Hello Doc,
Anything I could add to your upstreams would be redundant. My experience on the other board has taught me not to break my arse going head to head with vested interests on public fora. No matter how profoundly informative (right or wrong), my words merely get shunted to ground. No thanks.
Investors don't want "information" posted on their stocks' sites. Rather, they want to read things that "boost" their stock prices. Anything else, and they go into tantrums.
But I *would* like to comment on this matter concerning removing the technician from the MSO calculus. Who said this? Can he pass a drug test?
I don't mean to bash people who work for a living stringing coaxial cables and wrestling with bi-di amplifiers. My own background, in fact, on the other side of the fence, fosters a kind of affinity with those folks, so that is not at the heart of what I'm about to post.
Despite the advances that many MSOs have made in their most recent builds, the majority of all existing cable plant remains extremely diverse, and demanding of extreme care due to the sensitivities of imprecise analog constructs that date back to the turn of the "20th" century.
This extends right down to the side of the house and into the home. Some of the topics below reinforce this, straight from the mouths of those who work in the cable industry. The following is from today's SCTE email list, which I try to read every day. There's nothing particularly onerous, or forgiving, for that matter, about this one.
SCTE = Society of Cable Telecommunications Engineers.
Remove the technician, indeed!
Enjoy,
FAC
---------------------
SCTE-LIST Digest 2110
Topics covered in this issue include:
1) Re: Return path "cable modem" amplifier (DOCSIS) by "Chris May" <chris@westernet.net>
2) Tri-Shield verses Quad-Shield Drop Cable by Keith Boyd <keithboyd1@home.com>
3) Re: Tri-Shield verses Quad-Shield Drop Cable by "Chris May" <chris@westernet.net>
4) .412 Cables by tvrex <tvrex@earthlink.net>
5) Re: .412 Cables by Brent Nunley <brent@lcc.net>
6) RG-11 by "Steven Attaway" <srattaway@hotmail.com>
7) Re: RG-11 by Brent Nunley <brent@lcc.net>
8) Re: .412 Cables by "Norman Cheatham" <Norman.Cheatham@worldnet.att.net>
9) Re: RG-11 by "Chris May" <chris@westernet.net>
10) Re: Return path "cable modem" amplifier (DOCSIS) by <RHranac@aol.com>
11) Re: Return path "cable modem" amplifier (DOCSIS)by "Jerry K. Thorne" <jthorne@qrf.com>
12) Re: Tri-Shield verses Quad-Shield Drop Cable by <RHranac@aol.com>
13) RE: .412 Cables by "Steve Allen" <stallen@cisco.com>
14) RE: .412 Cables by Doug Stanfield <DOUGS@oceanic.com>
15) Re: Return path "cable modem" amplifier (DOCSIS) by Mark Millet <mmillet@cisco.com>
----------------------------------------------------------------------
*My* workaround for the tap by the amp is as follows: use a sufficiently low value tap plate (29-26-23-etc) so that the modem, allowing for drop loss and a 2-way, will have an 8 dB margin or so. Use quad drop if possible. Use a 2-way to go right to the modem. Use quad if possible. Put a sufficient FAM attenuator on the other leg, a data trap (HPF) and whatever splitter you need to get down to
your spec'd level at the sets. Tell your customer that the data line is just that, not for TV. Or tag it "data only".
For your modemless customers on that same tap, use FAM's there too. Or cut in another tap of the proper value....c
Mark Millet wrote:
> >From: "Jerry K. Thorne" <jthorne@qrf.com>
> >
> >Brian,
> >
> >I had a cable system in Virginia call me one day with a modem that would not
> >put out enough signal to overcome large tap losses. He asked us to build
> >him a passive forward drop amp with 20 dB gain in the reverse path only to
> >place in the line directly in front of the modem. It gave the modem the
> >range it needed to get online.
> >
> >Jerry K. Thorne
> >Applications Engineer
> >Quality RF Services, Inc.
> >jthorne@qrf.com
>
> For those of you considering a similar project in the garage, keep the following in mind:
>
> The DOCSIS cable modem will transmit between +8 and +58 (sometimes +60) dBmV, and that 20 dB gain block right smack at the modem will be generating +80 dBmV if it is doing what you think it should be doing. If it is operating as a normal CATV part does, it may give at best +72 dBmV, and spurs from DC to light, somewhere around -20 dBc, and not where you want them either...
