CCHS GGHC: CCHS element14 group

Team moment of the week

We had a great evening last night rounding out the project, with most of the team members able to join in on the last tasks to complete the project. There were many highlights during the evening, including getting the final product working in its nice shiny enclosure:

however the highlight of the night was team member Rob's unveiling of the secret experiment: a Ruben's Tube:

we've designed OpenLab's hardware so that it will be able to drive a Ruben's Tube from the available outputs:

Key Thoughts and Observations

We considered a number of times whether our ambitions had gotten the better of us, and perhaps they had. But in the end, we've gotten to a first alpha prototype product, with lots of potential to expand it. This is by far the most feature rich and complex project we have undertaken together as a hackerspace. It stretched our skills as a group and as individuals. We found that our core team members were severely taxed, and it took the efforts of the entire team to get us to a final working product. A stellar effort from all the crew: congratulations everyone, but particular thanks go out to our friends at Freetronics for their support and assistance, and to the team: Andy, Luke, Marc, Stuart, Antoinette, Bernd, George, Rob and Clifford for your hard work and commitment.

We're very excited by what we've built, and the potential to improve it going forward. We're already looking at presenting this to the local Science Teachers association.

Next Steps and post-project activities

There are various activities we intend to investigate, including:

     - Organise further school trials.

     - Establish a way for schools to purchase the project in various forms, including as a partially built kitset, or as a complete finished product ready to use.

     - Present to the local Science Teachers Association conference

     - Improve the hardware in second version update

Spotlight: The OpenLab

Our finished product is a general science and physics project platform, usable by schools to more easily conduct and monitor various experiments. Our working title during the competition was ArduinoLab, but as you can see in the pictures we've now settled on the name OpenLab.

The OpenLab platform provides a number of things:

- A damage proof, protected, experimentation environment where the circuits are designed to protect the inputs and outputs from damage.

- An onboard Lithium Ion battery, providing mobility and portability of the lab around the classroom.

- Charging of the portable power source over USB, enabling the OpenLab to be powered and charged from a computers USB port, from a multi-port USB hub, or from mains power using an adaptor wall plug with USB socket.

- Graphics LCD display, enabling waveforms to be displayed.

- Intuitive navigation using buttons and control knobs, using a soft-label menu system.

- A built-in Arduino: an Atmega 28p, lowering the cost of purchasing the microprocessor separately.

- A Xbee header, enabling the addition of communications features using any of the "bee" form factors modules: Xbee, GPSBee, RFbee, BluetoothBee, etc.

- A multi-tasking screen/ application handling framework.

- The ability to monitor multiple sensor inputs and to generate multiple outputs simultaneously.

- based on a design using easy to source and well known components, enabling the project to be replicated internationally.

Documentation

We'v used this blog pretty much exclusively for our updates and information on our project.

We've also uploaded some video to the agentsilverfox FTP server, so keep an eye out for that footage appearing at some stage.

In a few weeks we'll have a project entry updated on the Hack Melbourne Project Page

If you'd like to follow our progress on the software, have a look at the software repository here on github.

And github also provides a walkthrough of the software structure, use, etc.

And if you want to review the hardware work, or see where version two goes next, see the previously mentioned hardware design repository here on github.

And and walkthrough of the hardware structure was in a previous blog post.

Closing comments

Well, it's been tiring but it's been a real blast! Congratulations to all the other Hackerspace teams who have joined in the challenge: we've enjoyed reading about your projects and your progress. We hope you've enjoyed hearing about ours just as much.

And if you're ever in Melbourne, come and look us up. And if you can't make it in person, drop by the website from time-to-time to see what we're up to, or look us up on IRC.

Happy Hacking from the Connected Community HackerSpace (Melbourne)

http://hackmelbourne.org

CCHS GGHC: CCHS element14 group

Over the last couple of weeks, the electronic design for our microcontroller-based electronics experimenter's instrument has been completed, I've completed the PCB layout and sent the CAM files off to the PCB fab, we've sorted out the BOM lists, and we've ordered a whole bunch of components from our friends at element14 and Digi-Key.  We've sourced essentially all our components from element14 and Digi-Key... but these vendors aren't the only vendors that sell the components needed... these are almost all very common components available from many suppliers.

Above: The PCB layout of the main board.

Right now, tentatively, its working title is the ArduinoLab. (I'm considering calling it the Mystik Spiral project... because we're thinking of changing the name. )

So, I thought that now, while we're waiting for the delivery people to bring our PCBs after the Easter holiday, might be a good time to look at the electronics designed by Marc Alexander and myself, and go over how it all works.

Microcontroller

At the heart of it all we have a microcontroller... the Atmel ATmega328. This is undoubtedly a familiar microcontroller to many readers. It is, of course, the same AVR microcontroller used as the basis of the Arduino system, in most cases. This widely available, popular microcontroller is surrounded by several other core components of the microcontroller system on the system's main "motherboard".

To give us the most flexibility when it comes to packaging up the finished PCBs into an enclosure, and because I'm using the free "light" license for Eagle, which limits the PCB dimensions, our best option actually turned out to be dividing up this system across several PCBs, instead of putting it all on a single monolithic PCB. We'll take a look at these peripheral PCBs in a moment.

The AVR is running at a clock frequency of 16 MHz... the same most Arduino devices. This board is intended to be fully compatible with the Arduino system - the Arduino bootloader and the Arduino IDE - for easy programming and software development. And of course, just like Arduino, everything is open-source too, for hackability, repairability, and expandability, as well as for the educational value of open-source. We're writing firmware for the device based on the Arduino system, running on top of the Arduino bootloader, so the firmware is easy to understand and easy to change.

Because this board is almost entirely based on compact surfact-mount components, a QFN-32 version of the ATmega328 ATmega328 is used. This has the added advantage that it gives us a couple of extra I/O pins - the AVR's 7th and 8th analog inputs. These inputs are present on the chip die, but they are not broken out to pins on the DIP version of the chip.

Included on the motherboard is the popular FT232RL USB virtual-UART IC, just like on the Arduino Duemilanove, providing USB connectivity for programming and communications with your PC. We've also included a reset button for the AVR, too. A USB mini-B connector is used for the USB interface on this board instead of the traditional USB 'B' connector, which is excessively tall and bulky.

For even more expandability, flexibility and hackability, I've also added some pads surrounding the AVR, in a standard Arduino "Shield" footprint. This provides lots of exciting hacker-friendly potential for expansion and access into the microcontroller pins!

Power Supply

So, basically, it's just like an Arduino, right? Well, not exactly. There are some further features we've implemented in this case, whilst still maintaining a high degree of Arduino compatibility.

To start with, we need some portable power. I've included a 3.6 V lithium-ion AA cell on the board, for convenient wireless power and portable operation, with no wall-warts, power cables or batteries dangling off. A Microchip MCP73831 Li-ion charge management IC is used to charge the cell from power supplied by a USB host device, or a 5 V mains power supply (a plugpack or wall-wart) with a USB connector.  There is no need to remove the battery and plug it into a battery charger - you simply plug it into a USB host, and it charges the battery. User friendly, no-fuss, uninterrupted portable power that just works - just like your laptop or cellphone.

The lithium cell's nominal 3.6 V is stepped up to 5 V using a Texas Instruments TPS61202 buck/boost converter IC. This allows us to run the microcontroller's positive supply rail at 5 V, which is a common, familiar choice for most Arduino-like devices, and gives us a little more flexibility than a 3.3 V supply rail. One small disadvantage of this device, however, is that it is a very small QFN-10 SMD package. Probably going to have to bust out a stereomicroscope for this one.

Since the MCP73831, the FT232RL and the TPS61202 are only available as surface-mount devices, we could really make use of the extra 2 ADC channels on the ATmega328 ATmega328, and we need to fit a relatively large amount of electronics onto a relatively small PCB, we have decided to make the system's motherboard almost entirely built from SMD components.  For realistic electronic engineering projects today, with nontrivial devices, one needs to learn to stop worrying and love surface-mount technology, because you just can't get many interesting and valuable ICs in DIP packages.

Graphical Display

Next up, we've got the LCD display. Here, we've decided to step up a little from our traditional choice of the common HD44780 character displays, and we've chosen a nice little 128 x 64 graphical LCD based on the common KS0108 chipset, with white images on a blue backlit background. In fact, with a little shopping around in Asia, we've found that these graphical LCDs are no more expensive than similar 20x4 HD44780 alphanumeric LCD display modules.  But we don't have very many spare pins to play with on our AVR... so how are we going to wire up this LCD display, with its 8-bit data bus and several extra control lines compared to a traditional HD44780? In order to do more with less microcontroller pins, I've used a cascaded pair of cheap and readily available CD4094 8-bit shift registers (or HEF4094 HEF4094... or whatever. You could use any 4094 from one of the different digital logic families), to let us latch the data and instructions out to the LCD display in a serial fashion, using relatively few microcontroller pins.

