Subscriber Discussion

Testing Coax Before Ethernet Over Coax Install

UE
Undisclosed End User #1
May 25, 2017

Is there a test that can be done to determine if an existing coax cable can support Ethernet communications?  

I have read previous posts that speak of taking a set of converters and testing each cable before deciding to proceed. However as a end user, I would like to test the cable infrastructure in advance of issuing an RFQ to provide assuarance my cabling is able to support Ethernet traffic. Thereby  reducing the risk of getting into a project and receiving a big change order to pull new cable or being able to disprove I have a cable problem after receiving a change order request.

Does a piece of test equipment exist to "certify" coax cable like there is for Ethernet cable?

 

(2)
(2)
UE
Undisclosed End User #1
May 26, 2017

Any ideas?

Or would you just a set of translators and a man-lift out?

Avatar
Marc Pichaud
May 26, 2017

Personnaly during IP courses, I advise people to use an Iperf bandwidth test.

So you need a pair of IP over coax units (most can provide network and POE up to 500 m on Kx6 green cable, to checkout on the web) , and  2 Windows PC with fixed IPs, or an IP Monitor which embeds an Iperf client/server. One PC is a server , using a command "iperf -s" (server) the other one using a "iperf - c + IPserver + windows packet size+ return test (as client)

The test , is also useful for Wifi/Wireless bridge .. with real life measurements...not I think that... but real throughput

IP over coax is currently a 100 Mbit/s technology and should return a 94/ 95 mb/s symetric results. First test PCs, with a pointy to pint short EThernet cable to see if firewall and antivirus don't slowdown the bitrates. Then you can test on site.

You should also checkout if the BNC head is well plugged and doesn't have micro power cuts.

(1)
(2)
UE
Undisclosed End User #1
May 27, 2017

Thank you, for the great information.

is there a max cable length when using Ethernet over Coax? 

Would you recommend testing all of them or sample of 20% or so?

Avatar
Kyle Folger
May 27, 2017
IPVMU Certified

It all depends on the the type of EoC. We have successfully used Paluns from Everfocus because we found them to work and are inexpensive. They have a limitation though of 150M on the coaxial cable itself. The one thing to make sure of is that there are no splices in the coax cable and if there are, make sure there are high quality connectors and couplers. If the cable is run exposed outdoors, don't use it as it should be in conduit (personal experience). Also make sure the run is coax end to end. We had one site where every cable was coax end to end except for three of the runs. The baluns were hidden above the drop ceiling, just hanging out on the tile. It was one of those cable parties above the ceiling. Altronix, Veracity, and NVT make some that will go for a longer distance at a higher cost than the Paluns. 

As far as a test is concerned, there is not one that I have tested or used other than testing for continuity or for line tracing. I generally look at the brand of cable, the type of connectors, and the way it was ran. Connectors can be re-terminated. Thankfully, you can generally have better luck with poorly ran coax compared to poorly ran Cat cable. Some think they must pull cable with gorilla strength and far exceed the pulling tension. 

Another good sign is the image quality of existing coax cameras. Pay attention to your run lengths and choose a manufacturer of EoCs to consult with after you have your estimated run lengths. Then specify which product you want use in your RFQ. Get at least one set and test a few lines with that set and document your results with the cabling as is. 

(4)
UE
Undisclosed End User #1
May 27, 2017

i too am concerned about hidden splices, gorilla pulls, and cable type changes not readily viewable. Thus I was hoping that someone may know of a test that would help identify these issues like the OTDR test does for fiber. 

Thank you for the great information and experience.

(1)
Avatar
Kyle Folger
May 30, 2017
IPVMU Certified

Not that I know or have used personally. I generally test and if the cable doesn't work, I go to plan B. Is this coax running to cameras and is your question related to changing to IP cameras or is this for general communication? You don't mention this in your initial post and I would assume this is for cameras. The problem with most testing is you need to unplug one end or put something on the other end of the cable to test against. If they are cameras such as on a warehouse building high up, testing these can be more difficult and time consuming. I would buy a set of EoC devices though to at least check the actual link speed on at least one coax run.

