atcomsystems.ca/forum Forums Wiring - Telecom Carriers - Trunks & Services Cabling Telephones Systems Wiring Coaxial Cable

While zipping around the Internet, I came across an article that discusses coaxial cable.

As I read the article, I began to wonder about some of the things stated by the author.

For instance, the author refers to coaxial cable having a copper center conductor. My understanding is that the most commonly used coaxial cable is copper coated steel. Solid copper would be used primarily for satellite installations.

For connectors, twist and crimp connectors are mentioned, but, not compression.

RG11 is mentioned, but, that type of cable wouldn't be common in residential installations.

There are some other things stated that I'm not too sure about.

Here is the link to the article....

click

Is my understanding of coaxial cable not quite right?

Uggh! One of those consumer articles.


Probably better than the person who wrote that piece.

Copper center conductor- coax can be had either way, copper or copper clad steel. Solid copper costs more but you would use it where the cable is used to carry current also, like for a preamp, satellite LNB, switch, etc.

RG-11, we only used that for long drops. The only time we used it in buildings was in the "early days" for large apartment and commercial installations for distribution between closets. After that, when bandwidth increased, we switched to .500 hard line instead. Always had that around.

Compression connectors are the standard. They probably don't mention them because the typical DIY that this article is aimed at wouldn't want to spend the money on tools nor know how to use them. The older hex crimp connectors are OK but are no longer used. They are not water proof for one reason. They had to be used with a boot filled with silicon grease outdoors. Twist-ons should NEVER be used!

What other things aren't you sure about?

Hal, Well said......

I read that runs of RG11 that are less than 100' long can run into problems with reflections. RG6 and RG59 are less susceptible to it. I didn't see that addressed in the article.

Unused ports on splitters should be terminated rather than left open

Even though I don't regularly work with coaxial, I think that my compression tool was worth the cost. I've had crimp connectors fall apart. Compression connectors have been rock solid.

Micro reflections. Umm, yeah, I've heard that you should avoid using less than 2 foot cables to interconnect equipment. The impedance match isn't perfect and a short cable length can cause the reflections that will occur to interfere with the digital QUAM signal. Supposedly anyway. Never heard anything about what you are saying. Wouldn't make sense.

Yes, unused splitter ports should be terminated. By the same token unused wall plates off those splitters should be terminated too- but they never are.

-Hal

Here is the article that looks into RG11. The author describes the reflection factor...
RG11_Cable

Yes, 2 feet and I've never seen that to even be a problem. But that stuff about RG-11? I would like to know where he got that from. Why then shouldn't RG-6 not be shorter that like 20 feet? Sounds like there are some "vidiots" (cousins of audiofools) who think that RG-11 would be the ultimate cable to wire their houses with and that's who the article is written for. Like I said, I haven't seen it used in years. No reason to use anything other than RG-6 size cables.

-Hal

I have a pair of 3-way splitters in use here at home in order to serve three TVs and two WIFI extenders. In one instance, a 2-way splitter is being used to divide a single coax line to support a TV and extender. (Verizon Fios).

I know that splitters attenuate signals. Is there some rule of thumb as to a limit of splitters that can be used on a line before degradation becomes a problem?

I'm using, very roughly, 200' to connect the TVs and extenders.

All cable is Belden 1189A and 1613R.

The only rule that I ever follow is to never feed a splitter with another splitter.

The signal level will drop real fast. Take a 2-port splitter that has 3.5dB loss per port, one splitter will cut your signal level by half, add a second splitter and the signal is cut by half again. You need up with a signal level that is only a quarter of what you started with.

No. You have to know how to do a design. First, you have to know what you have to work with then what you need at each cable box.

For the sake of simplicity lets say that you have +10dbm at the ground block (or video output of the FiOS ONT). +10 is what we would normally shoot for to supply each house. Then figure that at each cable box you want between 0dbm and no less than -10dbm. That range is what cable boxes are designed to operate with. We're going to ignore signal "tilt", which is the difference in levels between the lowest and highest carriers, as well as the return signals from the cable boxes back to the system.

