Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!uunet!seismo!rutgers!mcnc!unccvax!dsi
From: dsi@unccvax.UUCP (DataSpan R+D)
Newsgroups: sci.electronics,rec.video
Subject: Re: Sync Generators
Message-ID: <792@unccvax.UUCP>
Date: Mon, 17-Aug-87 12:14:47 EDT
Article-I.D.: unccvax.792
Posted: Mon Aug 17 12:14:47 1987
Date-Received: Tue, 18-Aug-87 04:37:51 EDT
References: <225@mv03.ecf.toronto.edu>
Organization: Univ. of NC at Charlotte, Charlotte, NC
Lines: 53
Keywords: genlocking, colour black video generator
Xref: mnetor sci.electronics:1170 rec.video:1711

In article <225@mv03.ecf.toronto.edu>, drascic@ecf.toronto.edu (SpIKe) writes:
> I have an application where two CCD video cameras have to be genlocked.
> 
> There seem to be two routes available:
> 	1) for those cameras accepting a composite genlock signal, we
> 	   can use a distribution amplifier to split the video output
> 	   of one camera into two parts.  The first part is used as normal,
> 	   the second is used as the genlock input to the second camera.
> 
> Question: does anyone know of a (cheap) amplifier that would be appropriate?

        For synchronization pulse distribution, you could even stoop as 
low as using a TL 081 or (preferably) an MC 33074 opamp.  This is not going
to preserve the edges (150 nS rise) obviously, but it will work pretty 
darned well.  Better still, if the camera will genlock to composite video,
use a plain old emitter follower stage. The best, cheap approach, is to 
use a decent comparator (LM 311 comes to mind) and clip off the sync pulses
and distribute that. 

> 	2) for cameras expecting HD/VD, we need a sync generator.  The only
> 	   ones I have been able to find cost $3K - $4K, which exceeds our
> 	   budget enormously.  We do not need something that is broadcast
> 	   quality (it would be wasted on the equipment we have!).  I'm
> 	   amazed at the difficulty/cost we have faced.  After all, every
> 	   cheapo video camera and graphics board has a sync generator built
> 	   in.

      Most cameras I have seen expecting HD/VD will accept composite sync
for HD, and integrated vertical sync for VD. Use a two stage integrator
and a comparator to generate the vertical drive signal, and just clip the
sync pulses for the horizontal. This works very effectively. You could use
a one-shot and some combinational logic if you really **must** have true 
horizontal drive.

      The best way to do this is to have a master sync generator. Ferranti
has a sync generator chip, the ZNA 134 H; as does Fairchild (the number 
escapes me). They are both broadcast quality when used with the appropriately
stable timebase. You can build this for well under $25. If  you need colour
sync generation (black burst, etc.) this will be considerably more difficult
but not impossible to build; the Fairchild chip would then be your choice.
The Fairchild chip runs at 4 Fsc in colour mode, for which either an oscillator
can be made using an ordinary colour burst crystal, or you can always get
14.3182818 mHz crystals through commodity parts junk dealers like DigiKey.
]
     Most cheapo graphics cards use less than full NTSC sync generation. This
would include the TMS 34061, which we currently are using in a commercial
product. I have seen Cohu cameras genlock to even this chip, even though
it does not provide the equalizing pulses (the TI chip, that is). If you
were desparate, you  could probably build what you need out of a few sequential
PAL's or SSI.

David Anthony
DataSpan, Inc