The IPC-8600 does have a TDR option if you get the tester version with this option. It's very basic. It basically says if the cable is shorted or what length the cable is. There are many options for cable type as well. I have this tester with that option. I haven't used it very often other than to see if it does what I described. it seems to work in a crude basic way which is sometimes all you need. If you are expecting 400' and it comes back with 100', there is most likely an issue. For a product like this, I would test against a known length spool of say 500' or 1000' to see how accurate it is. I would make sure the spool is close to the spec of the installed coax.

Maybe someone else knows of a device/tool that does testing with a readout that you can quickly go from cable to cable. Marc's test would work, but if you have a lot of cables, this might become tedious. You would probably want a handheld device for testing. 

(2)
UM
Undisclosed Manufacturer #2
May 27, 2017

Some EoC sets have a cable tester built in. A Panasonic one I worked with allowed you to press a button or give a command in the web UI and it would test the channel and tell you the performance or throughput that it can get. 

I think the newer veracity ones have a similar function - I think it was called. Power star or something... 

 

As for distance length, each manufacturer will tell you their specs for each type of cable. 

(3)
Avatar
Marc Pichaud
May 27, 2017

In old installations better test all cables.... The job is supposed to be done once by somebody... you can not trust in a simple sampling ... like for Twisted pairs.. or fiber

If you want to take the risk : the longer coax cable will be a good indicator: if it's working on 600 meters then you sure knwo it should work on 300 m

You have to test bandwidth max limits AND POE remaining power if you plan to use it (and then think about to protect it by UPS)

 

(1)
(2)
UE
Undisclosed End User #1
May 27, 2017

Thank you for the added information. Yes will likely check all then, also thinking of adding a TDR test to the requirement as to determine the actual run length, we all know cable runs seldom goes in a staight line.

AM
Alastair McLeod
May 30, 2017

Just to add to this discussion from the perspective of one of the manufacturers of Ethernet over coax devices (Veracity):  Our HIGHWIRE Powerstar devices have built-in diagnostics with LED displays which will give you data rate status and POE power level information.  On most coax installations you will get full 100BaseTX performance (300m on RG59, 400m on RG6 and 500m on RG11).   You have full status information about both ends of the cable, from either end, which is very useful sometimes.

The advantage of HIGHWIRE Powerstar is that the same product that you use to provide POE and Ethernet over coax also constantly tests the cable and shows you the results.

(4)
Avatar
Marc Pichaud
May 30, 2017

Alastair.. please do not trust the blinking leds during an IP test... I have also a lot of VERACITY and other brands here. Some brands can reach 1Km and more with decreasing bit rates.. how do you measure that with blinking leds ?

When you have 5 Mbits left, it will still blink at 100 mb ....

The POE status is ok, Veracity gives you the remaining POE budget after powering the Base/Camera units but the Real bandwidth throughputs in IP isn't here .... the leds give you the theorical value... this is a 2 pairs. techno (TP and coax) .. 100 Mb .. that's it...  up to 500 meters if your coax is a good one, only 270 meters with an old one.

Only an IP test can measure what really comes to your recorder. And that is exactly the same, but in in worst - when you are using a wireless bridge...  do you really believe that you have 150 mbps or 300 mbps as mentionned on the datasheet or your switch leds...?   with an IPerf test you will see, you don't have it. Sometimes also, your Anti virus and firewall will drop your theorical performances to half what you expected. (wires, switchs and  PC limitations)

(2)
(1)
UI
Undisclosed Integrator #5
May 30, 2017

I agree Marc.  I wouldn't trust LED's and am unsure how they could properly show throughput on Coax.  IPerf is a great tool for measuring throughput and gives you a real world idea of how much data the Coax will be able to handle.  I've also used the version with the Java based GUI JPerf.  