So now it's just a matter of simple math.

If you don't know a 2-way splitter has a 3.5db attenuation from the in to each of the outs. A 3-way comes in two varieties. Either symmetrical or with a "hot"port. Either it's -7 all around or there is a -3.5 port with the rest -7. A 4-way is just two 2-ways. So -7db from the in to the outs.

Now look up the specs for the cable you are using. Look for the attenuation at 1000Mhz, for the 1189A it's 6.55db/100ft. It's given per 100ft so divide it down to per foot (.065/ft). Now determine all your individual cable run lengths and where they go.

Get your pencil and paper out and start with +10. Subtract .065 per foot for the length of cable between the ONT and the first splitter. Subtract 3.5 or 7 for the 3-way splitter. Subtract from that number .065/ft for each cable run from the splitter to the cable boxes. Do you end up with between 0 and -10 at each device? If so all good!

Hint- it's a good idea to make a drawing starting at the ONT of how you have your splitters connected along with cable lengths and where they go. That way you can write your levels at each point in the chain.

-Hal

Memory lane here .
My prior ISP installed a 2 way splitter.
One going to the tv's and one going to the cable modem.
However the one port was a passthu or 0 db and the port to the tv"s had like a 3db loss.
Problem was it kept failing and sometimes when they came out they did not have one to replace it..

What do you call that type of splitter?
I seem to recall it was not called a splitter.

It was not a powered device.

Will be crunching the numbers this evening. 3-way splitters are Philmore CS203NE and a 2-way splitter that Verizon includes with WIFI extenders.

Insertion and Return loss are frequency dependent.

Taking pictures of various hardware pieces along the way for archiving.

Those 3-ways are -7db all around. Any difference between 5 and 1000Mhz is not material. Just figure -7.

I'm giving this to you as a exercise more than anything so you can see how it's done. From what you are showing, you would be hard pressed to to have a 20db loss from ONT to any set, which is what would need to happen to be below the -10dbm threshold.

Incidentally, cable boxes have a huge amount of diagnostics built in and can display the signal levels coming in and going out (return). The cable company can access this feature remotely when you call with a problem. I don't know what cable box Verizon gives you so I can't say how to access this information. I do know that with the old SA boxes we have here it's something like holding down the volume up and down buttons on the front panel while pressing another. Then you get to scroll through a dozen pages or so. You might try Google with your box model.

-Hal

That's called a tap. The thru loss is as low as possible while the tap port can be had in many values. That one is kind of strange though, because when the tap port approaches 3db the thru loss winds up being 3db also. So that would really a 2-way splitter. Are you sure the port connected to the modem wasn't 10db with a 3db through? I've seen that done because the modem return signal is too hot.

You normally would use taps instead of splitters to equalize the levels from each one when they are spaced along a long line. Say you came off an amplifier with a level of +40dbm and you want to supply +10 to each house in a row of houses. Your first house would be supplied by a tap with a value of 30 (-30db on the tap port). Then the next maybe would be a 25, then 20 and so on. You calculate the tap value from the preceding tap thru level minus the cable attenuation to get you as close as possible to +10.

-Hal

Our new boxes are Arris. The older ones were Motorola.

I found the diagnostic menu, but most of the data fields are marked "N/A". I remember a similar screen with the Moto boxes and they always showed actual data.

Back to crunching numbers.

I crunched the numbers into a Microsoft Word document.

Some of the cable lengths were guesstimates (particularly the 2nd floor bedroom).

All connected devices...on paper anyway...are better than the -10db threshold.

Unfortunately, the Arris boxes don't show level readings.

As you can see a few feet +/- isn't going to break the bank. It was nice when they printed the length every foot on the cable jacket.

