After quite some time without development activity on VoIP32, I began investigating modifying Retro BSD to run on it.  Retro BSD is a port of 2.11BSD Unix to the PIC32.  It operates by loading the Unix kernel and hardware drivers into the PIC32′s Flash memory, then loading individual executable files from an SD card and executing them from RAM.  Of particular note, the project includes a BASIC compiler and a MIPS assembler and linker in the default filesystem image, so development for the system can happen on the system itself.   While the source for a C compiler is included, it is too large to run from RAM and is currently not operational.  Still, to say that the Retro BSD project is impressive is an understatement.

Due to the hard work of Serge Vakulenko, who started the PIC32 Unix effort, modifying the project to run on new hardware is as simple as editing a couple of Makefiles.  After doing so, using MPLAB to download the kernel to VoIP32 was pretty straightforward.

The project includes drivers for an SD card connected via SPI, and supports the use of serial or USB for the console connection.  Over the weekend, I was able to modify my existing HD44780 LCD driver and graft it onto the Retro BSD kernel.  The LCD appears as ‘/dev/lcd’ in the Unix filesystem, and writing to it is as simple as piping text to that file.   As time allows, I would like to add simple networking functionality using the board’s onboard PHY.

Retro BSD is an extremely impressive project, and I think having a real operating system like Unix running on our hardware has the potential to greatly increase its value.  We’ll be sure to write more about future successes with this software.  It’s awesome!

As you may know, one of the projects in the pipeline here at NBitWonder is a software-defined radio receiver, loosely modeled after Jeri Ellsworth’s receiver published earlier this year.  As of this past weekend, NBW-SDR (as we call the project) is operational!

The video features a live reception of WWV, one of the United States’ atomic clock stations (much more information here).  The station is located near Fort Collins, Colorado, some 2000 miles from the receiver’s location in Virginia.

There is much more work to be done on the software, as you can tell from watching the video.  For instance, only AM reception is implemented right now.  The signal processing functions still need some work, and more appropriate gains should be chosen for the entire signal chain.  Still, it is rewarding and reassuring to see the project working at some level.

Be sure to watch the github repository and the forum for much more information as project development proceeds.

Recently, the first issue of Open Hardware Journal was published, to the largely unheralded glory that it deserves. It features many interesting articles, but in particular, features a fantastic article written by Christopher Olah on the 3D printing of optical elements, specifically lenses.

I was unable to find photos used in his paper online, but an email to Chris and he generously sent the images my way, under a Creative Commons Attribution license (thanks Chris!). Those photos have been added to NBitWonder’s flickr account for all to access, and attributed/licensed accordingly.
(continue reading…)

I took a couple of hours tonight to put together the Software Defined Radio PCB, as the parts arrived a couple of nights ago.  Software development has already started — check out the git repository for the progress so far.  Be sure to watch the project in the forum to stay abreast of project progress!

Introduction:

Ah, Halloween. Next to Ramahanaquwanzmas, one’s birthday, and a handful of other possible days, is there a better day of the year? Not only is it socially acceptable in this small time window to walk around in public dressed up as a robot, vampire, etc., but if you’re maker-savvy you get to build something fun! I enjoy Halloween a great deal, and have gone to some pretty impressive lengths in years past to get a kick out of the holiday, and this year would be no exception!

I’ve always had a thing for Iron Man. He’s often hailed as the engineering superhero, and any super hero that can be seen on the big screen holding a soldering iron is grade-A material in my book. With this in mind, I set out to build an arc reactor, like the one seen in Iron Man.
(continue reading…)

Over in the forum, we’re designing a software-defined radio modeled after Jeri Ellsworth’s Hackaday post from earlier in the summer.  We ordered PCBs for the first revision perhaps a bit too hastily, as some forum members with valuable experience in the area found a number of shortcomings in our design.  Those have been corrected (we hope!) with this latest round of updates.

Changes that have been made include moving to a better oscillator module (wider frequency range, better stability, less phase noise) and to an external 24-bit ADC (for higher dynamic range).  We have a few boards of the old design already fabbed, but things like work and school keep getting in the way of building up a prototype or two to start coding.

Check out the schematic and layout over on github (or just grab the latest NBitWonder Eagle Library update), and let us know what you think in the forum!  We might not always respond right away, but we value your input.

An updated version of the Class-D Amplifier project is being worked through the pipeline as we speak.  It features a number of improvements over the breadboarded version, the greatest of which perhaps is the fact that it’s not on a breadboard.

You can see more information regarding its development in the forum (registration required to see the thread) or in its github repository.

The PCB pictured above was ordered a few days ago — be sure to follow our progress on the project to get any and all updates about its development!

For a long time now, I have put off breadboarding a Class-D amplifier.  A Class-D amplifier operates the output transistors as switches instead of as variable resistors, as in more traditional amplifiers.  In doing so, nearly all conduction losses in the output power devices are eliminated, leading to a very efficient amplifier.  Much more information is available at Wikipedia.

This circuit is based on a design by Johan Sorensen, who developed this circuit over a period of time in the early 2000s.  His work appears here and is a great read on the practicalities of Class D amplifier design. Notable differences between his circuit and mine center around my use of a MOSFET gate driver IC instead of his discrete component solution. His solution has the advantage of working at higher voltages and (likely) with cleaner, more precise switching waveforms; mine has the advantage of simplicity and the reduced need for many different supply voltages.

Let’s begin by having a look at the circuit, shall we?

(continue reading…)