Prev: Write CBM disk in IBM floppy drive? (Corrected)
Next: Eprom question..
From: BruceMcF on 28 Feb 2010 12:49
On Jan 28, 6:13 pm, "mrtinbs...(a)gmail.com" <mrtinbs...(a)gmail.com>
wrote:
> I was thinking about a simpler design.

> A simple expansion pass-thru board with ribbon-cable-pins in the
> middle.

Simpler to build, except someone selling such a cartridge would also
have to sell pre-wired ribbon cable for various configurations, at
which point it is no longer a simpler design.

A *simpler* design is a two-port expander, a cartridge port where you
have ribbon cable pins.

To avoid putting chips on the board (and make it genuinely simpler but
also still generally usable), there would be a jumper block for the
vertical and out the back ports, with each select line jumper when
installed pulling the select line high with a wired-or. Simply allow
it to be set up to split the select lines between the vertical port
and the pass through and leave any more complicated arbitration to the
cartridges plugged in.

Keeping the design as simple as possible also allows the board to be
as short as possible. If it had holes at the back for an optional "U"
bracket as a support, it could be extended back tinker-toy fashion -
and allow the 3 port expander with switches to be converted into a
four port expander.
From: Jim Brain on 28 Feb 2010 21:39
On 1/28/2010 5:13 PM, mrtinbspam(a)gmail.com wrote:
On Jan 28, 6:13 pm, "mrtinbs...(a)gmail.com"<mrtinbs...(a)gmail.com>
> wrote:
>> I was thinking about a simpler design.
>
>> A simple expansion pass-thru board with ribbon-cable-pins in the
>> middle.
>
> Simpler to build, except someone selling such a cartridge would also
> have to sell pre-wired ribbon cable for various configurations, at
> which point it is no longer a simpler design.

And ribbon cable is "horrid" for the unbuffered cart port signals.
> To avoid putting chips on the board (and make it genuinely simpler but
> also still generally usable), there would be a jumper block for the
> vertical and out the back ports, with each select line jumper when
> installed pulling the select line high with a wired-or. Simply allow
> it to be set up to split the select lines between the vertical port
> and the pass through and leave any more complicated arbitration to the
> cartridges plugged in.

Could you explain this in more detail?

(Sorry, I did not see this when it was originally posted)

Jim

From: Jim Brain on 28 Feb 2010 21:47
On 2/27/2010 11:21 PM, BruceMcF wrote:
> On Jan 27, 1:24 am, Jim Brain<br...(a)jbrain.com> wrote:
>> Trying to fill a small gap, I'm working on a 3 slot expansion port
>> solution ala EX-3 of FB-EX3. Here is a first cut of the PCB design:
>
>> Comments are appreciated. Some things I'm curious about:
>>
>> * Is the switch location OK, or do people need them off to the side?
>
> Where is the switch location on the diagram, in the white circles in
> front of each slot? I like a reset switch to be easy to reach (for the

Yes, it is.
> longest time the only cartridge I had was a WarpSpeed), but the
> settings of which select line goes to which cartridge doesn't have to
> be.
OK.

>
>> * Does anyone want a top slot for the 3rd slot? or is the back
>> entry desired?
>
> I prefer the back entry. A program cart or an el-cheapo wired-or
> compact flash cart could be sticking up, but I prefer that a serial
> interface cart not be.
>
Since the initial posting, some folks have commented, so the design
changed a bit. Notably:

* switches are now on the right side, opposite the side the FOTIOS and
CMD EX3 uses. The premise is that it makes it easier for C128D users,
and C64 users won't see much of a difference.
* Reset switch at back right of expander, like on normal cartridges.
* IO1/IO2 swap is not on right side of PCB, with the rest of the switches.
* Although not changed from the design, I am seriously considering
populating both the top entry and back entry connectors for the third
slot. Mind you, both slots will be configured with the same set of
switch settings, but it seems easier to populate both connectors and let
the user decide which to use. It also makes stocking easier, as I won't
need to stock two separate units.

Jim
From: BruceMcF on 1 Mar 2010 00:41
On Feb 28, 9:39 pm, Jim Brain <br...(a)jbrain.com> wrote:
> On 1/28/2010 5:13 PM, mrtinbs...(a)gmail.com wrote:
> On Jan 28, 6:13 pm, "mrtinbs...(a)gmail.com"<mrtinbs...(a)gmail.com>
>
> > wrote:
> >> I was thinking about a simpler design.
>
> >> A simple expansion pass-thru board with ribbon-cable-pins in the
> >> middle.
>
> > Simpler to build, except someone selling such a cartridge would also
> > have to sell pre-wired ribbon cable for various configurations, at
> > which point it is no longer a simpler design.
>
> And ribbon cable is "horrid" for the unbuffered cart port signals.
>
> > To avoid putting chips on the board (and make it genuinely simpler but
> > also still generally usable), there would be a jumper block for the
> > vertical and out the back ports, with each select line jumper when
> > installed pulling the select line high with a wired-or.  Simply allow
> > it to be set up to split the select lines between the vertical port
> > and the pass through and leave any more complicated arbitration to the
> > cartridges plugged in.
>
> Could you explain this in more detail?
>
> (Sorry, I did not see this when it was originally posted)
>
> Jim

In http://www.jbrain.com/vicug/gallery/xpansion/ext3_d imagine the
first (vertical) connector as close as possible to the computer, and
the second (back end) connector as close as possible to the first (and
corner post holes at the back end of the board in case someone wants
to give it legs).

Ah, what is "as close as possible"? It depends partly on what else
there is there other than port connectors.

Its been a while, but it seemed to me that with the more sophisticated
needs covered by the 3 slot port expander, a 2 slot expander could
sacrifice ease of swapping settings around just using jumper blocks.

The ordinary direct use of a jumper block allows a line to be cut by
running the line through the block. Unswitched has the jumper in
place, switched has the jumper removed. There may be a pull-up
resister needed for stability.

The next is to allow a Low to be masked by running an always on signal
through the block and wire-oring that with the select line. Unswitched
has the jumper removed, switched has the jumper in place. I'd expect
the wired or would need a diode between the cartridge port select line
and the extender select line and line from the jumper block to avoid
interfering with the other port.

I'd not sketched it out for ROM cartridges, and it seems likely if you
want to switch ROM cartridges, you want to turn them on and off, so it
may not apply for ROM cartridges ...

.... but for the /IO lines and the two /IO pages, one jumper block
gives a straight through or crossover:

/IO1-in -> 0-in, 3-in
/IO2-in -> 1-in, 2-in
0-out, 1-out -> /IO1-through
2-out, 3-out -> /IO2-through

As I said above, if I recall correctly, all four paths need diodes
somewhere to avoid walking stepping on each other's toes.

Jumper 0, 2 for straight through
Jumper 1, 3 for cross-over
Jumper 0, 1, 2 to hold up /IO1-through
Jumper 0, 2, 3 to hold up /IO2-through
Jumper 0,1,2,3 to hold up both

And then a7 through a diode and two inputs of a second jumper block
with the outputs wired to /IO1 and /IO2 to mask the top half of a
either /IO1 and /IO2.

A simple /IO1 cartridge mirrors eight registers across the whole page
and a, for instance, GEORam which uses the /IO1 page for the memory
page and the top two bytes of /IO2 to set the address could be plugged
in, with the simple cartridge swapped to /IO2 and the top half of /IO2
masked out.

If two hex jumper blocks are used, there are two spare jumpers.
Probably not any use, but maybe.

First  |  Prev  | 
Pages: 1 2 3
Prev: Write CBM disk in IBM floppy drive? (Corrected)
Next: Eprom question..