-Hal

Hal explained that last post very well

.I cabled my house with new RG6.

Home runs from attic to 5 rooms.

All identical cable.

Living room didn't like something?

Possibly the 1 ghz wall port snap in?

Re cabled living room.

Eventually ran RG11 from street to attic. Then from splitter in attic to living room installed more RG11

Seems still a discrepancy in living room only with older cable box and with my HDTV antenna.

Some TVs get more channels?

2 of the tvs are mounted 15' apart. Identical 2017 Samsung 65" HDTV

Go figure?

How hard was it to work with RG11?

I don't think that I've come across it before, and, the way I've seen it described, it reads like an extremely stiff/rigid cable.

When my church signed up for Comcast Internet, Comcast ran Hardline (~1.6 mm diameter) from a tap to an external demarc, and had me run Perfect Flex P6ET77VVWLD from the external box to the phone closet (aka my office).

Total distance was ~140'.

No, not at all. It's just a bigger RG-6. Same construction.

Actually PPC is owned by Belden.

See that cable you installed as well as the 11 series here

https://www.ppc-online.com/products/coaxial?Network=Premises

-Hal

About 15 years ago I ran the video install for the CNN studios in Rockefeller Center for Connie Chung and Paula Zahn.

We did the whole 9 yards - Electrical, Telephone, Data, Audio, Video.

Every desk got a telephone jack (Cat 5E), a data jac (Cat 6A) and a Video jack (RG-6).

We ran RG-6 to every desk, but the feeders to the Video amps were RG-11.

There was no problem working with the RG-11. A little stiffer. A different connector.

It all worked well and came out beautifully.

When I did Microwave we used RG-8. That was different. Much bigger. But still Coax.

Sam

After running the coax at the church, I did a check on PPC. Was surprised to see that the company is owned by Belden.

I asked the owner of the electronics store that I've done business with for decades if he could order a spool of the RG6 so I could use it at home.

He led me to his office and accessed the Belden website where he orders product. The PPC numbers were not recognized by the system.

I suppose Belden is treating PPC as a completely separate entity and limiting availability to industry as opposed to consumers (I don't think that Graybar or Anixter sell it). Wouldn't surprise me if the cable for both brands is produced in the same factories on the same equipment.

Perfect Flex is tri-shielded while Belden 1189A is quad. Belden 1613R is tri, but, the retailer doesn't stock it, so, he special ordered a box for me. He doesn't carry the Belden compression connectors, but, Lowes does.

Belden also bought West Penn, a company that makes cables for the audio industry. If you don't like the competition, buy them out.


We use PPC connectors here. We keep it the same as the cable company although they always cut them off and install new ones even if their own guys put them on.

-Hal

It is not hard to work with, but you obviously need the larger compression connectors. Those connectors may take a little more time to push on to the cable

I have read that RG11, because of its thicker design, has a larger bend radius than RG6, thus making it unsuitable for residential use. Spec sheets do confirm the larger bend radius.

For loss calculations, I suspect that couplers would attenuate. I have (4) 90 degree bends installed (cable management) as well as a coupler (baseboard jack). Whatever the amount of loss added in, all three Arris boxes are functioning properly.

Ugggh! Stop reading that internet garbage! Unsuitable for residential use? Why, do you live in a tiny house? All cable has a minimum bend radius including 6 size cable.

Couplers? You mean F-81 barrel connectors like on wall plates? Yes they do have a loss but it's so minimal that it's not an issue.

What I have no use for are those 90 deg connectors. I believe the loss is more than a barrel, dunno. I just never liked them, maybe because a length of cable is so much cheaper.

-Hal

Nope...looking at manufacturer documentation.

Belden 1189A: Minimum Bend Radius - 2.9"
Belden 1523A: Minimum Bend Radius - 3.9"

While 3.9" is far better than Hardline cable's bend radius, it can be cumbersome to run in tight/confined areas.