>
> Last time somebody pointed to a +80 dBmV RF source, I ran the other way.... Thus, such a gizmo is only useful near the ground block, and for the purposes of raising the noise floor in the entire building up to inconvenience levels.
>
> Granted, there are plenty of non-DOCSIS proprietary modems out there than give up the goat at +48 to +52 dBmV, and in this case a properly biased and installed hybrid chip could reasonably pump out +70 dBmV, with low level harmonics.
>
> BTW: Be sure to rewire the entire house with Quad Shield, and high power RF splitters from Mini-circuits, or the CPD from the saturated cores will toast the remainder of the network. (P=V^2/R)
>
> IMHO, if you need more than +60 dBmV output inside the home, it is time to re-design the home wiring, remove tap face plates above +29 dB, and anything with more than +32 dB of flat loss from the housing to the tap port (including DCs)... Downstream architectures with +49 to +55 dBmV RF output, and +17 dBmV out of the first tap (with little or no cable loss), are simply broken for two way designs.
>
> -----
> On the 950 foot RG-11 drop, that length is clearly fine, it comes on a single roll. <groan>
>
> Mark Millet
> CSE Service Provider Engineering
> Direct: 408 526 6275
> Cisco Systems Inc. Fax: 650 526 5807
> 170 West Tasman Dr Pager: 800 365 4578
> San Jose, CA 95134 mmillet@cisco.com
-----------------------------------
List;
I was asked by an in-house accountant why we have not switched to Quad
Shield cable
in our system. My answer was that Tri-shield is adequate for our systems
requirements and we would have another fitting type to stock, etc,etc,etc.
I believe that we were ordering drop cable and the vendors only have Quad
available.
I would like to know what most systems use out there, Tri or Quad and why.
We have over 1600 miles of fully loaded two-way plant, channel loaded to
700 Mhz with 8000+ cablemodems, 25,000 Digital Cable boxes, etc,etc and we
have not faced a real reason to switch to Quad Shield drop cable as of yet.
We are in a suburb environment and would understand using Quad if we were
in a city environment.
Any feedback on this is appreciated.
------------------------------
>From the small system perspective, we use quad in the noisy environment (ie, the
headend) and for superdrop (higher than normal RF levels feeding or inside a
lockbox ). So it's not a standard installer thing. That would be a nightmare.
Can you imagine trying to train a guy (sorry, a person) who refuses to recognize
the differences between plain, flooded and messenger already? AND double
his'n'her connector choice? You'd need one of those log tables previously
mentioned to calculate the chance of a drop being done right!
So we only need a roll every now and then, and keep them and the fittings in the
headend.
You apparently have no broadcast considerations (TV or FM). We do.
Your key words are right below: if you don't need it, don't buy it. The
beancounter will be happier, and if you are supervising the final installation
product, so will you....c
Keith Boyd wrote:
> List;
>
> I was asked by an in-house accountant <snip>
------------------------------
Just reading the messages on today's List concerning RG-11 and .412 cables.
With all of the .412 cables still out there today and the number of people
who have installed it and replaced it, I wonder that no one asks why all of
the rest of our feeder and trunk cables are manufactured in quarter-inch
sizes:
.250; .500; .625; .750; .875 and 1.000. Why was .412 cable even
manufactured? The answer is one you probably wouldn't have ever
guessed....Rex
------------------------------
Message-ID: <39EA13D6.A59EF029@lcc.net>
Date: Sun, 15 Oct 2000 15:30:14 -0500
From: Brent Nunley <brent@lcc.net>
MIME-Version: 1.0
To: scte-list@relay.doit.wisc.edu
Subject: Re: .412 Cables
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Don't forget .860. :P
B.
------------------------------
From: "Steven Attaway" <srattaway@hotmail.com>
To: scte-list@relay.doit.wisc.edu
Subject: RG-11
Date: Sun, 15 Oct 2000 13:59:33 PDT
Mime-Version: 1.0
Content-Type: text/plain; format=flowed
Message-ID: <F22PdLFYxc9NUpuabjK0000401f@hotmail.com>
I think the point is that why would you use such a small cable at all with
today's architectures??? With most systems going to 750mhz. <or in our case
860mhz.>...why would any system leave that old stuff around <I know, I know,
the bean counters won't let you replace it :-(>?
I remember a thread a while back about the problems with using P1 .500 for
90VAC powering. I can't imaging using .412 much less RG-11 for long runs.
