Generation 2.0 of RepRap, the self replicating 3d printer, is approaching realization. Code named “Mendel”, the new design will be wedge shaped rather than a box which offers a few benefits. The overall design is smaller than the original RepRap but the printable area is larger. This means more functionality with less building material. With each new generation of this project the assembly gets easier and total parts price drops making the RepRap available to a much wider audience.
The RepRap blog has put forth some design specs, a picture of the assembled prototype, and has links for general assembly information (ZIP). We won’t see a finished version of this released for a few months but so far it looks like a big leap forward.
[Knut Karlsen] put together a prototype set of solar rechargeable batteries. He always seemed to have batteries laying around on his worktable and figured they might as well be charging. The flexible solar cells were given to him by researchers at the IFE and are rated at 1.8V. He used superglue to secure them to the C cells. A silver conductive pen plus flat wires from a Canon lens connect the solar cells to the battery terminals. The batteries just trickle charge for now, but he’s going to try to build cells with built in charge controllers in the future.
The last time we checked in on Pandora it was just being shown in dev unit form. Embedded above is a video of the first case prototype. It doesn’t have any components yet, but it definitely looks like a good formfactor with a lot of potential. The Pandora is a Linux based portable game console with an 800×480 touchscreen.
[Alex] was frustrated by the amount of time it took to start prototyping with an AVR ATtiny. To make things easier, he built headers that carry the 8 and 20 pin chips and plug directly into breadboards. The boards include a 6pin ISP header, resonator, pull-up resistor, reset, and blocking caps. The ATtiny2313 version also has a serial connection header. This is a prototype though, and he forgot to route one of the connections. He plans on having a large batch of boards ready for next month.
If you haven’t checked out SparkFun Electronics’ prototype collection yet, you’re missing out. They unearthed many of their old prototypes and published them to show what kind of mistakes could be made. You’ll see plenty of errors and get hints on what to look for while developing your own hardware. This pairs well with their Design for Manufacture post. Along with the pile of broken board iterations, they also walk through how the company developed. Finally, they specifically cover the individual iterations of the BlueSMiRF.
One of the interesting modules in the gallery that never saw full release was the SparkFun Toys line pictured above. The individual units used the standoffs as the power and data bus. The four posts were arranged so they could only be connected in one orientation: power, ground, TX, and RX. It’s an interesting idea that seems like it might be worth exploring further. SparkFun says that it worked fine, but didn’t feel they had the resources to market it to the intended audience.
When you’re prototyping a new project, sometimes all you need is a switch. The folks at oomlout were tired of constantly having to rewire things, so they built a universal switch box for the Arduino. It has five potentiometers plus three switches. They’ve put together a software package that monitors the switches and can show you a live view of the knob positions. Have a look at the video below for a demo.
The writeup actually hints at what we can only assume is the next kit they’re releasing: a robot arm. Continue reading “Arduino switch box”
An article in EETimes suggests that we may see a memristor-based memory prototype in development as soon as 2009. The memristor is claimed by many to be the theorized fourth passive circuit element, linking the fundamental circuit variables of charge and flux. This news may not sound that exciting to most computer geeks, but this new component could usher in a new era of computer memory by forming the basis of RRAM (resistive random-access memory).
Scientists at HP labs have finally confirmed that the memristor behaves as their theories predicted. The reason that the component will work so well for memory is that the process is nonvolatile and the bits themselves will only change after the CPU tells them to. The bits in current DRAM systems slowly fade out and require a refreshment every 50 nanoseconds.