The Virtual Engineer - Broadcast Engineering Forums

Hi. Recently acquired BVH-2000 (NTSC). Along with it there was no TBC (no plug-in 6-9 cards, BVT-2000 or something). Unit works, but there is a problem with floating colors (video). I'm sorry for asking this stupid question but this is because of the lack of TBC or is it a servo failure? And if it is the absence of time base corrector, will this VTR work with BVT-800 or BVT-810?
Thanks.

Yes indeed, the unstable color is because there is no time base correction. The BVH-2000 needs to run either with an external BVT-2000 using a special umbilical cable, or the four-card BKH-2015 TBC (assuming that the machine isn't one of the very early versions).

If there were a servo failure, you wouldn't be getting a stable enough picture to record -- so what I see on your YouTube post is actually a pretty hopeful sign. I doubt that a BVT-800 or -810 will work: because of the length of the video track and the distance around the drum between guides (about sixteen inches), the amount of time base error is considerably greater than on a U-matic machine -- so I don't think the 800 series would have enough memory. Moreover, the TBC needs to return signals back to the VTR to provide advance reference and other handshaking. And if you can get a BVT-2000 or BKH-2015, the dynamic tracking head will let you do variable speed or still playback -- which definitely won't happen with any other TBC.

Jeff

Thank you so much for detailed answer! Yes, this BVH-2000 allows to install 4 cards on it. I asked about the BVT-800 because I can't find anything better on Ebay. And I'm not sure that BKH-2100/2150 kit is even possible to find separately from the VTR (maybe I'm looking in the wrong place?)
Also, I discovered that if I connect a video signal with black fields (Servo Ref switch to "Ext") to the "Ref In" input and then I play the video, colors become a little more stable. I'll try to buy sync generator to make sure.
P.S. My DT head is almost dead - the signal output level is less than 80%, still-frames lose color and vertical sync (but they are clean nevertheless). Also, head drum makes strange noises when running on it. The R/P head is more than alive instead.

VCD wrote: ↑

Wed May 08, 2019 4:55 pm

Yes, this BVH-2000 allows to install 4 cards on it.

Even very early revisions of the BVH-2000 had the four slots assigned for the plug-in TBC, but you would need to modify some of the other cards before it would work. Only a machine above a particular serial number would take the BKH without doing field modifications. Just seeing the empty slots is a bit misleading.

I'm not sure that BKH-2100/2150 kit is even possible to find separately from the VTR (maybe I'm looking in the wrong place?)

I don't think I've ever seen it sold separately from a machine. As you've discovered, a BVH-2000 without time base correction isn't very useful.

I discovered that if I connect a video signal with black fields (Servo Ref switch to "Ext") to the "Ref In" input and then I play the video, colors become a little more stable. I'll try to buy sync generator to make sure.

That doesn't surprise me, but even supplying external reference to the machine will never be sufficient to stabilize the playback. One of the problems that's fundamental to videotape recording is that there is always some mechanical imprecision with the tape and head motion; the playback head will never precisely duplicate the path that the recording head wrote. Minor differences in the slant guide around the lower drum, entrance and exit guide position, drum velocity and phase, head tip projection, and a host of other things add up so that the playback signal recovered from the tape doesn't exactly align in time with how the original video was recorded. These discrepancies create constantly changing delays in the playback signal, which (if extreme) can be visible as horizontal instability; even very small changes in delay become phase shifts at the color subcarrier frequency -- which become drifting hue in the picture.

What a time base corrector does is to store incoming video in a small digital buffer (enough for eight video lines, on a BVH-2000); the sync at the beginning of every horizontal line is compared to your stable reference, and the pointer that's grabbing data from the buffer to play out automatically adjusts to offset the error, so that the feed from the tape is given a correcting advance or delay to compensate for the mechanical error. It also does a much finer correction by comparing the color burst coming off the tape against the color burst in your reference signal, and making the corresponding correction on the output. (If you remember the old Ampex 2" quad machines, the corrector for horizontal error was called AMTEC, and the corrector for color was COLORTEC.)

Doing all this on a BVH-2000 is not a trivial exercise, because the length of the video track is extremely long (about 16"); Chuck Hintz, my instructor at the Sony maintenance training school, used to say that making the playback head exactly scan the same track that the record head laid down takes the same level of precision as driving from New York to LA without touching the steering wheel and without drifting out of your lane. One of the reasons that the machine needs a specially designed TBC is that the machine actually needs to be playing about four video lines early, compared to your house reference (and your other sources). Ideally, your stabilized playback video should be coming from somewhere around the middle of the eight-line buffer memory in order to allow for error in either direction. So the TBC produces an advanced sync reference that drives the tape machine's servos. There are other signals that go back and forth, including information from the demod circuits that trigger the dropout compensator. There's an awful lot going on!

P.S. My DT head is almost dead - the signal output level is less than 80%, still-frames lose color and vertical sync (but they are clean nevertheless). Also, head drum makes strange noises when running on it. The R/P head is more than alive instead.

Don't give up on your DT head just yet. The DT-04 board that controls it depends on a lot of feedback from the time base corrector. If it's not getting that feedback, and if there is no correction going on, the moveable head tip is going to go nuts trying to produce a stable picture -- and it will make a very noticeable chattering noise. Because the control system is sending the head all over the place, it's not surprising that the recovered RF level is low compared to the R/P head. Finally -- if you do get a TBC designed for this machine, DT chattering can be caused by dirty or worn sliprings, which are much cheaper and easier to replace than the upper drum (you need special tools and an alignment tape to replace the drum).

Realistically, I would say your best hope would be to either find a BVT-2000 with the umbilical cable, or another BVH-2000 that already has a BKH card set. A machine with a bad head would still be a good source for the cards (plus other spares).