I did use LONG RG-11 drops when I ran a 300mhz. rural system several years
ago, but that won't work with the expanded bandwith needed nowadays. Like
Tom said, you should run hardline back to the house. What are we to
do...have a 3 LE cascade all ran by RG-11?
Steve
----Original Message Follows----
From: "JR" <mcneilj@interlynx.net>
>>If you take a look at the comparison between Neal (Northern Electric
Aluminum) 412 and current Series 11 (RG 11) cables you may be
surprised what you find:<<
***************************
Steve Attaway
www.geocities.com/teckstuff
***************************
_________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at hotmail.com.
Share information about yourself, create your own public profile at
profiles.msn.com.
------------------------------
Message-ID: <39EA4A45.989A61F9@lcc.net>
Date: Sun, 15 Oct 2000 19:22:30 -0500
From: Brent Nunley <brent@lcc.net>
MIME-Version: 1.0
To: scte-list@relay.doit.wisc.edu
Subject: Re: RG-11
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Unfortunately, some of us are still at 330....sigh.
B. a.k.a. dinosaur tech
------------------------------
Message-ID: <000401c0370f$d5655940$ed024d0c@presario>
From: "Norman Cheatham" <Norman.Cheatham@worldnet.att.net>
To: <scte-list@relay.doit.wisc.edu>
Subject: Re: .412 Cables
Date: Sun, 15 Oct 2000 21:24:27 -0400
MIME-Version: 1.0
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
I think .412 was manufactured for the military in WW2 as comm. cables,. =
It was light weight, and messenger was easy to run and when the war was =
over, .412, 500, and .750 became the norm for CATV in it's =
infancy....But there was so much left over and the dies, tools to make =
it still in place, they just kept making it...and in its day, it was =
trunk...with twin lead, and .59 as drops and feeder =
respectively.......correct me if I am wrong.... and I know there is =
tons of it still in use, especially down long driveways, and as feeder =
in places it difficult to replace.....(i.e. Boca Raton and Palm Beach, =
FL) where there is shellrock, and easements with picky customers that =
dont want digging, and the .412 cable tdr'd ok.....just more actives in =
those places to compensate for the cable loss...in use in 550 and 750 =
systems...
------------------------------
Message-ID: <39EA624A.B75C813@westernet.net>
Date: Sun, 15 Oct 2000 20:04:58 -0600
From: "Chris May" <chris@westernet.net>
MIME-Version: 1.0
To: scte-list@relay.doit.wisc.edu
Subject: Re: RG-11
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Don't lose hope. There's a small mountain system near here, that will
remain nameless, that's only 300mhz. Owned by the metro district. All
UG, spectacularly high-dollar homes, only built a few years ago (OK,
maybe ten). No way are they going to dig up these yards!
They were losing to DBS, despite ordinances requiring whole-house
prewires to the ped for cable enforced by the metro bldg inspector
(doncha wish we could all have that?). All design reviewed by staff of
course...
So what did they do? Went all digital. Or at least substantially,
since there are no local channels, due to the mountains. Plenty of
bandwidth for all now. Plus the added box revenues. No unhappy
homeowners due to dig-ups...c
Brent Nunley wrote:
> Unfortunately, some of us are still at 330....sigh.
>
> B. a.k.a. dinosaur tech
------------------------------
From: <RHranac@aol.com>
Message-ID: <9.bda269c.271bc021@aol.com>
Date: Sun, 15 Oct 2000 22:21:21 EDT
Subject: Re: Return path "cable modem" amplifier (DOCSIS)
To: scte-list@relay.doit.wisc.edu
MIME-Version: 1.0
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit
In a message dated 10/14/2000 10:47:40 PM Mountain Daylight Time,
mmillet@cisco.com writes:
<< Last time somebody pointed to a +80 dBmV RF source, I ran the other
way.... Thus, such a gizmo is only useful near the ground block, and for the
purposes of raising the noise floor in the entire building up to
inconvenience levels. >>
Actually, +80 dBmV is just over one watt (1.33 watt), not really all that
much power. Still, Jerry brings up a good point about getting CATV hybrids to
operate at that lofty level. Most tend to hit an upper crash point around +70
to +72 dBmV. Several years ago while doing some lab research at Jones
Intercable to see just how much RF output could be achieved with various
amplifier technologies (push-pull, feedforward and power doubling), I decided
to do a little experiment of my own. I managed to modify a Jerrold/GI
QuadraPower line extender to reliably output +80 dBmV on one channel, while
keeping third order 60 dB down. To make things more fun, I did this in the
70cm amateur band, and used the modified line extender to transmit a "fast
scan" amateur TV signal (conventional 6 MHz-wide NTSC) some 22 miles through
the air to another amateur. This was all perfectly legal, by the way (I'm a
ham operator, and operation was confined to ham frequencies).
Regards,
Ron Hranac
Cisco Systems
------------------------------
Message-ID: <018201c03719$4c71a440$142c0ad8@jkthorne>
From: "Jerry K. Thorne" <jthorne@qrf.com>
To: <scte-list@relay.doit.wisc.edu>
Subject: Re: Return path "cable modem" amplifier (DOCSIS)
Date: Sun, 15 Oct 2000 22:32:14 -0400
MIME-Version: 1.0
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
Ron,
The MHW7205C power-doubled chip will get up in that range. Two of these in
a quadra-power QRAM will get it 2.5 dB higher.
JKT/QRF
-----Original Message-----
From: RHranac@aol.com <RHranac@aol.com>
To: scte-list@relay.doit.wisc.edu <scte-list@relay.doit.wisc.edu>
Date: Sunday, October 15, 2000 10:22 PM
Subject: Re: Return path "cable modem" amplifier (DOCSIS)
>In a message dated 10/14/2000 10:47:40 PM Mountain Daylight Time,
>mmillet@cisco.com writes:
>
><< Last time somebody pointed to a +80 dBmV RF source, I ran the other
>way.... Thus, such a gizmo is only useful near the ground block, and for
the
>purposes of raising the noise floor in the entire building up to
>inconvenience levels. >>
>
>Actually, +80 dBmV is just over one watt (1.33 watt), not really all that
>much power. Still, Jerry brings up a good point about getting CATV hybrids
to
>operate at that lofty level. Most tend to hit an upper crash point around
+70
>to +72 dBmV. Several years ago while doing some lab research at Jones
>Intercable to see just how much RF output could be achieved with various
>amplifier technologies (push-pull, feedforward and power doubling), I
decided
>to do a little experiment of my own. I managed to modify a Jerrold/GI
>QuadraPower line extender to reliably output +80 dBmV on one channel, while
>keeping third order 60 dB down. To make things more fun, I did this in the
>70cm amateur band, and used the modified line extender to transmit a "fast
>scan" amateur TV signal (conventional 6 MHz-wide NTSC) some 22 miles
through
>the air to another amateur. This was all perfectly legal, by the way (I'm a
>ham operator, and operation was confined to ham frequencies).
>
>Regards,
>Ron Hranac
>Cisco Systems
>
>
------------------------------
From: <RHranac@aol.com>
Message-ID: <a.38d8c42.271bc338@aol.com>
Date: Sun, 15 Oct 2000 22:34:32 EDT
Subject: Re: Tri-Shield verses Quad-Shield Drop Cable
To: scte-list@relay.doit.wisc.edu
MIME-Version: 1.0
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit
For many years I've recommended the use of tri-shield drop cable as the
minimum shielding configuration for two-way systems. Quad-shield can be used
in areas of strong ingress--for instance, residences located near paging
transmitters, two-way radio repeaters, local broadcast stations, and so
forth--but otherwise is usually not necessary.
In general, here's the typical shielding performance you can expect with
various shielding configurations: bonded foil, four-end braid (67% on 59
series and 60% on 6 series cables) is good for 90 dB to 95 dB. Tri-shield is
good for around 100 dB to 105 dB, and quad-shield will get you to the 105 dB
to 115 dB range. The actual shielding effectiveness will vary from
installation to installation, how the cable is handled during installation,
age of the cable, how well it was waterproofed, etc. Wind flexure will
degrade these numbers, as will improperly installed connectors.
As you might imagine, craftsmanship plays a big role in drop reliability. I
recall a system many years ago that used quad-shield for all of its drops,
but leakage (and ingress) was horrible. The problem? Poor workmanship. The
best materials do no good if connectors aren't installed correctly or are
left loose.
Regards,
Ron Hranac
Cisco Systems
------------------------------
From: "Steve Allen" <stallen@cisco.com>
To: <scte-list@relay.doit.wisc.edu>
Subject: RE: .412 Cables
Date: Sun, 15 Oct 2000 19:35:38 -0700
Message-ID: <000101c03719$c504a560$2a04150a@stallen-pc.cisco.com>
MIME-Version: 1.0
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
Dear Rex, with everything I have had going on the last 30 years, I never
stopped to think about why .412 was made that way. I guess I figured it had
to do with the "inside diameter of the outer conductor, and the outside
diameter of the center conductor, and the type of dialectric material, and
the ambient temperature, that determined the characteristic impotence of the
cable", to make it 75 ohms and able to be put up with an C-Lasher when they
took down the RG-11 and the Chromataps, etc , etc.
But now I am dying to know why .412 is .412
Steve Allen
PS, Archer said Yes. <grin>
-----Original Message-----
From: owner-scte-list@relay.doit.wisc.edu
[mailto:owner-scte-list@relay.doit.wisc.edu]On Behalf Of tvrex
Sent: Sunday, October 15, 2000 12:42 PM
To: scte-list@relay.doit.wisc.edu
Subject: .412 Cables
Just reading the messages on today's List concerning RG-11 and .412 cables.
With all of the .412 cables still out there today and the number of people
who have installed it and replaced it, I wonder that no one asks why all of
the rest of our feeder and trunk cables are manufactured in quarter-inch
sizes:
.250; .500; .625; .750; .875 and 1.000. Why was .412 cable even
manufactured? The answer is one you probably wouldn't have ever
guessed....Rex
------------------------------
Message-ID: <8457258D741DD411BD3D0050DA62365907A3A7@huina.oceanic.com>
From: Doug Stanfield <DOUGS@oceanic.com>
To: "'scte-list@relay.doit.wisc.edu'" <scte-list@relay.doit.wisc.edu>
Subject: RE: .412 Cables
Date: Sun, 15 Oct 2000 17:21:07 -1000
MIME-Version: 1.0
Content-Type: text/plain;
charset="iso-8859-1"
[ Rex asked:]
> .250; .500; .625; .750; .875 and 1.000. Why was .412 cable even
> manufactured? The answer is one you probably wouldn't have ever
> guessed....
[ Steve Allen said:]
> Dear Rex, with everything I have had going on the last 30
> years, I never stopped to think about why .412 was made that way.
---snip---
> But now I am dying to know why .412 is .412
My best guess is that it has something to do with Metric Madness, but I'd
really like to know also.
-Doug-
------------------------------
Message-Id: <4.2.0.58.20001015213247.00a97600@malone.cisco.com>
Date: Sun, 15 Oct 2000 21:39:38 -0700
To: <scte-list@relay.doit.wisc.edu>
From: Mark Millet <mmillet@cisco.com>
Subject: Re: Return path "cable modem" amplifier (DOCSIS)
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
>Mark,
>
>You missed the point. The 20 dB of gain of the block allowed the modem to
>use only enough output level to get recognized. Meaning in this case, the
>modem output went down from maximum, so that the output from the hybrid in
>the amplifier was somewhere around +62 to +65 dBmV in the locations where it
>was used. Think man, ALL the answers that have been posted in the recent
>threads talked about self adjusting modem output levels. This one adjusted
>itself also. It just had a slightly higher range, but it did not use it
>all. And by putting the amp directly on the modem, it did not amplify any
>other noise or ingress from anywhere in the house.
Thus, the solution was bringing the modem up from +58 dBmV maximum to +65 dBmV maximum via a large external amplifier device. Not exactly the optimal design choice, but there are reasons for each case, and sometimes the flat loss to the return amp just can't be bypassed.
I have seen some customers change the CMTS marshalling algorithm to permit such modems to remain online without this kind of amplifier. It works, but can result in lower C/N for those modems, and some lost packets, not to mention a higher probability of packet error in general.
hfc(config-if)#cable u0 power-adjust continue ?
<2-15> Power level in dB
where the "standard" compliant setting is 2 dB, and 15 dB will cause a modem to stay online, even if almost all data packets would be lost. 8-10 dB "kind of works" for QPSK, and 5-6 dB "kind of works" for 16QAM, but it is not the right solution. The external amplifier is the work-around for when the plant / in home design can't be fixed.
Mark Millet
CSE Service Provider Engineering
Direct: 408 526 6275
Cisco Systems Inc. Fax: 650 526 5807
170 West Tasman Dr Pager: 800 365 4578
San Jose, CA 95134 mmillet@cisco.com
------------------------------
End of SCTE-LIST Digest 2110
****************************
I've gone to my daily |