(1)
(1)
UI
Undisclosed Integrator #3
May 30, 2017

I've converted/installed over 100 transceiver devices successfully in NYC (Elevators mostly) and the tri-state area.  I have a special coax tester that I use to measure distance and determine if the wiring is good prior to converting.

For 100% coax installations I use Vigitron and NVT. As a secondary backup, I specifically use NVT equipment that has the ability to use the 18/2 copper wire (from a siamese cable) to convert to RG59 connectors (2 wire) and I am successful with a 10/100Mb/s connection up to 1000ft (that's my max limit so far to date).

 

I've used Veracity before and it does work. However, based on POE load I've determined that NVT works better with older coax.

 

I hope this information helps you make your decision for IP over Coax/Copper solution.

(1)
(2)
UE
Undisclosed End User #1
Jun 01, 2017

Can you tell us more about your special coax tester?

(1)
UI
Undisclosed Integrator #3
Jun 05, 2017

It's no longer available. However it's a BelMerit LM 50 LAN/COAX

U
Undisclosed #4
May 30, 2017

I have been designing and installing industry standard ethernet cabling infrastructure for more than 20 years as a BICSI RCDD. There is to my knowledge no standardized cable certification test for Ethernet over coax because Ethernet over coax is not recognized by BICSI as a ratified standard. You need to know this and recognize that when you are using non-standard cabling infrastructure as is common in the security video surveillance industry that your success will hinge directly on the non-standard electronics solution you are applying to the cable plant. Cable performance specifications, environmental listing, maximum cable length, channel resistance, connector types etc. all must comply with the electronics (media convertors, switches, lightning suppressors etc.) manufacturer's published performance requirements. Do not make the mistake of assuming that an installed cable plant that successfully operated over manufacturer-A electronics solution will also be successful operating over manufacturer-B electronics solution. Make the effort to retrieve the applied electronics manufacturer's performance specifications and installation instructions and full understand their application before you quote a project. If the requirements are incomplete then contact the manufacturer's tech support line and get printed confirmation of detail in writing. You are otherwise assuming responsibility for application and performance. Most often if you don't have details in writing then it's the same as if the conversation never took place when it comes time to accepting responsibility for performance. I hope this explanation is helpful.    

(2)
(5)
UE
Undisclosed End User #1
Jun 01, 2017

UD4 that is very interesting concerning the BICSI standards and makes me think that just because we get it the camera up and communicating today when it comes time to repair it in two years the same EoC device may not be on market thus may have a problem getting it repaired or communicating again.

(2)
U
Undisclosed #4
Jun 01, 2017

That is correct. There is no standardized cable certification test for IP over coaxial cable. Equipment manufacturer's have created proprietary/ non-standard solutions to enable IP transmission over a number of different media types. To ensure correct functionality the applied network/ media convertor solution must be chosen and the applied cable performance requirements must be field verified. I know Fluke makes/ made a Coaxial tester that will sweep test Coax for performance however you should make sure the values required by the media conversion manufacturer are included in the tester's capabilities before you buy/ rent the tester. This is a lot of work to perform up front but is the only way you can be sure the solution will operate correctly and if it doesn't operate correctly you have a measurable metric to hold the system manufacturer responsible to correct.  

(1)
(2)
Avatar
Marc Pichaud
May 30, 2017

that's why an Iperf test is independant from any manufacturer technology... could be Power Line communications or wireless.. or IP over coax A, B, C  (hey Jackson's)

(1)
UE
Undisclosed End User #1
Jun 01, 2017

Marc, you did say in your first post to use a set of Ip over coax units to do the Iperf test, thus I would fathom that those results would hinge on utilizing the same ip over coax units in the future. As UD4 and others state from my understanding so it's not a true independent test from any manufacturer. Correct?

Avatar
Marc Pichaud
Jun 01, 2017

It's like for a Wireless bridge.. you are using a specific technology (IP or proprietary bridge) and some devices which could be different from a manufacturer to an other

As EOC isn't standardised with the EEEC.. you will need a pair of specific devices, yes sure.

