[QuadList] Ampex Hint of the Week, Video Levels in VR-1200 and VR-2000
Chill315 at aol.com
Chill315 at aol.com
Mon Feb 27 08:07:42 CST 2012
Setting video levels is an interesting problem. What is the philosophy
that one should follow. At the production facilities I worked at, we took
the view that the video operator did his job correctly. That was the heart
of the question. We therefore had the following procedure.
Set the Proc Amp to Unity. Then use the Demod Level out to set the video
out of the machine.
To do this it was necessary to do the following.
Set unity levels in the proc amp. I used a scope with the invert and add
function to null out the levels at the highest gain possible. That way it
was as close as possible to perfect.
The book has the alignment procedure. Here are the following pots numbers
and what they do.
R-5 on the test point panel sets burst level.
R-6 on the test point panel sets video level.
R-9 on the test point panel sets the set-up level.
R-39 Board 17 sets the blanking clamp level to the same as input.
R-51 Board 13 sets the sync level.
Once we did this, we found that the quality of the video shot up. We were
able to get more generations. The reason being that we did not make
slight changes to the video quality each pass. Miss-set the black level by a
couple of IRE's on the bars and over a few passes, you either wash it out or
darken it. Also it was the intention to keep the dynamic range at the
optimal point for each piece of the chain.
The thing we found is that we rarely had to take the proc amp out of
unity. The times we did this was usually when we got a tape from a station.
They seemed to never have a set of best practices. It was just throw up a
tape and push buttons. Some even would record tapes without bars and send
them to us. Try to make a silk purse out a sows ear on those situations.
Chris Hill
WA8IGN
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://quadvideotapegroup.com/pipermail/quadlist_quadvideotapegroup.com/attachments/20120227/a1dd0118/attachment-0004.html>
More information about the QuadList
mailing list