Tear down/reverse engineering of Hasbro Electronic Scrabble Flash Cubes game
17 August 2011¶
18 August 2011¶
Thinking about the (assumed to be) IR communication I started down the road of using a logic analyser of some sort. But now that I think of it maybe I should be using a scope instead.
Anyway, I had access to a LogicPort Logic Analyser but couldn't get the associated software to recognise the device was connected when using WINE (Darwine, actually) on an OS X 10.5 machine. (I didn't try particularly hard--it was also a reasonably old version.)
While it might've been possible to try with WINE under Linux under a VM I'm short of memory and disk space at the moment so I skipped that option.
I then discovered sigrok which is an FLOSS cross-platform application for interacting with a variety of logic analysers--including the LogicPort.
You can follow along with my unsuccessful adventure in: Notes on (not) compiling sigrok logic analyser GUI for OS X
Since then I've discovered two further possible options:
I've not yet followed up with either of these options.
*By being SUMP-compatible it is possible to use OLS logic analyser GUI software with it--and OLS even has a
.dmg download for OS X (in addition to support for other plaforms)!
23 August 2011¶
Looking at the "back" (non-display) side of the PCB, when power is applied, the bottom left component of the covered pair of black top hat shaped LED-like components seems to have a signal on its bottom-most pin.
The signal is not constant it appears to be a stream of high-low followed by a pause before the sequence repeats. This was found using the Arduino-based Logic Analyser and an oscilloscope.
24 August 2011¶
After further analysis with an Open Bench Logic Sniffer I seemed to have a closer look at what's happening.
AFAICT, according to the OLS "LogicSniffer" software:
- the pause between data bursts is ~44ms;
- the data bursts are ~27ms long.
These measurements were taken with a sampling rate of 100 Khz.
The signal seems quite noisy in places--not sure which of the million possible causes that is. :)
The data seems to have a pattern rather than being a constant signal.
Oh, actually, maybe the signal's not noisy--I might not have zoomed in far enough... :)
Will need to compare signal with another unit to see if it's the same.
I got some weird signal behaviour when I tried to trigger the capture but I think that was due to interference from another cable or something. (Something to watch out for. :) )
Here is an overview of the signal captured:
This is the same capture zoomed in to show a whole "databurst":
And here is an arbitrary portion of the signal:
28 January 2012¶