The Captain (captain18) wrote,
The Captain

  • Mood:

Outsmarting Sony (or, Stupid U-Matic Tricks!)

So, part of the reconfiguration of my edit suite/man cave has been to recable everything to use S-Video instead of composite video for the highest quality video I can get from my collection of professional and prosumer gear. This was pretty easy on the S-VHS front since that's where S-Video was developed to begin with, and my Betacam deck does S-Video natively as well. Same with my Hi-8 camera.

This left one format conspicuous by its absence -- my trusty old BVU-870 U-Matic deck which was designed and built before the idea of S-Video (Y/C 3.58) output was conceived of. However, U-Matic machines did use a special 7-pin component output (officially Y/C 688 but generally just called "Dub") which was intended as a short cable, straight-through connection for tape duplication. And while the luminance (the high-detail B&W part of the video signal) is identical in output to modern S-Video, the chroma (the low-detail color part of the video signal) is encoded differently and is not S-Video compatible.

Now, years ago there were two companies who made gear that would transform dub output to S-Video but it was never made in large quantities and is now very hard to even find information about it. But while Googling I found a forum posting with a not-very-detailed account from someone with a different model unit than mine. He was able to attach wires to specific points on the circuit boards inside the unit and obtain a Chroma signal that was actually S-Video friendly. Since I have the service manuals and wiring schematics for my machine, I spent a night pouring over the diagrams until I found exactly which circuit boards I needed to study and some likely locations where I could "steal" the signal. My guesses turned out to be pretty shrewd because I was able to get a watchable signal the first time. It wasn't perfect, of course, and it took a few more tries to find the best points to tap the signal from, but suddenly I was getting video that was visibly better quality than anything I was used to seeing from the U-Matic format!

From there, it took me three trips to the local electronics shop to figure out what I actually needed to wire things up permanently, and I spent a night Googling just to find that special 7-pin connector so I could obtain the luminance signal from its proper location. Ultimately, I decided to use female BNC connectors so I could use my existing video cables to run the separate chroma and luminance signals to the rest of my gear. After a couple evenings with my soldering iron I'm proud to say I have something that looks pretty damn good and works a treat!

Here's where I pull the color signal from the Chroma Decoder board's test points. Only after spending an hour trying to find the test points from the schematic did I realize the service manual also included a map of the circuit board.
The custom cable feeds out the back of the unit to this nice connector.This takes just the luminance signal off the Dub Out port.

To give you an idea of just how much of an improvement this makes, here are screenshots from the composite output (above) and the S-Video output (below) from the same segment of videotape. Note how much "cleaner" the color bars are!

And lastly, I've captured one frame of video using both sources and made a composite using a checkerboard pattern to show off the differences. I deliberately used a low-quality piece of video which has been copied several times from a very old (1973) source tape. This is why the S-Video output appears much grainer than the composite video. But even so, look at how much more detail is in the reporter's jacket!

Damn, I'm proud of this!
  • Post a new comment


    Anonymous comments are disabled in this journal

    default userpic

    Your reply will be screened

    Your IP address will be recorded