Like some Wireless bridges (never mix brand A with brand B) , but also lots of "less common technologies" like long span, VDSL twisted pair, POE repeaters and port duplicators (with POE 30W, 60W, 95W..)

Better avoid Coax technologies that are POE powered only if a POE device is connected... it requires an additionnal POE splitter...

Veracity /Axis OEM are very good for this up to 500m, simple to use and no need for POE end device. It wil cover most needs.

Iperf is used in IT since 30 years now.. and can not be contested. Analog times are over with I believe, I think it's working , and I want to use a single device to test everything.... Now you need a PC without firewall in admin mode, a good PC card, and lots of freewares, and utilities, and accessories to build a network, setup and test it.

And anyhow, most technology (EOC , PLC, VDSL2, POE Repeaters) are limited to 100 Mb/sec so the results should be symetric and should reach 94/95 Mbit/s on TCP. sometime more when using UDP.

Let's add that EOC is only one technology, using actually proprietary frequencies, ... but technicians need also to test Cat5/6A/7 TP,  VDSL, VDSL2, Long Span or Wireless bridges, optical fiber??  They need also to test remaing bandwidth on the network ... with propre test.... Iperf / Jperf can be used the same way ... so it's a kind of universal test tools now embedded  in some Onvif Handheld devices (no need for a second PC)

 

 

(3)
Avatar
Chris Daniels
May 30, 2017
IPVMU Certified

I don't disagree with your iPerf test to validate existing conditions if the existing network electronics solution will be reused and if no new cabling is installed. However if the network electronics solution and or cable is to be replaced or expanded then you need to research the entire solution before proceeding with quote or installation. Just a word of caution about non-standard cabling infrastructure and applied electronics.  

(1)
(1)
UI
Undisclosed Integrator #5
May 31, 2017

If the original poster picked a good EoC solution that met the specs of the cable and ran an iPerf test through them and the cable it would be about as good a test as you can get.  iPerf actually puts packets on the media and measures the actual throughput the media can handle as opposed to the theoretical bandwidth stated on the spec sheet.  However this method may not be practical for an end user to perform.  

(1)
(1)
U
Undisclosed #4
May 30, 2017

Same design logic applies when applying PoE levels 802.3af and 802.3at to IP network powered equipment. Also be aware of NEC sections 110-26 and 110-27 regarding exposed conductors/ terminals and required working clearances. Many non-standard equipment providers do not comply.  

(1)
Avatar
Ross Vander Klok
May 31, 2017
IPVMU Certified

First off who would mark disagree on this initial question!?!?!  I marked agree to balance it out....  

I would think that Fluke would have some sort of tester for this function?  Our old Flukes could pretty much tell us anything about the coax.   I would suggest sending them an email and asking them if they have anything that would do it.  If they do, but it is too expensive you might want to check with some of the companies that pull wire for a living and see what sorts of tests they would suggest.  Might be worth it in the long run having them in to certify all your existing cable.

(1)
(1)
Avatar
Kyle Folger
May 31, 2017
IPVMU Certified

I added agree to the initial question as well. The problem with testing any of the cable as mentioned above is that you still need to test the cable with the specific EoC solution you are looking at. This is because not all EoC solutions are created equal.

It's kind of like HDMI baluns before HDBaseT was available. Some of them would work at a certain distance better than others with the same cable. There is now more consistency with HDBaseT because it's more of a standard than just a balun. 

(2)
(1)
Avatar
Ross Vander Klok
May 31, 2017
IPVMU Certified

So basically if you have X throughput on your test, depending on what EoC devices you use, it could be X, Y or Z that you actually end up with?  Makes sense Kyle!  I have never used these, but would be very interested in seeing what my 20 year old coax would be capable of.

(1)
(1)
UE
Undisclosed End User #1
Jun 03, 2017

Thank you to everyone that replied! You have given my team a lot information to help make an informed decision, which is exactly why I am a member of IPVM. 

(1)
New discussion

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

Newest discussions