LED Screen Documentation
Tom Mudd
I've compiled all the documentation I have on the LED from Piccadilly
Circus.
Its currently on our companies ftp:
ftp://public.displayled.com/LED_Screens_Infodump/
Username - dlpublic
Password - tannoy
There's loads of info on there, some relevant, some not. It is all
however from the same manufacturer and would certainly give you a
better idea of how these screens were used / configured.
Enjoy !
Tom
ps - Ill be clearing the ftp in approx 48 hours so please download all
you need by then.
Sam Cook
Tim Storey
\t
On 10/01/2012 16:35, Sam Cook wrote:
> Cheers for this Tom, will start downloading now.
>
> S
>
> On 10 January 2012 16:33, Tom Mudd <mist...@gmail.com
Dave Ingram
D
Sam Cook
Tom Mudd
hack the LED is to look at the data on the ribbon cable joining the
three LED tiles.
The SIB and whole serial data side of Lighthouse kit is seriously high
bandwidth custom protocol stuff (loosely based on HD-SDI video). I
think trying to re-engineer anything on the serial side is going to be
one almighty headache.
Mind you - I believe there's enough electronics IQ up there in
hackspace to reverse engineer NASA so - by all means give it a go :-)
Tom
Mike Harrison
Here is the datasheet for the LED drivers used in these tiles
http://pdf1.alldatasheet.com/datasheet-pdf/view/31602/TOSHIBA/TB62706.html
Looks like it does not do onboard PWM, so to get greyscales you need to drive them fast enough to
PWM them via the data. The tile master PCB is simple enough to be able to trace out the
connections, even if you need to sacrifice one for better access.
Should be a doddle to get a single tile displaying non-greyscale data though, and could probably do
some PWM on one tile with an Arduino - you'll run out of bandwidth and memory pretty quickly on
anything bigger.
.
Not looked in detail but considering how little there is on the 'master' board on each 16x16 tile, I
think the addressing is by simply selecting which data line on the ribbon the drivers connect to.
there is an eeprom on each tile, which I imagine would contain intensity calibration data.
Tom Mudd
You're quite right about the PWM and that's a spot on guess about the
eeprom.
The eeprom was a unique feature of the Lighthouse tiles at the time.
When manufactured, they were burnt-in for about 1 week and then put in
a calibration chamber.
The calibration data (individual LED intensity data) gathered would
then be written to the eeprom on each tile and be used to subtly
adjust the intensity of every LED on the tile.
When all the tiles were assembled into one big screen, the calibration
data would get compared and equalised among all the tiles and (in
theory) this would mean every tile would be calibrated with every
other tile in the screen.
Fairly cutting edge back in the day.
Tom
On Jan 14, 3:21 pm, Mike Harrison <m...@whitewing.co.uk> wrote:
> On Tue, 10 Jan 2012 09:03:50 -0800 (PST), you wrote:
> >On the reverse engineering subject, I think the best way for you to
> >hack the LED is to look at the data on the ribbon cable joining the
> >three LED tiles.
>
> >The SIB and whole serial data side of Lighthouse kit is seriously high
> >bandwidth custom protocol stuff (loosely based on HD-SDI video). I
> >think trying to re-engineer anything on the serial side is going to be
> >one almighty headache.
>
> >Mind you - I believe there's enough electronics IQ up there in
> >hackspace to reverse engineer NASA so - by all means give it a go :-)
>
> >Tom
>