MEM11 update

[Guy Sotomayor]

> On Feb 8, 2016, at 1:03 PM, Ethan Dicks <ethan.dicks at gmail.com> wrote:
> 
> On Mon, Feb 8, 2016 at 3:30 PM, Guy Sotomayor <ggs at shiresoft.com> wrote:
>>> On 2/8/16 10:09 AM, Guy Sotomayor wrote:
>>>> So, things are moving forward.  I also wanted to get folk's opinion on
>>>> the need to actually produce
>>>> an SPC form factor board.  In other words (and sort of in line with how
>>>> peripherals were done on the
>>>> original 11/20) is it OK to have the MEM11 be outside of the 11/20
>>>> chassis and connect via BC11A (my replica) cables?
> 
> For my own case, I have a 3-box 11/20 that I need to restore (as I've posted
> before, it was chopped apart and dumpstered, and I recovered it from there).
> The majority of the second and third boxes is MM11-E core memory units
> (I have N-1 because one of my co-workers nabbed a core plane to hang on
> the wall).  My plan with the MEM11 has been to restore the CPU cabinet
> and use the memory on the MEM11 instead of the MM11-E units, leaving
> me plenty of space and power supply for peripherals with the ultimate goal
> of running UNIX v1 on it (I also have an RK11C that would be a secondary
> restoration project, or an RK11D that should "just work")
> 
> With that, I had been expecting that the MEM11 would be an SPC board
> that would just sit in a DD11CK with some other periperhals.  Apparently,
> it's sounding like there's too much "stuff" for a single quad-height card then?

I don’t know at this point but I’m getting “uncomfortable” with the amount of
“stuff” on this board (most can’t really be helped).  I won’t know until I sit down
and do serious board layout.  The prototype will tell me a lot.

> Is it component density or having to go to a 4-layer board that's an issue?

It’s component density.  I have been assuming a 4-layer board from the get-go.

The other issue is daisy chaining of some of the signals.  I’m already running
out of pins on the PQFP and that just adds to the congestion.  I could probably
handle some of that with a small CPLD but again that adds to the board
congestion.

> 
>> As Ian stated in a subsequent post, it'll probably be something like a 1U
>> box with appropriate power supply.
> 
> That will certainly be functional, but it seems to up the cost quite a bit.
> 
> I would rather have an external something that works than not have
> anything at all, but I think an SPC, if possible, would be the most
> portable of solutions.

Yep.  I’m going the external route for the prototype for ease of development
and debugging.  I’d like to be able to do it in an SPC form factor.

> 
> If this was an external device, would it just have a pair of 60-pin cables
> to your replica BC11A?  Would it then have an onboard terminator or
> option to install a terminator?  I get that you won't be selling the MEM11
> as a bare board, but it doesn't seem that the intended audience would
> be put out by soldering their own 60-pin connectors and/or a bunch of
> terminating resistors if it needed to be at the end of a chain or in the
> middle.

Yes, I’ll probably put in two sets of connectors (in & out) so that it could be
in a chain or the option to install terminators.

Anything optional will be in sockets.  I’ll be putting the UNIBUS transceivers
in sockets because I can’t afford the overage that I’d need to provide to the
board house for assembly.

TTFN - Guy