[AT91SAM7_prototype] - about the possibles energy supplies
Jorge Pinto aka Casainho
Donald, can you please tell what cells would you use? NiMh? Li-ion?
Please tell me about the prices for the batteries, holders and how it
should be charged.
I think that the best way is to use 3 X AAA NiMh or alkaline cells.
There are holders with springs but where batteries are kept inside a
plastic case with springs - I bought them recently for about 1,3€.
I was thinking not put the batteries on PCB but instead at center of the
wheel.
About your idea of using just a shootcky diode instead of a LDO
regulator, I would prefer to use the LDO, for being more secure that
when using alkaline batteries there is no problem, however, we can just
also put on the schematic and PCB that diode in parallel with the LDO
and in the end we just assembly one of them :-)
As for firmware, I saw that Atmel have an application note with title
"USB Mass Storage Device Driver Implementation" - the source code; for
the AT91SAM. Also there is drivers for DataFlash memory :-)
DPZiems
certainly has weight and power density benefits, it requires
specialized chargers and can explode if handled wrong. Considering
these things will be whizzing around the inside of a bicycle tire in
mud, rain, and UV exposure, I'm not so sure Li-Ion technology is
rugged enough yet.
I agree about the LDO. I was not aware that LDO regulators had that
low of a voltage drop, I thought it was more along the lines of 0.5V,
in which case the Schottky would be more efficient, but I was wrong.
No real need to put the Schottky in parallel, the LDO is cheaper, more
efficient, and smaller, might as well go with it.
Putting the batteries in the center is a good idea, as close to the
hub as possible. Lowers the moment of inertia, and allows multiple
light units to be driven from a single battery pack, if desired.
As to the firmware, I know nothing about ARM yet, so, uh, good find!
On Jul 9, 7:24 pm, Jorge Pinto aka Casainho <casai...@gmail.com>
wrote:
Jorge Pinto aka Casainho
> Personally, I would use NiMH for this application. While Li-Ion
> certainly has weight and power density benefits, it requires
> specialized chargers and can explode if handled wrong. Considering
> these things will be whizzing around the inside of a bicycle tire in
> mud, rain, and UV exposure, I'm not so sure Li-Ion technology is
> rugged enough yet.
Ok.
> I agree about the LDO. I was not aware that LDO regulators had that
> low of a voltage drop, I thought it was more along the lines of 0.5V,
> in which case the Schottky would be more efficient, but I was wrong.
> No real need to put the Schottky in parallel, the LDO is cheaper, more
> efficient, and smaller, might as well go with it.
Ok.
> Putting the batteries in the center is a good idea, as close to the
> hub as possible. Lowers the moment of inertia, and allows multiple
> light units to be driven from a single battery pack, if desired.
Ok.
> As to the firmware, I know nothing about ARM yet, so, uh, good find!
Well, I am just looking, to know if later we have code ready for put in
pieces and make all the system working - for example, I saw that Atmel
have code drivers for the AT45 DataFlash so I need to change for that
version to have sure that later it will work.
THE IMPORTANT for now is starting the schematic design(firmware we just
need to worry after having boards assembled at our hands - I will
assembly them and sell online on Ebay and sell empty PCB) and at same
time find the cheap components. Would be possible for you start working
on the schematic?
The ARM7 schematic is the same of this one board:
http://www.olimex.com/dev/sam7-h256.html
I have that board with me and I even made a LCD 16x2 working with it - I
am familiar with the development tools of the Atmel ARM7. I also am
familiar with the Phillips LPC ARM7 but I found this one from Atmel with
128K Flash more cheap and also with all the code we need :-)
We should use http://code.google.com/p/armopendous/ and
http://code.google.com/p/avropendous/ as a source for the USB connecter
and DataFlash KiCAD libs :-)