Step 4: Prepare PIC

Now we need to do a bit of work on the PIC chip before it's ready to be soldered in place.

First, a note on pin numbering, in case you're not familiar with the subject. Pin 1 is indicated by a recessed dot on one corner of the top of the chip; the rest are sequentially numbered, going counter-clockwise around the chip. If you don't see a dot in one corner, you're probably looking at the bottom of the chip - flip it over!

These before & after pictures show the pin 1 dot in the upper left - however that is NOT the orientation in which it will be installed in the JuiceBox. Pay attention to the pictures! A PIC installed upside-down will probably not kill your JuiceBox, but neither is it going to accomplish anything...

Using needle-nose pliers, gently bend pins 1, 4, and 8 until they point straight outwards from the chip. The idea is to keep them from touching anything on the circuit board when the rest of the pins are soldered into place.

Apply a bit of solder to the top sides of pins 1 and 8 - that will probably be easier to do now rather than later. Don't bother with pin 4, it will not be connected to anything.
Would it be possible to do this with a Pixster?
As far as I know, there isn't any sort of accessory for the Pixter that lets it use external storage cards, which is a fundamental requirement for what Pixecutor does. The minimal amount of hardware that would let you run your own code on a Pixter would probably be an entire cartridge with ROM and a storage card interface, a much more complicated project than the Pixecutor.
can you run Mac OSX ????????
If you're asking if the ARM toolchain can be run under Mac OS X, the answer is yes - in fact, all of the demo programs I supplied were compiled on a Mac. If you're asking if Mac OS X could be run on the JuiceBox, well... the concept is good for a laugh, at least.
This is an excellent instructable. I vaguely followed the juicebox dev process on www.linux-hacker.net, and am very pleased to see this project and instructable in the light of day. Excellent work, especially in light on how much work it took to get to this point!
Went to your page JuiceBoxPixecutor but its blank, any probs with this ?
Looks like that wiki got restored from an old backup. Hopefully they'll be able to restore everything eventually, but at the moment the best I can suggest is that you do a Google search for {{{pixecutor site:elinux.org}}} and click the &quot;Cached&quot; links for each page.<br/>
Nice source code; well commented/etc. Sorta unusual for this sort of project. (Good Programmer!) I wonder if a lot more can be done with this sort of self-contained "jtag bug." Desperately in need of those demo programs, though. Should be compatible (except perhaps physically) with jtag debugging if you connect a spdt switch from the reset pin to GND/VCC. Connected to GND, the chip will stay reset, with all the pins tri-stated.
Sorry about the delay in getting the demo programs up - there was a random flakiness in some of them that I wanted to be absolutely sure was gone before making them public. Anyway, they're now available <a rel="nofollow" href="http://www.elinux.org/wiki/JuiceBoxPixecutor">here</a>.<br/><br/>Holding the PIC in reset would indeed allow it to stay out of the way of a JTAG debugger. However, there's no general way to get the JTAG debugger to tri-state <em>its</em> outputs, to get out of the PIC's way. You'd probably have to physically disconnect and reconnect the JTAG pod every time you ran the program.<br/>
Your pictures are good. Not enough of them though. It would be nice to see the product you're working with; Without having to google for it. I would say maybe 2% of us know what the product is at all... (I'm not one of them.) It would also be nice to see your finished project in operation. That would give us the idea that it actually worked.

About This Instructable


10 favorites


More by jasonharper: Pixecutor - run software on a Mattel JuiceBox
Add instructable to: