75 ohm Termination Question

[Bennett]

A friend replaced her Comcast TV box with an over-the-air antenna but
continues to use Comcast Internet service. This leaves her
Comcast-supplied splitter with an unterminated socket - does it matter?
Should I bother to terminate it?

(A barrel connector isn't a good replacement for the splitter as the
splitter provides a grounding point.)

---
This email is free from viruses and malware because avast! Antivirus protection is active.
http://www.avast.com

 

[Geoffrey S. Mendelson]

David Platt wrote:

The cable companies usually have "signal leak" detection teams driving
around occasionally, looking for emission from their cable
transmissions which ought not to be present. I've heard of people
who've had their cable company employees arrive at their door, and
insist that their in-home wiring either be fixed or disconnected.

If the FCC catches them (usually because of a complaint from the user
of those frequencies) they can be heavily fined, so usually they are
agressive at tracking down "leakage".

Geoff.


--
Geoffrey S. Mendelson, N3OWJ/4X1GM/KBUH7245/KBUW5379

 

[hrhofmann@sbcglobal.net]

On Thursday, February 13, 2014 12:40:26 PM UTC-6, Bennett wrote:
> A friend replaced her Comcast TV box with an over-the-air antenna but continues to use Comcast Internet service. This leaves her Comcast-supplied splitter with an unterminated socket - does it matter? Should I bother to terminate it? (A barrel connector isn't a good replacement for the splitter as the splitter provides a grounding point.) --- This email is free from viruses and malware because avast! Antivirus protection is active. http://www.avast.com

COMCAST might even furnish your friend with a splitter.

 

[hrhofmann@sbcglobal.net]

Oops, I meant furnish your friend a terminator

 

[David Platt]

A friend replaced her Comcast TV box with an over-the-air antenna but
continues to use Comcast Internet service. This leaves her
Comcast-supplied splitter with an unterminated socket - does it matter?
Should I bother to terminate it?

(A barrel connector isn't a good replacement for the splitter as the
splitter provides a grounding point.)

It is a good idea to do so.

Cable-TV systems often use frequencies on the cable that are assigned
to non-television services for over-the-air use (e.g. aircraft, ham
band, etc.). If you have loose or unterminated cable connections,
some of the cable-TV signal will "leak" out and can be radiated. This
can cause interference in those other services.

The cable companies usually have "signal leak" detection teams driving
around occasionally, looking for emission from their cable
transmissions which ought not to be present. I've heard of people
who've had their cable company employees arrive at their door, and
insist that their in-home wiring either be fixed or disconnected.

Also, if you don't terminate a splitter port, you'll create an
impedance mismatch at that point, and this can cause a signal
"reflection" within the cable. Depending on the quality of the
splitter (its port-to-port isolation) this reflection can cause
"ghosts" in an analog TV transmission, multipath distortion in analog
FM, and inter-symbol interference (leading to a higher bit-error rate
and possible dropouts) in digital TV.

75-ohm F-connector terminators are cheap and widely available. It's
good practice to use them on all otherwise-unused splitter ports.

 

[Jeff Liebermann]

On Thu, 13 Feb 2014 10:40:26 -0800, Bennett <bjprice@cal.berkeley.edu>
wrote:

A friend replaced her Comcast TV box with an over-the-air antenna but
continues to use Comcast Internet service. This leaves her
Comcast-supplied splitter with an unterminated socket - does it matter?

Yes, it matters.

>Should I bother to terminate it?

Yes, you should terminate it.

(A barrel connector isn't a good replacement for the splitter as the
splitter provides a grounding point.)

Sorta. Apparently she has a 2 way splitter installed with one port
left unterminated. A barrel connector is a very good replacement.
However, you're correct that you'll need to do something about the
grounding point. If this is indoors, just remove the grounding wire,
wrap it around the barrel connector, and hold it in place with a small
hose clamp. For outdoors, use a proper grounding clamp:
<http://www.solidsignal.com/pview.asp?p=pvgb1&d=perfect-vision-pvgb1-single-grounding-blocks-%28pvgb1%29>
<https://www.google.com/search?tbm=isch&q=cable+tv+grounding+block>
For outdoors, also don't forget to recreate the coax drip loop so that
water doesn't run into the connector block, or into the house.
<http://activerain.com/blogsview/3168531/drip-loop-really-do-you-need-it-wenatchee-and-leavenworth-home-inspection>
<http://images.meredith.com/diy/images/2009/03/p_SCW_240_02.jpg>
--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

 

[Michael Black]

On Thu, 13 Feb 2014, Jeff Liebermann wrote:

On Thu, 13 Feb 2014 17:41:28 -0800, Bennett <bjprice@cal.berkeley.edu
wrote:

On 2/13/2014 5:17 PM, Jeff Liebermann wrote:
On Thu, 13 Feb 2014 10:40:26 -0800, Bennett <bjprice@cal.berkeley.edu
wrote:

A friend replaced her Comcast TV box with an over-the-air antenna but
continues to use Comcast Internet service. This leaves her
Comcast-supplied splitter with an unterminated socket - does it matter?

Yes, it matters.

Should I bother to terminate it?

Yes, you should terminate it.

(A barrel connector isn't a good replacement for the splitter as the
splitter provides a grounding point.)

Sorta. Apparently she has a 2 way splitter installed with one port
left unterminated. A barrel connector is a very good replacement.
However, you're correct that you'll need to do something about the
grounding point. If this is indoors, just remove the grounding wire,
wrap it around the barrel connector, and hold it in place with a small
hose clamp. For outdoors, use a proper grounding clamp:
http://www.solidsignal.com/pview.asp?p=pvgb1&d=perfect-vision-pvgb1-single-grounding-blocks-%28pvgb1%29
https://www.google.com/search?tbm=isch&q=cable+tv+grounding+block
For outdoors, also don't forget to recreate the coax drip loop so that
water doesn't run into the connector block, or into the house.
http://activerain.com/blogsview/3168531/drip-loop-really-do-you-need-it-wenatchee-and-leavenworth-home-inspection
http://images.meredith.com/diy/images/2009/03/p_SCW_240_02.jpg

Thanks for all your replies. It's an indoor splitter. Since I had a 68
ohm resistor and appropriate connector (but no hose clamp), I just built
a terminator. Factory-builts are cheap - if you want 10, 25 or 100 but
I only needed one.

Not good enough. If you just crammed a 68 ohm resistor into the
F-connector jack, you probably have a fairly good approximation of an
antenna. It's going to radiate junk, and pickup more junk (ingres).
If you insist on doing it like that, wrap the resistor with some
insulating tape, and then cover it with aluminum foil.

One time I was looking into switches that couldn't be accidentally
toggled, and I thought of type F connectors.

I'm sure I took a bit of wire for the center conductor, then a bit of thin
copper sheet. PUt the wire into the centre of the copper sheet, soldered
it, then soldered the copper sheet to the outer part of the connector.
So switching was by screwing in the connector or unscrewing it.

I'm blank about soldering the copper sheet to the outer part of the
conenctor though, but I can picture the finished connector.

Yeah, you're right that nobody sells them in small quantities. The
best I can do on eBay is 10 pcs for $3.85 including shipping.
http://www.ebay.com/itm/261230167868

That's why I tend to keep junk. Terminators may seem like the most
mundane thing found in a pile of junk on the sidewalk, but if you need
one, you actually save money and effort, while the 60gig iPod that I found
somewhere else needs a battery in order to be useful.

Michael

 

[Jeff Liebermann]

On Thu, 13 Feb 2014 22:32:15 -0500, Michael Black <et472@ncf.ca>
wrote:

One time I was looking into switches that couldn't be accidentally
toggled, and I thought of type F connectors.

No need to build your own. Most switch manufacturers make "locking
lever" type of switches. I've used them in various devices where I
didn't want any accidentally toggling. They're also common in
aerospace and military equipment:
<http://infinityaerospace.com/LL_Toggle_Sw_for_Preoiler.jpg>
<http://www.chiefaircraft.com/edm-7103k2zqe.html>
<http://www.ebay.com/bhp/locking-toggle-switch>

--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

 

[Jeff Liebermann]

On Thu, 13 Feb 2014 15:25:34 -0800 (PST), "hrhofmann@sbcglobal.net"
<hrhofmann@sbcglobal.net> wrote:

>Oops, I meant furnish your friend a terminator

Careful. Comcast has an early termination fee.

--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

 

[Bennett]

On 2/13/2014 5:17 PM, Jeff Liebermann wrote:

On Thu, 13 Feb 2014 10:40:26 -0800, Bennett <bjprice@cal.berkeley.edu
wrote:

A friend replaced her Comcast TV box with an over-the-air antenna but
continues to use Comcast Internet service. This leaves her
Comcast-supplied splitter with an unterminated socket - does it matter?

Yes, it matters.

Should I bother to terminate it?

Yes, you should terminate it.

(A barrel connector isn't a good replacement for the splitter as the
splitter provides a grounding point.)

Sorta. Apparently she has a 2 way splitter installed with one port
left unterminated. A barrel connector is a very good replacement.
However, you're correct that you'll need to do something about the
grounding point. If this is indoors, just remove the grounding wire,
wrap it around the barrel connector, and hold it in place with a small
hose clamp. For outdoors, use a proper grounding clamp:
http://www.solidsignal.com/pview.asp?p=pvgb1&d=perfect-vision-pvgb1-single-grounding-blocks-%28pvgb1%29
https://www.google.com/search?tbm=isch&q=cable+tv+grounding+block
For outdoors, also don't forget to recreate the coax drip loop so that
water doesn't run into the connector block, or into the house.
http://activerain.com/blogsview/3168531/drip-loop-really-do-you-need-it-wenatchee-and-leavenworth-home-inspection
http://images.meredith.com/diy/images/2009/03/p_SCW_240_02.jpg

Thanks for all your replies. It's an indoor splitter. Since I had a 68
ohm resistor and appropriate connector (but no hose clamp), I just built
a terminator. Factory-builts are cheap - if you want 10, 25 or 100 but
I only needed one.

 

[Jeff Liebermann]

On Thu, 13 Feb 2014 17:41:28 -0800, Bennett <bjprice@cal.berkeley.edu>
wrote:

On 2/13/2014 5:17 PM, Jeff Liebermann wrote:
On Thu, 13 Feb 2014 10:40:26 -0800, Bennett <bjprice@cal.berkeley.edu
wrote:

A friend replaced her Comcast TV box with an over-the-air antenna but
continues to use Comcast Internet service. This leaves her
Comcast-supplied splitter with an unterminated socket - does it matter?

Yes, it matters.

Should I bother to terminate it?

Yes, you should terminate it.

(A barrel connector isn't a good replacement for the splitter as the
splitter provides a grounding point.)

Sorta. Apparently she has a 2 way splitter installed with one port
left unterminated. A barrel connector is a very good replacement.
However, you're correct that you'll need to do something about the
grounding point. If this is indoors, just remove the grounding wire,
wrap it around the barrel connector, and hold it in place with a small
hose clamp. For outdoors, use a proper grounding clamp:
http://www.solidsignal.com/pview.asp?p=pvgb1&d=perfect-vision-pvgb1-single-grounding-blocks-%28pvgb1%29
https://www.google.com/search?tbm=isch&q=cable+tv+grounding+block
For outdoors, also don't forget to recreate the coax drip loop so that
water doesn't run into the connector block, or into the house.
http://activerain.com/blogsview/3168531/drip-loop-really-do-you-need-it-wenatchee-and-leavenworth-home-inspection
http://images.meredith.com/diy/images/2009/03/p_SCW_240_02.jpg

Thanks for all your replies. It's an indoor splitter. Since I had a 68
ohm resistor and appropriate connector (but no hose clamp), I just built
a terminator. Factory-builts are cheap - if you want 10, 25 or 100 but
I only needed one.

