took us quite a few hours, and a little cursing, but eventually we’ve got the 3D printer moving and extruding to our hearts content.
There’s a lingering bug with build ‘burping’s, but we do have to same something to work on for next weekend.
Max gave a demo of his quad-copter
Phi started work on a furnace
Adrian demod his ServoShield units
Jeremy gave a demonstration on how spontaneous symmetry breaking is done
Member's Projects -- 'Light Interrupted' by Adam Synnott
Introducing a new category here on the blog, Member’s Projects. We’d like to show off some of the things people are making, taking apart, remaking or whatever! Our first project is by Adam Synnott, and is called ‘Light Interrupted’. It was recently featured in ‘Serial Space’ in Chippendale.
Light Interrupted is a rear projected FTIR multi-touch display using a combination of Touch Designer and Max/MSP the screen. Adam is currently in development of a muti-touch 3d painter/modeller, an RSS fed globe, some reactive grass and a 3d fish tank.