PDA

View Full Version : No legal SDI signal - anyone checked the TC?



pro---studio
06-09-2010, 02:13 PM
Hi,

we had a problem today with a Leitch SDI distributor that produced massive dropouts when it was connected to our TCB. I thought the Leitch had a problem but we checked the unit with a Tektronix WFM 8300 HD.

The unit performed without any problems when connected to another vision mixer. Then we checked the SDI signal of the TCB which showed why the Leitch produced drop outs:

The TCB has a non legal SDI signal. The jitter was far out of legal values and we got lots of audio and gammut errors. In fact the Tektronix didn't stop logging errors of the TCB SDI signal (we tested preview and programm out).

The technician said that he had similar experience with another TCB some month ago. So my question is: Does anybody in this forum checked the SDI signal of the TCB with a high end video measurment tool like the WFM8300HD?

If anyone has a SDI measurment tool could you check this pls?

We will do this also with a second TC unit. Maybe our TCB has a problem.

Regards

pro.

ZachSchuster
06-09-2010, 11:32 PM
I experienced a somewhat similar problem with my TCXD300. We were feeding SD-SDI out to a DA that fed projectors for our graduation ceremonies last month. I was providing the production side and an outside company was providing the projection. I sent color bars and had to leave for a while, during which time they finished their setup and began to tweak the projectors.

When I returned they had placed a override switcher (maybe an Extron... something that took VGA, SDI, analog, etc... and you could do cuts only switches) in the chain between the TCXD300 and the DA. The tech told me they were getting a bad signal off the Tricaster. He suspected it was because of cable length, which was about 100 feet.

Anyway, the new device stabilized the signal. I was not able to do any further testing on it, and do not have a WFM with SDI inputs to test.

So, another question that I would tack on to this topic: is there a legnth limit on SDI coming out of a Tricaster?

pro---studio
06-11-2010, 10:49 AM
Thanks for your reply Zach!

In fact it is no problem to "repair" the SDI signal with a SDI unit that does resync. But you lose another frame and that is not acceptable in IMAG applications.

2.) For me it is not acceptable if I can only use my TCB only in conjunction with a unit that "repairs" the bad signal.

3.) Audio errors (which are produced a lot) cannot be repaired by a SDI synchronizer.

4.) It is not a good feeling if you know that your PGM signal is full of jitter and errors - even if you have a synchronizer behind your TC.

5.) There are SDI specs. If the TC doesn't meet them, then it is not professional SDI switcher.

But for now I hope that this is a faulty SDI card inside my TCB. Otherwise we cannot use it anymore.

BTW:

The specs say that you can use a cable lenght about 300m (which are about 900ft) with a SD SDI signal (SMPT259M).

HD-SDI specs say that the maximum cable lenght without repeaters is 100m (300ft).

That all depends on the quality of the source signal and the cable quality. Some gear auto equalizes. Others don't. Don't forget ground loop issues. In large venues this can be a significant issue.

Regards

Pro.

bbeanan
06-11-2010, 10:56 AM
Have you called tech support? I personally have never had a problem with the output of any of my NewTek gear and actually quite to the contrary I had to send a feed out to press once and I was told by 4 different engineers (from the major networks) that it was one of the cleanest feeds they have seen and wanted to know what gear I was using… they were very surprised to learn it was a TriCaster Broadcast.

There is one thing I do know about NewTek is that they are normally the only ones to comply with specs so it must be an issue with your box.

pro---studio
06-11-2010, 12:42 PM
I hope so.

The thing is - a broadcast engineer here in germany that builds machine rooms for all great networks told me the exact opposite:

He said that no Tricaster machine that he measured had legal SDI signals. Another thing that he said is that you always should use the TC with an external reference signal because the internal timing is very poor. Well - I have no measurement protocols right now but I will have them shortly and let you know the results.

I really don't hope that he told the truth because I like the TC really. But what I like more is professional work which means providing correct signals.

Another Newtek story with correct levels is the poor audio quality of the TCs. A year ago I asked Newtek here in this forum about the exact audio levels of the TC. The last thing I heard was "We'll get these infos for you". But until today (over 1 year later) I have no official answer to that question:

http://www.newtek.com/forums/showthread.php?t=102483

I measured the levels myself and found big issues with the levels. They do not meet the studio specs at all. The TC is consumer level only (-10dB). If you feed a studio level (+4dB) to the TC you will distort the internal amplifiers. So without a pad cable (14dB) you cannot connect the TC to professional mixing desks. But that is another story. I hope Newtek has more experience in video- than in audio processing. Even their own promotion videos are always distorted.
I'll figure out if it is a faulty board on my TC and let you know the results.

BTW - @ bbeanan

You cannot "see" these errors until one machine in the SDI chain produces drop outs because of the high jitter. The signal "looks" clean everytime. It's SDI - digital - that means you have a picture or you don't have it. There is no degrading in SDI signals. You have to measure the SDI signal to see if it fulfills the specs - and by measuring I don't mean to look at a waveform monitor. You must use a SDI qualtity measurement unit, that costs more than any TC to look at the signalflow, echoing, equalizing and jitter. You cannot "see" it. We had exactly this problem: Our digital recorder (Bonsai 1:1 SDI) had no problem withe the jitter signal but the Leitch didn't like it at all and produced drop outs. So if engineers told you they got a "clean" feed, they probably ment they got a good looking picture.


Cheers

Pro