Subscriber Discussion

Analog HD Or HDCVI Etc Crosstalk Problem

aw
alon wallach
Apr 08, 2015

Hi.

Has anyone heard of crosstalk problems in all the analog HD systems.

From analysing how it works, there is an RF carrier of about 58Mhz, modulated with I and Q signals, very much like the old analog TV transmissions decades ago. Because of the common carrier to all cameras, one would expect that two cables running in parralel, especially a short one influencing onbe that comes from afar, will lead to noises.

Alon

aw
alon wallach
Apr 08, 2015

For 720p/60Hz transmission the carrier is at 58.559489MHz

http://www.edn.com/design/analog/4402915/2/Transmitting-HD-video-over-RG-59-cable

aw
alon wallach
Apr 13, 2015

No comment from anyone ?

U
Undisclosed #1
Apr 13, 2015
IPVMU Certified

Are you saying you've seen such alien crosstalk with in sheilded cables with HD-Analog signals?

Avatar
Ethan Ace
Apr 13, 2015

I've been testing CVI and TVI from multiple manufacturers, with cables ranging from the junky micro-coax included with Q-See kits through RG6 factory cables. I have not seen any issues, and our cables are generally laying all over each other, coiled up together (even worse than running in parallel), etc.

I'm not saying it's not possible. I do think it would take a tight bundle of very long cables to create an actual issue, though.

aw
alon wallach
Apr 13, 2015

I do not own or use HDCVI so I was speculating or at best making an intelligent guess, based on the ( undisclosed ) theory of operation.

Ethan Ace claims he has not seen anything bad. This is a good indication, but is not a guaranty that such a problem is not existing. Crosstalk of carrier wave is a physical inevitability, and since the bandwidth is a minimum of 20 Mhz for an 1080p signal, one should expect it, and its not a question of IF but rather of how much. It will be naturally more pronounced at assymetric and longer cable runs with parralel and close routing. It would be an interesting test to be performed by IPVM . I am willing to suggest a testing method and protocol for John.....

The effect if existing would most probably affect all vendors as they probably use similar method of analog RF carrier modulation.

I will try this week to make a quantitative lab measurement ( with RF generator and spectrum analyzer ) of crosstalk or leakage of 58 Mhz signal from a TX RG59 to an RX cable running parralel, per meter of parralel run, for several distances between cables.

If this will yield a result that raises an interference possibility, I will reprt it in this forum.

Its a new technology in the market invented and developed by Chinese scientists, not necessarily experts in RF interference issues......

See a random google search on crosstalk issues in normal CCTV. In wideband analog systems the problem should be much worse. http://www.cctvforum.com/viewtopic.php?f=6&t=34331

Alon

Avatar
Ethan Ace
Apr 13, 2015

Alon, I think you don't realize that I'm the one doing testing at IPVM. You can feel free to suggest a testing method and protocol to John, but it's going to end up with me, anyway. I understand what you are saying and I agree that it's a matter of how much. I do not deny it could be an issue. The reality is that in hours and hours of CVI/TVI tests, we have seen no practical issues.

BTW, CATV design is in my background as well, so these are not new concepts to me.

Also, your inference that "Chinese scientists" cannot be "RF interference" experts is insulting to much of our readership.

And that CCTV Forum thread you mention is discussing multiple analog cameras on a single UTP, an entirely different issue. That is not something we have tested with analog HD.

Finally, this thread likely sat unanswered for five days because no one has reported these issues. We have multiple members using CVI/TVI in the field. No one has reported these issues to us, either privately or in public. So again, you can go on the hunt for them, but as far as being a practical issue, it's a different story.

aw
alon wallach
Apr 13, 2015

Thanks Ethan. Yes , I did not know that you are from IPVM and that you are the testing person.

Secondly, English is not my mother tongue as you probably noticed, but at any rate I never intended to insult anybody. Pls read again what I wrote and note that I did NOT imply what you might have understood about the Chinese scientists. I just wrote that the Chinese scientists ( Dahua engineers are most probably Chinese, aren't they ? I just wrote Chinese because as far as I know ALL analog HD solutions come from China ) who designed the new technologies we are discussing here are NOT NECESSARILY RF interference experts. I still stand behind this assumption. If analog HD technologies would have originated in the US, I would have written US or American Engineers, again without any disrespect.

Let me explain : I must humbly "confess" that just because of a very unique profesional CV I became a seasoned CCTV and RF expert, and that these two expertises are rarely found in the same person or even electronics Engineer. Its just two different disciplines - Baseband, and RF, or carrier if you wish. What these engineers designed is ingenious to anybody that understands how difficult it is to send high bandwidth signals over long distances with no negative effect of the cable frequency slope, attenuation etc. Even SD or old NTSC video is sometimes challenging for 1000 Ft cables, with apparent color loss, wrong colors, bad contours, shadows, noise etc. So HD that needs 6 times more bandwidth and flatness is really really hard over long distances. Almost magic !!!!

