Posted a screen shot mockup of my proposed
LPC2103 Protoboard. This protoboard
will be used to develop the ARM port of the TRAXMOD firmware.
I'm currently stuck on trying to figure out how to cram all of the electronics
on this board together with means for installing 4xAA NiMH batteries. It's
like the HOODMICRO all over again. This one is going
to be designed around the dimensions of a standard Altoids tin.
Altoids tins are pretty close to the form factor desired and they are very
easy to obtain. If it worked for Lady Ada, it should be good enough for us
too, right?
A set of 4xAA batteries fits nicely inside the Altoids tin. If you lay down
a standard thickness PCB with DB25F connector attached, the lid closes
perfectly. But there is no room left to spare. There is definitely not enough
space for a plastic battery holder. Anything that makes the PCB
slightly thicker makes the lid fail to close. We're going to have to try
keeping a flat space on the bottom of the PCB, one big enough for the
batteries to rest in. We're also going to need some springs/tabs for
connecting to the battery terminals. I think this might be possible to
do with existing parts available from Digikey/Mouser, but I haven't
gotten all of the details worked out.
The other alternative is to use smaller AAA batteries. I don't really want
to go that route though. AA batteries pack far more power in each cell and
don't cost much more than AAA batteries. A single AAA NiMH battery holds up to
850mAh of power, while AA NiMH batteries hold up to 2600mAh and only cost $0.45
extra.
Moved the Helix MP3 Decoder link to the FAQ. MP3 is not the main focus of
TRAXMOD.
01-21-2006
Not sure if the LPC2103 will have enough memory to support MP3 decoding,
but added a link to the Helix MP3 Decoder for ARM any way. At least there is a chance of possibly
being able to play MP3s. Of course, MP3 playing is not the main focus of the
TRAXMOD: MOD playing is the main purpose. I'm won't be shedding many tears
if we still won't be able to play MP3s.
Still doing work on the JTAG Debugger and starting some
work on a LPC2103 protoboard.
01-18-2006
Been reading up on the LPC2103 chip like mad. It looks to me that the
PWM can be clocked at the same speed as the core, i.e. 70MHz. That should
be excellent, maybe even possible to achieve 10 bits of output audio
resolution without using an external DAC chip.
Programming and in-circuit debugging of the chip can appearantly be done using
the industry standard JTAG interface. I've been drawing up schematics for
a cheap DIY JTAG Debugger for the parallel port. Some of
the people on the LPC2000 yahoo group say that
parallel port "wiggler" clones
are slow and act flakey -- they work sometimes, but other times just don't
want to work at all. I changed the circuit slightly based on their comments
and suggestions. Hopefully the one I build will work well enough to use for
developing the TRAXMOD ARM port.
Flakiness of the in-circuit debugger was one of the reasons why development
on the dsPIC dropped off. If the cheap DIY JTAG Debugger
flakes out on me, hopefully all will not be lost. Besides the JTAG interface,
the LPC2103 has a built-in serial bootloader that should hopefully work even
when the JTAG debugger isn't working. As long as I have one reliable way to
program the chip, I'll still be able to continue work.
In other news, yesterday I got an email from Philips saying that they didn't
ship my samples order because the chips are now at Digikey. Ugh. So
I had to call up Digikey this morning to get an order in before all 150 of
those chips disappear. Looks like they're down to 141 just overnight.
For the first prototype player board, I would like to have one DB-25F
connector along the bottom of the board. This connector would contain
access to the JTAG pins, serial port, audio out, and power. The rest of
the board would consist of the LPC2103, SD Card socket, a 3.5mm stereo
audio jack, maybe some buttons, and maybe some additional header pins holes
for attaching any additional prototype circuitry. A custom cable would connect
the player's DB-25F to the JTAG debugger and the PC's serial port. I am
still up in the air about what to use for a case...
01-03-2006
Started investigating the new
32-bit LPC2103 ARM7TDMI-S chip from Philips. This
chip costs less than the cheapest dsPIC while providing over twice as much
MIPS and SRAM. Submitted an order for some samples. Hopefully it will be
easy to get up and running.