Sun monochrome monitors, 13w3 cable, keyboard cable

Sun monochrome monitors, 13w3 cable, keyboard cable

Post by Joel A. Shapi » Sat, 11 Apr 1992 22:07:36

A group in our Physics Department has bought monochrome Sparcstation II's for
each of its members, but they find the fan noise very disturbing. So
they wish to keep their CPU's in a central location and use long
cables to reach the monitors and keyboards in their offices. NuData
sells such cables but they are very expensive, over $400 apiece.
   I tried to make a substitute, and this is a combination progress
report and a request for info and help. As the 13W3 connectors are
(apparently) impossible to buy, my approach was to take a short Sun
cable and split it in the middle. Here is a summary of what I found
and some info I got from Sun. If anyone can correct or enlarge the
info, I would be very grateful.
   Here is the 13W3 connector and its connections within the Sun

(   A3    A2       5   4   3   2   1    A1      )    A1, A2 and A3 are coax
 \    ^     ^      *   *   *   *   *      ^    /     1-10 are twisted pair
  \ < o > < o >                         < o > /
   \  -     -        *   *   *   *   *    -  /       coax pins are about
    (_______________10___9___8___7___6______)        4mm diameter (ground)
                                                with 1mm hole (signal) on cable
Internal wiring
        A1 red coax     1=black twisted with 6=brown   (Twists are rather tight
        A2 green coax   2=red  twisted with (7?)=orange   about one turn/inch
        A3 blue coax    3=yellow twisted with 8=green
                        4=blue twisted with 9=purple
                        5=gray twisted with 10=white
        [(7?) i wrote down 3 but this must have been a mistake]

All pins are connected straight through, including the shell.
There are no connections between the pins or the 4 grounds.

I plugged it into a Sparcstation I and measured DC voltages (Remember,
this is a monochrome machine)
When plugged into an SSI, the three coax grounds are connected to the shell.
DC voltages appear:   0.7 V on each coax signal, and
        On pin no   1   2   3   4   5   6   7   8   9   10
        DC volt     0   5   5   0  4.5 4.5  0   5   5   0
An oscillascope found signals on all the coax, with A1 and A2 possibly
identical, but A3 different. But there also seemed to be signals on
several of the twisted pair, at least relative to the shell.

The cable has a very heavy overall shielding of foil surrounded by
braid, and a thich plastic covering. The coax seems to have roughly
75 ohm impediance. Our setup for testing it was primitive, but 75
seemed clearly better than 50.

I got the following info from Sun:

> Colour Monitors

>       1 GND                    8 SENSE1
>       2 VSYNCH                 9 SENSE0
>       3 SENSE2                10 GND
>       4 GND                   A1 RED
>       5 CSYNCH                A2 GREEN
>       6 HSYNCH                A3 BLUE
>       7 GND

> Mono Monitors

>       1 No Connection          8 No Connection
>       2 No Connection          9 No Connection
>       3 Jumper to pin 4       10 H-V GND
>       4 Jumper to pin 3       A1 No Connection
>       5 H-V Combined Sync     A2 Video (coax)
>       6 No Connection      >
   A3 No Connection
>       7 No Connection

As we will see, this mono info didn't seem to work. But first, more
info from sun:

- Show quoted text -

>       Three pins are provided on the 13W3 connector for monitor
>       identification. These pins are internally pulled up. The
>       appropriate pins should be tied to ground for a zero and
>       left open for a one. The monitor ID may be read via a re-
>       gister in the Lego Frame Buffer Control (FBC chip). The
>       codes are tentatively assigned in the next table.

>       Monitor         Resolution              ID2   ID1   ID0

>       No monitor                               1     1     1
>       Reserved        640x480                  1     1     0
>       ECD             1024x768                 1     0     1
>       Apple           1152x870                 1     0     0
>       Sun Standard    1152x900, 66Hz           0     1     1
>       GPD Proposal    1280x1024                0     1     0
>       Sun Proposal    1152x900, 76Hz           0     0     1
>       Reserved                                 0     0     0

another message from Sun says

>       1 GND                    8 SENSE1
>       2 VSYNCH                 9 SENSE0
>       3 SENSE2                10 GND
>       4 GND                   A1 RED
>       5 CSYNCH                A2 GREEN
>       6 HSYNCH                A3 BLUE
>       7 GND

This would seem to be consistent, in that the connecting of 3 and 4
would make ID2=0 while the floating 8 and 9 would make ID1=ID0=1, and
tell the CPU my monitor is a Sun Stardard, which is almost surely

Now for my progress. As I had messured the impedience of the coax at
about 75 ohms, I used RG59 to connect this. A test with only this
line extended did very well. Then I tried extending the pin 5 pin 10
pair, at first using appletalk cable, which is shielded pair with
78 ohm impedience. This did not work, in a fashion which made me think
of impedience mismatch. I then bought some shielded twisted pair
claimed to be 100 ohm, and a long run to do a rigorous test. 200 foot
for this pair and the coax, and the video still seemed acceptible
although there was slight fading. That is, the black cursor rectangle
had faint darkness following it for about one more character space
(roughly 8 pixels), but it was not bad for a monochrome system, and
would have been acceptible except for:
Notice that sofar I have left all other lines hanging. So I figured
the CPU inquires at boot time what kind of console there is, and
finds no monitor because I haven't grounded pin 3. So I connected
pins 3 and 4 on the end connected to the CPU and tried again.
Unfortunately, the same thing happened, and I am discouraged, after
having spend an inordinate time on this. Much as I hate to tell them
to spend over $4000 on 10 cables, I think I will unless someone can
give me encouraging advice!
   By the way, I had great difficulty in making the join with the
coax, because I wanted to keep the braid surrounding the joint pretty
well intact, and because the central conductor of the 13w3 coax is
so fragile that it broke on me three times after closing everything
up. This job is not for the faint-hearted.

        Joel Shapiro
          (supposedly a theoretical physicist)