I was trying to unravel the mystery or the "magic" since nobody is officially giving the technological concept, or namely this RF modulation technique. I then registered to IPVM several days ago with hope that your reviews will help me to find out. It eventully did because it was the only place that confirmed that this 1080P analog CCTV systems really work at 1000 ft, and even work good, to my surprise. You might take it for granted as another market product, but I was amazed. I then got even more motivated and went and found out that 58Mhz story written by somebody not from Dahua. Assuming what he explained is correct, and it makes sense that he is correct, then there might be a crosstalk problem because its a single carrier frequency going in all cables, unless each camera has a DIFFERENT carrier freq, but this in turn might create a configuration issue and compatibility problems.

You mentioned that you did not see any crosstalk problems in any of your testings, but as far as I saw, your tests were mainly using SINGLE analog HD camera connected to a monitor or DVR through a long cable on a spool. If I am wrong, and you have a test where two camera were working in tandem with two cables running parralel and very close to each other, then pls direct me to the relevant test or article.

The ultimate and most challenging crosstalk test for this new technology would be a strong signal interfering a weak signal, with real world coax cables T-wrapped together over several hundred feet, and this can be done in several ways.

I am willing to bet that two Cat5 cables with baloons running in parralel would show a very strong crosstalk effect. Coax is better of course because of the shielding, but still I anticipate an issue.

And again, if anybody was disturbed from the Chinese Scientists remark, although no offense meant, I do appologize.

Alon

U
Undisclosed #1
Apr 14, 2015
IPVMU Certified

I just wrote Chinese because as far as I know ALL analog HD solutions come from China ) who designed the new technologies we are discussing here are NOT NECESSARILY RF interference experts. I still stand behind this assumption.

Actually, the Inventor is English and is both a Baseband and RF expert.

aw
alon wallach
Apr 13, 2015

Ethan. I just thought after reading again your first post, that you might relate to crosstalk between chroma and Luma within a single picture, while I am talking about crosstalk between TWO pictures from two cameras, two cables etc......

Alon

U
Undisclosed #1
Apr 14, 2015
IPVMU Certified

I have not seen any issues, and our cables are generally laying all over each other, coiled up together (even worse than running in parallel), etc. - Ethan Ace

aw
alon wallach
Apr 15, 2015

To everybody : Reference design from March 2015 of the HDCVI encoder and decoder from the inventor, a 7 employees strong company named Singmai Electronics located in Thailand, technically directed by Daniel Ogilivie who boasts a masters degree in arts history.....

http://www.singmai.com/Documents/aCVi/aCVI_Rev0.5.pdf

http://www.singmai.com/Documents/Literature/Company_overview_March_2015.pdf

There is no RF carrier for the Luma signal. Its just an extended Composite video like the NTSC signal with 30Mhz bandwidth, opposed to the 6Mhz of conventional old NTSC. This means about 30dB slope and loss at the highest frequency for a 300m long RG59 cable, requiring a SUBSTANTIAL to unusual high frequency pre emphasis at camera side, which of course depends on cable length and quality. How is the matching of slope done for each different cable in use, there is no answer.....

The RF carrier I was referring before is only for the COLOR part of the picture, unlike what I understood before, and there are several versions as far as I could understand, 41, 51 and 58 Mhz.....

Its a completely baseband design, and I stand behind my earlier suspicion, that there was no RF deep thought behind this design. For a single point to point video transmission there will be no problems, but running parralel lines, especially with unshielded or poorly shielded cables, will most probably end up with leakage of that color subcarrier between neighboring cables, and crosstalk and noise on the color part of the picture. Remember that NTSC when passing in CCTV was usually modulated with a SEPERATE RF carrier frequency per channel or video picture, but here all color signals of all cameras are at same RF carrier. This leakage can also occur at the DVR input because of poor grounding or ground loops.

And BTW, the reference design gives a lower cost version with lower bandwidth and performance, which could explain the different versions in the market or the minute differences in performance between Dahua and Hikvision etc.

Alon

U
Undisclosed #1
Apr 15, 2015
IPVMU Certified

Alon, believe it or not no one is arguing with you from a theoretical standpoint whether there might be crosstalk in some scenarios.

It's just that no one has reported, including yourself, actually observing this crosstalk. And no one on this forum is likely to be interested until it is observed.

I haven't been able to find any cases on the Internet myself. That doesn't mean they don't exist or that there is never any alien crosstalk, but it does suggest that the effect is uncommon and/or slight.

Furthermore even if some degradation were to be observed and attributed to this, it's unclear what the remedy might be, since the standard can't be changed anytime soon, and not replacing the RG-59 legacy cabling is a key selling point of HD-CVI...

But since you seem both motivated and knowledgable about the subject, I have created a new discussion on the feasibility of creating CVI to TVI adapters. If this can be done somehow cheaply, there might be a buck in it for somebody. Any ideas?

Btw, I will send Daniel an IPVM invite...

New discussion

Ask questions and get answers to your physical security questions from IPVM team members and fellow subscribers.

Newest discussions