Home arrow Blog arrow FreakZ arrow Status Update - 2009-10-19
Status Update - 2009-10-19 | Print |
Written by Akiba   
Tuesday, 20 October 2009

Finally working through my design backlog and getting ready to start the stack development again. Things took much longer than expected. After releasing the Chibi stack, I realized that hardware is needed or else the stack is meaningless (seems to be true for any stack I guess). So I took a couple of days to design a board specifically for it so that people could play with it. It's mostly targeted at the DIY electronics community which is composed of artists, hardware hackers, and enthusiasts so I had to put some extra thought into it to increase functionality and keep the cost low. I just sent the PCBs out yesterday so I'm hoping that they should be back in about a week and a half (including shipping time).

Some details on the Chibi board are that its using an ATMega32U4 AVR MCU (32 kB flash, 2.5 kB RAM, integrated USB) which is compatible with the AVR Dragon emulator. This is one of the cheapest in-circuit emulators available and goes for about $50 at Digikey. Beats the crap out of paying $300 for the JTAG ICE MKII which is the full blown ICE for AVRs. Along with that, it includes an on-board AT86RF230 radio with a tuned front end and an SMD chip antenna with an option for either an SMA or RP-SMA connector in case an external antenna will be attached. I also included a coin cell holder for battery powered operation in the case that it will be used in a remote application. The board also has the standard serial connector that I'm using for all of my MCU boards so all peripheral boards will be compatible with it. The peripheral boards will consist of sensors, relays (mechanical and solid-state), displays, random things like mp3 decoders, and of course communications peripherals, ie: the Microchip Ethernet MAC+PHY w/SPI interface. The secret purpose of the design is to be a gentle introduction to wireless sensor networking, since WSNs are a bit intimidating if you're completely new to it ;) I suspect that it should be fun to use and am looking forward to working on it with Tokyo Hackerspace who were the original target audience for it.

While I was at it, I finally got the Hackerspace kits out the door for MAKE Japan, which we're going to be helping O'Reilly Japan with. Its basically a hackable scrolling LED sign which has a USB interface. The default app will allow updates to the scrolling message via USB, but there is also a firmware bootloader and an exposed SPI interface for other crazies to turn it into something more interesting. I was surprised how difficult it was to design a scrolling LED sign, since it seems like kind of a standard thing for hardware hackers to make. Turns out that there were a lot of gotchas, like needing current source and sink drivers to make the display as bright as possible and having special code to remove ghost images from LED flicker. Anyways, the final prototype worked well and even my wife was impressed which is not easy since she usually takes no interest in what I do.

And along with all of that, I finally got my Zigbee development boards tuned and working (which was my main test system for Chibi). So for the past three weeks that I've been back, it's been nothing but design, debug, and hardware galore. Now that most of it is out of the way, I'm looking forward to start tweaking the stack again...which brings me to another interesting thing that happened...

I was contacted by Open Remote which is an open source project for home automation. They're trying to tear off the closed source veil of secrecy that seems to be prevalent in home automation, mostly due to proprietary standards and/or vendor silliness. To make a long story short, I'm excited to be collaborating with Open Remote to help them integrate Zigbee into their platform which will be good because it gives me a target to shoot for, as well as a live test bed for home automation. I'm also thinking of setting up an Open Remote server inside the Tokyo Hackerspace and getting some of the application programmers involved in the effort. It should be really fun. 

In the meantime, I still have quite a list of things to do, but working on the stack and documenting Chibi are the main priorities. And if you're interested, here are the final schematics for the Chibi Dev board. You can view them through the JPGs below. Here's the pdf since I know the JPGs make it hard to see the details:

PDF Schematic Link

Click image to open!
Click image to open!
Click image to open!
Click image to open!
Click image to open!

Hits: 4784
Trackback(0)
Comments (2)Add Comment
...
written by Kevin Townsend, October 20, 2009
Looks great ... looking forward to trying it out. (Not sure how much chance there is that the PCBs will arrive while I'm in Tokyo, though.) I've been working a bit on trying to switch between different power sources (battery vs. DC, etc.) ... I was thinking that a cheap supervisor could probably make things easier (no physical switch) since it can detect whether battery or dc are present and switch accordingly (and maybe charge the battery while DC is plugged in if you want to add another $1 or so in parts and get all fancy). I'm sure a cheaper battery-backup supervisor exists, but I've been playing with this one myself ... MAX6364 (3.08 V):

http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name;=MAX6364LUT31+TCT-ND

At least on paper it looks promising (though not cheap at $2,35/100) if I want to run off a single LIPO cell, but be able to plug in a DC jack or USB when battery power is low and charge the battery using an inexpensive MCP73833: http://search.digikey.com/scri...-AMI/UN-ND

I want to put a prototype together anyway once I have Chibi working myself. Light-weight, long life LIPO cells with easy recharging should be a useful combination, especially since I've seen some really small cells lately.

In any case, great job on the boards. Chalk me up for a couple when they're ready for ordering.
report abuse
vote down
vote up
Votes: +0
...
written by Akiba, October 20, 2009
Thanks!

As for detection of a DC input, you can probably forgo a supervisor and just use the onboard comparator on the AVRs or microcontroller you're using. If the DC voltage is present, it will probably be equal to or greater than the coin cell (3V at full charge) hence you'll see a 1 at the output of the comparator. Once the MCU detects it, it can switch over to the DC input via transistor switch.
report abuse
vote down
vote up
Votes: +0

Write comment

busy
  No Comments.

Discuss...
< Prev   Next >