Presented by: Only
On Saturday Night
|
Here's
the 8mb PCB. It's the same exact size as the NoCan series. In these old
photos it is only partially stuffed, as you can see. Now it is fully stuffed
and testing is complete. It all works and on the first time too, i.e. no
mods or patches! This board lays on top of the CoCo-3 and plugs into the
6809E, 74LS245 sockets and into CN4, CN5, CN6 connectors.
The
funny overlapping sockets in the lower right are for the 16550 or 6551A
option, user picks one. They are both wired to the MAX232 in parallel.
Now folks can pick which real serial port chip they want to use, just plug
it in.
Here
is NoCan3 mounted on top of a repacked CoCo-3. Also note a popular Cloud-9
AT-keyboard interface and how they fit together without much interference,
it just seemed to work out that way. In the photo: 16550 serial port chip,
68B21 printer port chip, MC68B09EP cpu, 74F245 buffer, 3 SMD buffers for
cpu, one delay-line chip, Lattice CPLD, Cypress DAT SRAM, two each 4MB
simms, MAX232 buffer, 1.8432MHz oscillator can, 74LS05 or equivalent for
16550 interrupt and reset, headers and jumper pins galore. This board uses
8mil runs and spaces for those of you who are interested.
The row of jumpers right behind the Lattice chip are configuration jumpers,
ie user address selections for the I/O chips. Check the "Detail" button
for more jumper info.
That reminds me, because only one serial chip is chosen, the left-over
chip select is still active and available. I suppose that you could piggy-back
a clock chip and use that un-used chip select. What clock-chip would you
like to see? Right now I am investigating the Dallas 12C887, a bus oriented
chip.
This page updated: 2000/05/08
Links removed, sorry.