Hi guys,
Quick update regarding the disc analyser... Latest news is:
* The USB interface is now running properly at full speed.
Write-to-buffer has gone from 84Kbytes/sec to over 180 with two small
firmware tweaks. Read speed is a little under 250Kbytes/sec, which means
the entire 512K buffer can be read out in a shade over two seconds. It's
not going to win me any awards, but I think it's plenty fast enough.
* It Has A Name! One of my friends suggested "DiscFerret" partly as
a slight dig at the Catweasel. The name has stuck to the point where
I've... um... registered a .com domain name for it... (hey, it only cost
?12). I'll be setting up the site as soon as I find a nice Wiki package
that isn't bloated to the size of a small main-sequence star (*koff*
Mediawiki).
* Schematics are basically done. I need to test the power supply
section, which is on the to-do list for a bit later on. Soldering down
the QFN-packaged power regulator IC will be interesting.
* There's an external power output for disc drives. It provides +12V
at 1.5A and +5V at 1.5A. Perfect for running small 3.5in drives, and
saves a mains socket. Haven't decided on a connector, but the +12 is fed
directly from the input power supply, so if you connect a 15V PSU you'll
see 15V on the EPS output.
* Power supply can accept between 9 and 15V, though it is specified
to operate with peak efficiency at approx. 12V. Input is
reverse-polarity protected with a grounded-gate low-Rds(on) MOSFET.
I'll probably be building up a prototype power supply board tonight or
tomorrow, and giving that a good beating with a homebrew DC load. I
think I've got some MJE3055s and trimpots somewhere, though whether
they'll work with Vce=1.2V remains to be seen (I suspect a MOSFET might
work better).
But first I need to get my paws on some 0.5mm carbide drill bits. There
are a couple of vias on that board that are in a really tight spot... I
make no secret of the fact that it was designed to be commercially
manufactured, not haphazardly assembled by some semi-skilled engineering
student in a garden shed (i.e. me!)
The plan for those is to drill out the holes, push in a resistor leg
flush with one side, solder on one side, cut and file flush on the other
side, then solder that side too. I figure that should be a tight enough
fit that the wire won't be going anywhere... and if that doesn't work
there's always 26SWG tinned copper wire.
Cheers,
--
Phil.
classiccmp at philpem.me.uk
http://www.philpem.me.uk/