We're going to write some modified KS0108 libraries which are compatible with existing Arduino libraries for KS0108 displays, so they can be plugged into existing programs, but which drive the LCD via the shift registers. This is relatively slow, but saves many microcontroller pins for other uses.

The KS0108 128x64 LCD displays we are using are essentially the same as those used for ladyada's delightful MONOCHRON clock. This gives you a good idea of what they look like:

(This image belongs to Limor Fried and Adafruit Industries. Used in good faith. Thanks!)

Lots of interesting peripherals

A header footprint for a Digi XBee 802.15.4 module (or any compatible 3.3 V serial module, such as those GPS, Bluetooth and RF communications "bee" modules available from Seeed Studios) and a corresponding 3.3 V regulator are also available on the board, as is a piezo buzzer for audible feedback to the user.  We've also got a digitally-controlled adjustable DC voltage output. MOSFET Q1 is switched by the microcontroller, and the resulting PWM waveform is smoothed out with the big-ass capacitor C2. Shunt resistor R11 lets us use the AVR's ADC inputs to measure the current flowing in the output circuit. This simple low-current power supply is really convenient for all kinds of electronics and physics experiments. Suppose that we want to measure the V-I curve of a diode. Well, over a certain range at least. We can program the system appropriately, attach the diode, and sweep the positive bias voltage from 0 to 5 volts while we're measuring the current flowing, and we can send the result to a plot on the graphics LCD! (Well, theoretically at least. I haven't tested this yet.)

Analog Data Acquisition

We've also got a couple of potentiometers for analog input from the user, as well as a couple of terminals for analog voltage measurements from external circuits or devices. To make sure that nothing gets blown up by students who are experimenting, these analog input ports are attenuated in a 4:1 ratio by voltage dividers, and clamped with 5.6 V zener diodes, making these inputs nicely hardened to prevent damage to the unit. The main 5 V rail is also clamped to 5.6 V by a 5 W zener diode to make sure that the ICs cannot be damaged if an excessive voltage is applied to the 5 V rail. Because the ADC inputs are attenuated in 4:1 ratio, the maximum analog voltage that can be measured by these analog inputs is approximately 20 V.  One of these digital inputs is connected to both a digital input on the AVR and an ADC input, so we can do things like pulse counting and digital data recording in the digital domain, as well as the measurement of analog signals. Lots of possibilities for experiments and educational activities!

Lots of buttons to push

We've also got some other bits and pieces, on the peripheral PCBs.  We've got a set of six pushbutton switches, for user input, control and navigation through the software menu system displayed on the LCD. These six pushbuttons are connected through a string of resistors, forming a voltage divider with an output voltage that changes as different buttons are pressed. By reading this voltage with of the the AVR's ADC inputs, we can receive input from a whole bunch of pushbuttons using only a single microcontroller pin.  This simple pushbutton pad is on its own little PCB - and this design is one that I suspect will probably have future value for all kinds of different microcontroller and Arduino projects. This board is connected back to the main board using three wires - the 5 V rail, ground, and the analog output. Since the component density is not especially high on this board and no specialised components are used, common through-hole components are used exclusively on this board, and it is a single-sided board, making it relatively simple even for beginners to solder together, adding another aspect of potential educational value.

Digital Outputs

The final one of the three PCBs in this system contains three digital open-collector transistor outputs, and an output for analog waveform synthesis. The main reason that a separate PCB was used here is because of the limitations on board dimensions in the free-as-in-free-beer "Light" edition of CADsoft Eagle, which I used to lay out the PCBs.  Three of the AVR's PWM-capable output pins are bought out to three BJTs, with 1 k pull-up resistors that be switched on and off, allowing either digital outputs or true open-collector outputs. 1N5819 1N5819 diodes are used in the pull-up part of the circuit for protection of the circuit, preventing the 5 V rail from being pulled up too high if the pull-up switch is closed and an excessively large voltage is applied to the output. An 18 ohm resistor is included in the emitter to protect the transistor from overcurrent, and a 1N4004 diode is connected to the output to protect the transistor from back-EMF from an inductive load.

