Wang 2200: video timing problem & questions

From: Jim Battle <frustum_at_pacbell.net>
Date: Sun Sep 30 02:02:45 2001

After a few years of searching, I finally have a Wang 2200 computer (2200-T
style CPU, to be exact). Parts of it have been arriving from Dallas for
the past four months. The last piece arrived yesterday, the 2226
keyboard/monitor. Yeah!

Unfortunately, the monitor just doesn't sync at all. Not having worked a
whole lot with video, I was hoping someone could steer me in the right
direction. I'll explain what I know so far.

First, this monitor, cable, and CPU were all part of a set -- I'm not
mixing and matching. Supposedly the video was fine when it left Dallas a
week or two ago.

The video goes from the CPU to the monitor over a single coax and has
composite sync. Video is baseband monochrome.

There are a couple hidden pots that can be twiddled, and I have, but the
best I can do is get it to a slow roll/tear. It is an impossible task as
just thermal drift causes the screen to wander. Neither vsync nor hsync
appears to work at all. At first I thought I just would have to tweak the
H & V pots into the lock range of the PLL that controls the sync but there
appears to be no locking zone.

I got a standard monitor that I use with my Sol and hooked it up. The
screen rolled wildly. By adjusting the H & V controls on that monitor, I
could get it to sync both axes, but the video was not readable -- it was
locking to some multiple/submultiple so the image was stable but not
coherent. Thus it seems to rule out the unlikely case that the original
monitor is OK and the video card is putting out signal but no sync.

So then I broke out my crappy scope that I picked up a hamfest, but which
is good enough for this job. I measured some timings. Interestingly, the
scope has TV V sync and TV H sync trigger modes; both worked properly to
lock the signal. Thus the timing is close enough to normal video for the
scope to think it was OK.

Timings below are as good as I can measure with this scope; 5% error
wouldn't be unlikely. I've put in square brackets the figure for normal US
television timing (from a web page that I just looked up)

vertical period: ~16.8 ms/frame [ 1/60 = 16.67 ]
v blanking: 4 scan lines
v front porch: 17 scan lines
v back porch: couldn't trigger it in a way to count them
v active: there are 24 rows of text, each on 11 scan lines, or 264 scans, =
15.3 ms

horizontal period: 58 us [63.5]
h sync: 2 us [4.7]
h front porch: 7 us [4.7]
h video: 47 us [52.6]
h back porch: 3.5 us [1.5]

dc level: 0.7v
sync: 0v
video white; 1.4 v

[ignore the voltages somewhat -- I was measuring them using a
high-impedance scope probe and a properly terminated coax would probably
lower those figures]


So here are my specific questions; if you have any thoughts and have read
this far, don't limit your responses to these questions, though.

1) Doesn't is seem odd that neither H nor V sync lock? What would cause
both to go out at the same time during shipment?

2) Doesn't it seem odd that the timing is pretty close to nominal but
     my 2nd monitor can lock but not get a proper picture?

3) Any thoughts about what I should try next?

I have no schematics or technical docs for any of this.

Thanks for any ideas, even crackpot ones.
-----
Jim Battle == frustum_at_pacbell.net
Received on Sun Sep 30 2001 - 02:02:45 BST