Not good enough. If you just crammed a 68 ohm resistor into the
F-connector jack, you probably have a fairly good approximation of an
antenna. It's going to radiate junk, and pickup more junk (ingres).
If you insist on doing it like that, wrap the resistor with some
insulating tape, and then cover it with aluminum foil.

Yeah, you're right that nobody sells them in small quantities. The
best I can do on eBay is 10 pcs for $3.85 including shipping.
<http://www.ebay.com/itm/261230167868>

Looks like you're semi-local. Send me a mailing address and I'll
stuff a few into an envelope for you. You could also buy one of the
outdoor grounding blocks, and use it indoors. Lots of options that
doesn't involve a leaky do-it-thyself terminator.

--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

 

[dave]

On 02/13/2014 06:07 PM, Jeff Liebermann wrote:

On Thu, 13 Feb 2014 17:41:28 -0800, Bennett <bjprice@cal.berkeley.edu


Not good enough. If you just crammed a 68 ohm resistor into the
F-connector jack, you probably have a fairly good approximation of an
antenna. It's going to radiate junk, and pickup more junk (ingres).
If you insist on doing it like that, wrap the resistor with some
insulating tape, and then cover it with aluminum foil.

Do you have any idea what the wavelength is at 600MHz? It's half a meter
(Meters times megaHertz equals 300). That makes the length of a quarter
wave is 1/8 meter or about 5". A tiny resistor is not going to radiate
efficiently enough. If the "splitter" is a hybrid the unterminated port
will cause higher attenuation through the splitter. Maybe a single dB.
Not important when receiving.

You'd be better off getting a DC block. The splitter loses at least 3 dB.

 

[Michael A. Terrell]

dave wrote:

On 02/13/2014 06:07 PM, Jeff Liebermann wrote:
On Thu, 13 Feb 2014 17:41:28 -0800, Bennett <bjprice@cal.berkeley.edu


Not good enough. If you just crammed a 68 ohm resistor into the
F-connector jack, you probably have a fairly good approximation of an
antenna. It's going to radiate junk, and pickup more junk (ingres).
If you insist on doing it like that, wrap the resistor with some
insulating tape, and then cover it with aluminum foil.


Do you have any idea what the wavelength is at 600MHz? It's half a meter
(Meters times megaHertz equals 300). That makes the length of a quarter
wave is 1/8 meter or about 5". A tiny resistor is not going to radiate
efficiently enough. If the "splitter" is a hybrid the unterminated port
will cause higher attenuation through the splitter. Maybe a single dB.
Not important when receiving.

You'd be better off getting a DC block. The splitter loses at least 3 dB.

A DC Block is nothing more than a capacitor between two connectors.
Some use more than one capacitor to extend the usable frequencies. I've
used up to five to have a flat response from <100 KHz to >1400 MHz at 50
Ohms for a production test line. The insertion loss was about .1 dB at
each end of the usable range.


--
Anyone wanting to run for any political office in the US should have to
have a DD214, and a honorable discharge.

 

[Jeff Liebermann]

On Fri, 14 Feb 2014 06:38:54 -0800, dave <ricketzz@earthlink.net>
wrote:

On 02/13/2014 06:07 PM, Jeff Liebermann wrote:
On Thu, 13 Feb 2014 17:41:28 -0800, Bennett <bjprice@cal.berkeley.edu


Not good enough. If you just crammed a 68 ohm resistor into the
F-connector jack, you probably have a fairly good approximation of an
antenna. It's going to radiate junk, and pickup more junk (ingres).
If you insist on doing it like that, wrap the resistor with some
insulating tape, and then cover it with aluminum foil.