Analog Synthesiser

One of these PWM output channels also has a second function - it is also used for analog waveform synthesis. An input PWM signal is filtered through a second-order low-pass RC filter, creating an analog waveform. A second-order Chebyshev LC filter might have a bit sharper transfer function here, but resistors are much cheaper and smaller than inductors, as a rough general rule, and the RC filtering gives good enough performance characteristics for our purposes.  To synthesis sinusoidal waveforms, for example, a sine lookup table will be pre-computed and incorporated into the microcontroller firmware, avoiding the relatively computationally intensive task of calculating sines in realtime.

The analog output is then buffered with a voltage follower, an Burr-Brown OPA340 rail-to-rail opamp. A rail-to-rail opamp is really valuable here, since we only have a  unipolar 5 V power supply. An AC coupling capacitor can be switched into the circuit to remove any DC offset - or bypassed if some DC component is actually desired. Another toggle switch is used on the output side in order to choose between either the analog synthesis output or the transistor output.  Hopefully these features, along with other hardening and protection measures in the circuit, make this device fairly hard to break electrically.  Like the pushbutton input panel, the nature of the components on this board and the density of these components did not justify the use of surface-mount components, and it uses only through-hole components, making it easy to assemble, even for beginners.

I've included portions of our hardware schematics in this post... but of course the entire thing is open-source. If you want to get the whole lot, in full, glorious high-resolution, feel free to check out the contents of my repository for this project over on GitHub - you'll find everything there. If you're not familar with git (and GitHub) it's a great thing to learn - it's not just for greybeard programmers.

Happy Hacking,

Luke Weston,

Connected Community Hackerspace, Melbourne, Australia.

http://hackmelbourne.org

CCHS GGHC: CCHS element14 group

Team moment of the week

We had a number of great moments this week.

Team member Andy was invited to join gmc and friends for an interview about the Great Global Hackerspace Challenge on the /dev/random show that broadcasts from Revelation Space on the Signal radio station. For those who don't know, Signal is dedicated to hackers, hacking, hacks and everything related. Team members Andy and Paul teamed up for a fun, relaxed chat with gmc about the challenge.

If you'd like to hear the interview playback, you can find it at the 2011-04-14-2200 entry in the archives. We're on a little after halfway (at about 33:15):

  http://signal.hackerspaces.org/archive/

Another big moment was the satisfaction of finally pronouncing "Go!" on the PCB production run. A small run of 8 boards, but still a satisfying milestone, none-the-less . A big thanks to team members Luke and Marc for getting us there.

And finally, a few of the team attended the ever amazing Trampoline Day, and team members Andy, Paul and Luke gave a talk on the Great Global Hackerspace Challenge and Hackerspaces in general. About 15 or so people came along to hear all about it and ask questions. Thanks to all those who attended, and Wade for the photo

Progress this week

Project: Due to "other life" commitments for various team members, we found ourselves slowing down this week (counter to our plans). We also found that the Easter long weekend introduced a number of headaches for getting PCB fabrication and parts supply completed to schedule. We got the PCB production under way, the bulk of the ordering completed, discussed existing software we could reuse and new work that would be required, and planned what we would need for some concerted build activities over the Easter long weekend.

Requirements: We reviewed what would be required to create a couple of the experiments we'd like to use to demonstrate the project.

Design: We made a final round of revisions to the PCB design, and sent the gerbers off for fabrication.

Build: We discussed the approach to a hardware rig to test the graphic LCD. We ordered the LCD's earlier than other parts, so expect they'll arrive early this week in advance of the PCB's. As soon as these arrive, we'll get the test rig up and running and start exploring a design for operating the LCD via the shift registers.

Plan & Schedule: Our main focus has been bringing together all the things needed to work through build and test activities over the Easter break. We've also needed to adjust our schedule to fit with the availability of team members. We've certainly felt the pinch of  "real world" commitments cutting into our project schedule.

Key Thoughts and Observations

We had the importance of local hacker support networks reconfirmed a couple of times this part week, both in terms of working together in group review of the PCB layouts, but also in overcoming the various problems associated with a short time frame, limited size PCB fabrication run. A number of folks in the extended community stepped in to offer advice and helpful contacts to getting done what we needed. In particular, team member Marc Alexander provided key advice and assistance through his company Freetronics in meeting our PCB fabrication and component sourcing challenges.

It's amazing the wealth of skill and experience a hackerspace can bring to bear. We're honored and extremely grateful to be able to work with such a diverse and supportive crew.