I can't imagine where you would be trying to stuff it into that that would be a problem. If it is, it just might be that an 11 size cable shouldn't be used or you need to make allowances for it.

-Hal

If i were to do anything with 11, it would be between the ONT and the first splitter. However, with attenuation calculations for all branches not exceeding -10db, the need isn't there at this time. When Verizon replaces the existing Tellabs ONT with an Alcatel-Lucent, I'll revisit.

Here is a picture of the Tellabs ONT in the basement.

Shouldn't make any difference,the output level should be the same. The carriers are generated by the ONT and I wouldn't be surprised if they were adjustable. You would just have to get a Verizon tech who knows what you are talking about.

If you are really serious about this save up your lunch money and get something like this:

https://www.markertek.com/product/b...8vsb-digital-qam-catv-analog-tv-analyzer

It's a nice installer's meter.

-Hal

Nice!

Would need to go lunchless for a bit of a while though.

The one I used to have was about $2500. That was back in the days of analog and it became a boat anchor. Broke my heart to take it to the scrap yard a couple of years ago, probably got $1 for it. If I was back in that business I would probably get that one that I showed you or something similar.


-Hal

I'm not adverse to splurging here or there for a test set or tool that can come in handy despite my not working in the communications business at this time. It might not be inexpensive, but, it's far less expensive than an Ethernet Certifier.

And more useful. You don't need to certify ethernet for it to work. But you are flying blind when doing cable work without knowing levels and that can definitely affect whether it will work or not.


-Hal

You could get an HST-3000C off Ebay and find an ethernet module for it too. It's probably a generation old, but it should handle anything you throw at it, with enough modules and software unlocks if viavi will still give out codes.

One more log on the fire here....

Different manufacturers tout how good their connectors are (referencing compression connectors). Knowing that signals utilize the center conductor (either copper or copper covered steel) and not necessarily the shield, is there really much of a difference between brands?

Comcast uses PPC cable & connectors.
Verizon uses Digicon/Arris brand connectors. Not sure which coax they currently use. When I had FiOS installed here at the house back in 2007, Verizon's contractor used TFC-T10 coax.

I don't think so. The big difference came from switching to compression from hex crimp. Not only a little better shielding but the ability to be used outdoors without boots and grease. But between compression connector manufacturers I think it's all hype.

The choice to use one over another is probably a decision made because of suppliers and price. I have always used PPC but do have some Digicon. Digicon may be easier to get if you don't do business with CATV suppliers. So I wouldn't worry about it, they are both good as long as you use the proper tools.

-Hal

I had my suspicions that would be the case. I purchased a small supply of both brands of RG6 connectors. That should last me well into the foreseeable future and beyond.

The standard Klein Tools RG6 connectors are not marketed as weatherproof. The company sells a version for outside installations. The problem with it is that it isn't designed to be used with quad shielded cables. Dual and tri shielded fit perfectly.

Before I write-off climbing onto the roof for good, I want to replace the old RG59 feed to my FM antenna with RG6.

I have an external antenna mounted transformer up there already and a spare in the basement. The new cable (including the PPC connector and boot) is ready to go.

TWC/Spectrum has standardized on PPC connectors here for many years.

PPC hardware seems to have become the go to for a portion of the industry. Verizon is bucking the trend. In a way I can understand it. Digicon/Arris supplies set top converter boxes to Verizon for FiOS customers. So, go all in with Arris I suppose.

Sure. When the sales person comes around you know it's going to be "buy everything from us and I can get you a better deal".

-Hal

Oddly enough, the original FiOS STBs were made by Motorola. Not sure if there is any connection between the two labels. Even back then (2007) Verizon used Digicon compression connectors.

It is a stretch, but Arris/Digicon is a division of Commscope, which absorbed the Systimax structured cabling product line from Avaya. In a way, that would make Arris/Digicon a distant relative to the Western Electric nameplate. Don't know if that figured the selection of Digicon by Verizon.