Jeff

PID_Stop wrote: ↑

Thu May 09, 2019 7:47 am

(If you remember the old Ampex 2" quad machines, the corrector for horizontal error was called AMTEC, and the corrector for color was COLORTEC.)

Just to be clear: AMTEC and COLORTEC are pre-digital systems, and worked by passing the RF signal from the tape through voltage-controlled variable analog delay lines. In AMTEC, house horizontal reference would reset a ramp generator at the beginning of every line; as soon as the raw playback signal from the tape hit a horizontal sync pulse, a sample-and-hold circuit would capture the voltage coming from the ramp generator. The result was a DC voltage that represented how much the tape playback was advanced or delayed compared to the stable reference. This voltage was applied to control the delay line, and the result was to compensate for the error. COLORTEC worked the same way, but performing much finer correction based on the color subcarrier.

Considering that this was done in the early transistor days of the late '50s and early '60s, well before even analog ICs, it's a darned impressive feat!

The reason quad machines could get away with the fairly limited correction capability of analog delay lines: unlike 1" helical scan machines like the BVH-2000 where a single very long track contains 262 lines (one field), one sweep of a quad head only recorded a sixteenth of the field, or about 17 video lines, and was about an inch and a half long. The quad format worked because it minimized mechanical errors through physical robustness and avoidance of difficult-to-achieve mechanical tolerances.

Jeff

Thank you again for the answers! Thanks to your explanation, I finally understand why the colors are floating on this machine. But it's just funny to know while having an early 70's Soviet b/w VTR also using a 1" tape with very similar tape mechanism design (single-head helical scan drum with almost 360° grip) that is surprisingly able to write in SECAM with stable colors (on a originally black-and-white machine that doesn't have any IC!).

What a time base corrector does is to store incoming video in a small digital buffer (enough for eight video lines, on a BVH-2000)

Recently I bought the manual to the BVT-800 and there is really not specified that this TBC is able to work with the BVH-2000 (it's designed mainly for the BVU-800 and BVU-200 U-machines, however, there is an option for "communication with the VTR other than the BVU-series VTR which is equipped with a capstan servo system"). Also, it can work with Dynamic Tracking system and also has a corresponding connector for Sony's CCZ-A (CCU) 26 pin cable.
But most importantly, it has 32-line memory for Y & C signals, according to that manual. So, I think to try BVT-800 with my VTR. Hope there's a chance that this will work together.

Don't give up on your DT head just yet. The DT-04 board that controls it depends on a lot of feedback from the time base corrector. If it's not getting that feedback, and if there is no correction going on, the moveable head tip is going to go nuts trying to produce a stable picture -- and it will make a very noticeable chattering noise. Because the control system is sending the head all over the place, it's not surprising that the recovered RF level is low compared to the R/P head.

Thank you so much for the hopeful information! I was almost disappointed with my machine when I saw the RF output level from Play-head.

VCD wrote: ↑

Sun May 12, 2019 4:09 pm

But it's just funny to know while having an early 70's Soviet b/w VTR also using a 1" tape with very similar tape mechanism design (single-head helical scan drum with almost 360° grip) that is surprisingly able to write in SECAM with stable colors (on a originally black-and-white machine that doesn't have any IC!).

There are ways to mask chroma instability enough for video to play on a monitor -- generally involving some sort of heterodyne circuit. The portable Sony BVH-500 had an optional heterodyne adapter that allowed the machine to make a good enough playback to have confidence in the recordings, but the overall signal still had significant time base errors, to the point where it was not stable enough for broadcast or to feed into a production switcher. At one point, Sony had a single plug-in card for the 2000 that did heterodyne correction, but I don't think they ever brought it out of the prototype state into production.

Jeff

Thank you very much, sir! I bought the BVT-2000 since BVT-800 only worked with color-under systems. And it works perfectly!
https://www.youtube.com/watch?v=tk4kkFS3ocg
One thing I haven't been able to fix is sometimes the color hue suddenly slides off and then can't be fixed by knob on the BVT front panel. What could be the reason of such behavior?

Any suggestions with BVT's color tint problem? I still have not figured out what cause this problem, except that the bug is in the TBC (not in the VTR).

VCD wrote: ↑

Thu Jan 09, 2020 4:40 am

Any suggestions with BVT's color tint problem? I still have not figured out what cause this problem, except that the bug is in the TBC (not in the VTR).

Hue is controlled by a DC voltage that varies between -5 and +5 volts, with ground (0v) giving you the nominal preset. The front panel knob feeds a variable DC voltage to the VAR/PRESET switch below, sending out either the variable control or ground (if the switch is set to "PRESET") as the local control. Try exercising this switch to see if that changes things. From there, the local control and the remote input from a rear panel connector go to a large slide switch (S13) to select either local or remote control. Again, try exercising this switch to see if that helps things any. Slide switches are notorious for getting dirty, and that sounds exactly like what's going on here.

The selected output of remote/local switch S13 goes through an op-amp configured as a voltage follower that buffers the signal before it gets sent out a ribbon cable (CN1 on the front panel board) to hit the motherboard. Make sure the ribbon cable connector is well seated, and that wiggling the cable doesn't make the hue shift. It shouldn't... that's a relatively unlikely problem.

From the motherboard, the signal goes to the SG-28 sync generator board. Try re-seating this board to make sure the edge connector hasn't gotten dirty... that's actually a fairly likely cause. Within the board, things get a bit more complicated... but the intermittent nature of what you're seeing is more consistent with a dirty switch or a dirty edge connector. If the things I've suggested don't solve the problem, the troubleshooting gets considerably more intense. Let's hope for a dirty switch...

Jeff