Summary: Matrix camera with TAAC (l

[CAMPBELL%UTOROCI.BITNET at cornellc.cit.cornell.edu]

Many weeks ago, I posted a request (to SunSpots, actually) for info about
hooking up a Matrix Multicolor video camera (aka Dunn 638HSC) to our Sun
3/260, which also has a TAAC, and which also has longish (say 20 ft) video
cables between the system and the monitor.  We were seeing ghosting on the
monitor when the TAAC was hooked into the video lines (whether or not it
was active) and the Matrix was connected.  The Matrix has a switch for
either 75ohm or Hi-Z termination, but it seemed that it was doing
something funny to the video even when told to be Hi-Z.

Anyhow, we got several responses, all of which basically said, "Let us
know what you find out."  The best lead was a phone call from a chap at
Transcept (as in the T in TAAC) - and I hadn't even put my phone number in
the posting!  A comparison of this to reports of U.S. Sun support will be
left as an exercise to the reader.  (Canadian Sun support is good, by the
way).

So this Transcept guy and our electronics shop conversed, and results
forthcame in the form of two active boxes, one at the TAAC output and the
other in-line near the monitor.  The fact that we wanted a switch unit to
select any of several source systems as input to the Matrix complicates
matters a bit, but not much.  What follows are the notes made by our
electronics shop at my request for this posting.  The schematic renderings
are mine.  . . .

The Sun and TAAC appear to have the following output configurations:

Sun:                                 Taac:
------75ohm---o (to 75 ohm monitor)   -----0 ohm---o (to 75ohm monitor)
 |                                    |
(2V)                                 (V)
 |                                    |
---                                  ---
 -                                    -
In either case the cable is terminated with 75ohm at the monitor, after about
20 feet of cable.  We think the Sun configuration is better, effectively
terminating the cable with 75ohms at both ends.  The TAAC output appears to be
a bit non-standard.

We built two devices, one mounted close to the TAAC output, the other near the
monitors.  The downline unit is actually a switchbox, switching the output of
several devices (The Sun/TAAC, a MegaVison image processing unit, and an
HP 9000 workstation) to a Matrix Multicolor video camera, without disturbing
the feed to the monitors.

The video buffers used a Comlinear CLC 400AJP operational amplifier. The first
unit effectively standardized the TAAC output (s and c mark signal and common
(shield) inputs from the TAAC).

               ------270ohm-
              |            |
   |---270ohm---| - \      |
  ---           |    \     |
   -            |     >-----68ohm---o (output)
                |    /            --- (shield)
s---------------| + /             |
  |                               |
  x 75 ohm                        |
  |                               |
c-|--------------------------------
  |
 ---
  -

The output stage of the switchbox is similar, with the 75ohm resistor
replaced by a 56kohm one to present a high impedance to the line.  Each
input line (four input devices times 4 channels [RGBS] equals 16 BNC
connectors) is fed through immediately to its output (16 more BNC
connectors).  For video switches, the box uses Teledyne RF relays, 712D12.
Both the signal and shield are switched. When selected, they are fed into
the points marked s and c.  The four shields for each input device are
connected together inside the switchbox, but each device's shield system
is separate, and only the selected system's shield(s) are connected to the
Matrix's.

The Matrix appeared to have some shunt capacitance, even when set to high-
impedance input.  Thus our switchbox presents a known high-impedance tap
on the video lines, and offers a 75ohm output for the Matrix (which we set
to 75ohm input mode). The teledyne RF relays ensured minimum cross
coupling.

The HP input still does not work correctly; although its output is usable on
the Matrix, the HP monitor cannot sync on the output from the switchbox.  The
HP seems to be working on a current-based specification.  Work continues.

                                    . . .

Hope this helps somebody.  If anyone's more interested and I haven't made
something clear, send me a note or give me a ring at (416)924-0671x4924.  I
could arrange to mail copies of the shop's notes and schematics.

Chip Campbell
VAX System Manager, Physics Division
Ontario Cancer Institute
Toronto, Ontario
bitnet: campbell at utoroci