Rough Milestones for next week

For week five, we’ll be working hard  on breaking the back of various activities, including:

     - Assembling and exploring prototype LCD hardware rig

     - Organising and running at least one  surface-mount build day

     - Concluding the purchase of remaining components from local suppliers

     - Building the first set of ArduinoLab boards

     - Software development

     - Draft enclosure layout and assembly

     - Sourcing components for the demonstration experiments

     - organising some school trials in the final week of the competition.

Spotlight

No spotlight this week, but we'll try to follow up with some additional guest posts on specific topics from some of the team members.

Stay tuned for more updates!

Happy Hacking from the Connected Community HackerSpace (Melbourne)

http://hackmelbourne.org

CCHS GGHC: https://www.element-14.com/community/groups/challengedownunder

Team moment of the week

We got pretty excited by one of the experiment suggestions provided by our teaching contact. It was something team member Stuart had suggested during our initial brain storming, and at that stage we pretty much laughed it off as too crazy. Now that our stakeholder has specifically mentioned it, we'll investigate whether we can pull that together as our main demonstration experiment. At this stage, we'll offer the obscure hint that it's a pretty hot number - feel free to offer your best guess in the comments. Stay tuned for further details!

Progress this week

Project: The feedback from our secondary education contacts reinforced the usefulness of the primary project we'd chosen. We're now focused on a completed solution as the main goal for our entry, rather than a development kit requiring soldering (something we'd considered earlier). We had a great session Tuesday night, taking feedback we'd received from our teaching contact and exploring the design.

Requirements: Our teacher stakeholder provided a number of ideas for lab experiments that we'll select from as validation and test scenarios, and to showcase our entry. One of those is particularly spectacular.

Design: We considered some alternative components, including replacing our initial choice of a 20 char, 2 line LCD screen with a Graphic LCD, and exploring a couple of battery charging IC options. We're thinking of using AA profile lithium ion batteries for portable power. After much discussion, we've decided to push ahead with a predominantly surface mount design.

Build: We've decided on surface mount for most components, and have begun the task of identifying specific components and reliable manufacturing sources. We're planning on using the project as a way to train a number of team members in surface mount assembly.

Plan & Schedule: We discussed a couple of schedule options around PCB manufacture runs that would put us in a good position to focus on test and validation in week five.

Key Thoughts and Observations

Besides the mystery experiment, probably our main design thought this week was to move away from a"standard" 20 char, 2 line LCD, and use a 128x64 Graphic LCD instead. This gives lots of possibilities for displaying soft labels, as well as waveforms. The part is a more expensive option, but we hope the added benefits will justify the expense, that other design choices will offset the additional cost, and we think we my have found some more reasonable suppliers. We're investigating the practical implications of that LCD move prior to making a final decision.

Rough Milestones for next week

For week three, we’ll be focused on:

     - Finalising the PCB artwork to send off for fabrication

     - Outlining the core software components

     - Deciding on the peripherals required for build & validation

     - Ordering components

     - Sourcing components to set up a prototype hardware rig

     - Starting software development

Hardware Design Overview

We've settled on the ArduinoLab, a project initially conceived by team member Marc Alexander. The basic idea is that ArduinoLab is a general lab experiment tool that is tolerant of user error and can perform basic functions, such as measuring voltage, frequency, and generating waveforms, sound and PWM output. We're aiming for two inputs and two outputs. Were using an ATMEL AVR as the base microcontroller to provide Arduino compatibility. Initially, we'd considered an Arduino shield approach with a standard Arduino off board, however integrating the AVR helps keep the cost down. We'll provide standard "Bee" headers (and probably standard Arduino shield headers) to enable additional expansion possibilities. We intend to use USB to both program the AVR (as needed), but also to power the unit to recharge its internal Li-ion batteries. Optionally, mains power to USB wall wart can be used to power the ArduinoLab for longer running experiments.

It's targeted for use in a school setting, mainly secondary school however its error-tolerant design may make it suitable in primary school settings. We are aiming to keep the component costs as low as possible, while providing as much functionality as possible. We'll certainly have a versatile experimenters lab that offers lots of bang for a fistful of dollars .

Here's an early mock-up of a faceplate for ArduinoLab, showing where user interface input controls and the display might be:

We have some busy weeks ahead of us. Stay tuned for more updates!

Happy Hacking from the Connected Community HackerSpace (Melbourne)

http://hackmelbourne.org