Re: Problem with old(ish) equipment!
- From: Phil <Phil@xxxxxxxxxxxx>
- Date: Wed, 23 Nov 2005 10:03:11 +0000 (GMT)
Phil: Go back to basic timing principles:
1/ Ensure a single source of common syncs+burst from a stable spg to each
and every unit including the camera ccus and the vision mixer (switcher).
2/ Set both Waveform and Vectorscope to External lock on this same source.
Typically place this Vectorscope on the Mixer Prgm Output or a
convenient AUX bus (more versatile later) - and select Pgm Out to it.
3/ Monochrome time each source by comparison to a SINGLE source ref:....
Eg Colour Bars from the original SPG - which, if on the same cable
length to the mixer as the black+burst ref, will be in sync UNLESS they
pass through Distribution Amplifiers - which may, or may not, have
equalisers in them. (I have seen installations with the CB arriving at a
different time to the BB because someone THOUGHT an equaliser was NOT
needed on a BB DA, but was inclused on the CB DS - the result is of course
a timing difference due to the filter delay)
Procedure: Select the BLACK/colour bar input and centre the wfm monitor
and vectorscope, using Ext Ref. This usually acts as your unchangeable
reference source. Set at 50% slice point
Now select in turn, each camera CCU (eg BARS from CCU mode if available)
and adjust ITS GENLOCK SYNC setting until it is in MONOCHROME phase on the
Wfm - (Rapidly switch between the BB/CB and Camera and adjust until there
is NO CHANGE - there may be an automatic mode in the mixer (switcher) for
this). Adjust when looking at the 50% slice point only.
ONLY THEN look at the Vectorscope and adjust the CCU GENLOCK SUBCARRIER
phase in the same way - to align the BURSTS.... but this may not be
visible because, for example of a Proc Amp setting in the Vision Mixer,
passing through only the reference syncs (Sure Sync, SureBurst Operation -
a single source as far as a recorder is concerned!)
If the burst is not switched, when changing source, judge by the colour
bars, or failing that, colour picture, and aim for symmetry of the PAL
waveform.
It is quite possible that when the Vectors are now aligned, the syncs will
be seen to be misaligned by more than spec (25ns monochrome/SECAM) You are
aiming for 3ms or less timing error (PAL). In which case, return to the
SYNC adjustment, and move the synces to 'the other side' of ideal, to see
if it is better when the Vectors are retimed.
A Potential problem here, is that whilst the 1979 redefinition of PAL to
specify Sc-H phase improves matters, by specifiyinq the relationship to
+/-20 degrees, (instead of +/-180), this was not (immediately, if at all)
adopted by the BBC, or all equipment from manufacturers. (In fact the BBC
DPS system deliberately changed the Sc-H phase!)
A typical Sony Genlock Circuit has a range of 4micro seconds on the sync
phase knob, and several cycles of subcarrier on the subcarrier knob in
270degrees of rotation. This allows a signal to vary from one side of Sc-H
to another simply by resting your fingee on the Sync-phase knob, without
attempting to turn it!!! - so be careful how you touch them!
I have known of a JVC mixer whose output was configured 180degrees off its
input reference - possibly for similar reasons.
So the final position, is monochrome sync 'as close as possible', but the
final timing is off the colour burst/vectors because 3ns or better
accuracy is required, and we then rely on the Vision Mixer strippin off
the old incoming syncs, and replacing ALL source syncs, with a stable set
of internally generated (genlocked) syncs - Sure Sync Sure Burst -
resulting in the mixer appearing as a SINGLE source to VT or Network,
unlike a 'router'/switcher which woulf pass through each individual
source's syncs, and result in a varying size and shape of sync pulse and
burst, which would then adverseley affect a Videorecording following it
(which would rely on consistant sync amplitude and burst amplitude for
auto-chroma, and scH phae to identify the PAL 8 field sequence.
ONLY if a source is OUT OF RANGE should the (analogue) mixer succumb to
passing through the mistimed source - and this should be after preventing
DISSOLVES OR WIPES to or from that source - typically a fade to black and
cut/ or V-fade up would result. Keying might be prevented (should be 8-) )
Hence, where you monitored back at the beginning, with your WFM and
Vectorscope will be affected by the Mixer's design and ProcAmp/not
PGM out should be protected, wheras the AUX buses will probably not be.
Monitoring on the AUX bus therefore allows you to see the source timing,
and also possibly PGM output.
Note that DIGITAL mixers, with sdi inputs ALWAYS have LINE SYNCHRONISERS
due to the nature of sdi decoding. Before that, YOU had to get it right!
TEKTRONIX publish wall charts and booklets about all this. Shouldn't it
also be a part of the basic core curriculum at your college?
--
Phil Spiegelhalter: Phil@xxxxxxxxxxxx
==== Technical Training for Broadcasters =====
*RE CUE Mobile DV Multi-Camera Production and Non-Linear Editing*
.
- References:
- Problem with old(ish) equipment!
- From: David Bibby
- Problem with old(ish) equipment!
- Prev by Date: Re: W/S adverts.
- Next by Date: Re: Children in Need
- Previous by thread: Re: Problem with old(ish) equipment!
- Index(es):
Relevant Pages
|
Loading