Do you have any idea what the wavelength is at 600MHz? It's half a meter
(Meters times megaHertz equals 300). That makes the length of a quarter
wave is 1/8 meter or about 5". A tiny resistor is not going to radiate
efficiently enough.

Whether it radiates efficiently or inefficiently, it still radiates.
Assuming a 1/2 watt resistor, with one lead wrapped around the
grounding screw, and the other shoved into the F-connector center
wire, there will be about 1/4" of exposed wire between the center pin
and the 60 ohm resistor. That's not much, can probably be tolerated,
and will be sufficient to slightly increase base line noise level in
the area. I would hate to think what might happen if EVERYONE decided
to use leaky terminators.[1]

I like to play ham radio (VHF/UHF) in my vehicle when driving. I
constantly hear the digital hash as I drive repeatedly by specific
locations. At first, I assumed that these were leaks from the
overhead cable plant. Nope. When I started banging on doors, I found
some really creative wiring and connector crimping. Most were caused
by distribution amps with unterminated cables, where the coax cable
becomes somewhat of an antenna. I also found either a defective drop
or tap on the pole (Comcast didn't tell me what they found, but they
did fix it).

Obviously, a leaky cable is a much longer length of wire than 1/4" of
exposed center conductor. Yet, after the gross leaks were fixed, I
still could still hear the test tones on the borrowed Trilithic
something leakage detector. I had to install terminators on literally
everything to reduce it to zero. Admittedly, I was working well below
the FCC Part 76.605 leakage limits:
<http://transition.fcc.gov/mb/engineering/605.html>
<http://www.cablefax.com/tech/sections/columns/broadband/Broadband-Signal-Leakage-in-an-All-Digital-Network_33882.html>
<http://www.cablefax.com/tech/deployment/bestpractices/A-Leak-Is-A-Leak&hellip%20Or-Is-It_52272.html>
but I needed to get the radiated junk down to a level where I could
use my radios in the neighborhood. Overkill yes, but necessary.

It's also a problem in the other direction, which is called ingress.
Nearby transmitters, such as mine, can easily get into the cable
system, and interfere with reception. Once it gets in, there's no
easy way to get rid of the signal. The best approach is proper
shielding. It isn't just my transmitters, but broadcast stations,
intermod with cordless phones, RF remotes, wireless security cameras,
and other RF devices. Mostly, they cause problems with cable modems,
which use the higher RF frequencies, but in extreme cases, can affect
TV reception. It's also difficult to see with a spectrum analyzer:
<http://www.jdsu.com/ProductLiterature/DSAM-D3-Spectrum-Analyzer-App-Note.pdf>
Can you see it? Spoiler:
<http://blogs.jdsu.com/hfc/Lists/Photos/dd6cc9ceb7e6a11e1006e1ad654edd7f.jpg>

More:
"Characterizing Signal Leakage from an All-Digital Cable Network"
<http://www.incospec.com/resources/webinars/files/Characterizing%20Signal%20Leakage%20from%20an%20All-Digital%20Cable%20NetworkMS.pdf>

If the "splitter" is a hybrid the unterminated port
will cause higher attenuation through the splitter. Maybe a single dB.
Not important when receiving.

Without a termination, the splitter begins to resonate with the coax
cable which can be made to radiate. I'm not sure of the unterminated
losses, but I suspect it's more than 1dB. I'll try it on the bench
when I have time (mostly to clean off the accumulated junk from the
bench).
<http://802.11junk.com/jeffl/pics/home/slides/test-equip-mess.html>

>You'd be better off getting a DC block. The splitter loses at least 3 dB.

A DC block won't do anything useful except block DC. I think you
might have meant a straight through barrel connector.



[1] Incidentally, you might be amused to look at my study of monopole
antenna gains versus length.
<http://802.11junk.com/jeffl/antennas/Monopole/index.html>
The title includes the wavelength. For example, 0_750 is 3/4
wavelength long and 5_250 is 5.25 wavelengths long. If you look at
the antenna gains, you'll notice that the gain doesn't really drop
very much for shortened antennas. From the antenna patterns:
Wavelength Gain dBi
0.250 5.19
0.125 4.85
0.050 4.75
That's a 0.45dB gain loss in an antenna that's 1/5th the length of a
1/4 wave radiator. That might explain why I'm concerned about even
small amounts of exposed center conductor.


--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

 

[David Platt]

In article <f7fsf99rlb7drdsn9h8b6v9b5gv1n74lg1@4ax.com>,
Jeff Liebermann <jeffl@cruzio.com> wrote:

It's also a problem in the other direction, which is called ingress.
Nearby transmitters, such as mine, can easily get into the cable
system, and interfere with reception.

Heh. I remember a discussion hereabouts, some years ago... it might
even have been you posting, Jeff?... in which the poster reported
having detected cable-leakage problems in his town, reported them to
the cable company, and was being ignored. He found that even the
cables system's central facility (where the downlinks and primary
head-end amplifiers were) was leaking like a sieve.

He visited the office, asked to speak to the lead engineer, and
explained what he'd found. The engineer poo-pooh'ed and said "No, our
system is tight, we're very careful about that." The OP took out his
2-meter radio, keyed down... and the signal monitors in the next room
went crazy, as the 2-meter signal leaked into the system and saturated
the first-stage amplifiers. Picture quality went to hell all over
town and the phones started ringing...

 

[Jeff Liebermann]

On Fri, 14 Feb 2014 11:16:43 -0800, dplatt@coop.radagast.org (David
Platt) wrote:

In article <f7fsf99rlb7drdsn9h8b6v9b5gv1n74lg1@4ax.com>,
Jeff Liebermann <jeffl@cruzio.com> wrote:

It's also a problem in the other direction, which is called ingress.
Nearby transmitters, such as mine, can easily get into the cable
system, and interfere with reception.

Heh. I remember a discussion hereabouts, some years ago... it might
even have been you posting, Jeff?... in which the poster reported
having detected cable-leakage problems in his town, reported them to
the cable company, and was being ignored. He found that even the
cables system's central facility (where the downlinks and primary
head-end amplifiers were) was leaking like a sieve.

Not me. I've reported specific leaks, with photos of my test
equipment and attached reports to Comcast. However, that's when I had
a friend employed by Comcast to expedite the process. Even so, it
would usually take a month for any kind of response. I no longer have
an inside track into Comcast, so I don't know what will happen if I
tried that today.

He visited the office, asked to speak to the lead engineer, and
explained what he'd found. The engineer poo-pooh'ed and said "No, our
system is tight, we're very careful about that." The OP took out his
2-meter radio, keyed down... and the signal monitors in the next room
went crazy, as the 2-meter signal leaked into the system and saturated
the first-stage amplifiers. Picture quality went to hell all over
town and the phones started ringing...

That definitely wasn't me. I would never do that. The most sensitive
point of the cable system is at the head end. That's where all the
gain is buried, where all the numerous RF sources come together, and
where there's a high probability of nearby RF sources creating
problems. I also doubt the story. I'm quite familiar with the
Comcast head-end in Scotts Valley. Nobody gets anywhere near the
equipment unless they have a very good reason for being there. A
leakage complaint doesn't qualify. Also, there's no "lead engineer".
All the Comcast engineers are either in Denver, or outside
contractors.

If you sit in the parking lot at the Scotts Valley office, you'll hear
some RFI coming from the building on VHF and UHF. I've walked around
2 sides of the building with a portable spectrum analyzer and didn't
see anything really disgusting. However, I will admit that everything
above 700 MHz was problematic due to the proximity of a nearby cell
site. That was also many years ago, so things may have changed (or
deteriorated).

So far, what I've seen of the local Comcast plant, it's quite well
done. There's some junk cable Comcast inherited years ago north of
Boulder Creek that should have been ripped out and replaced, but
otherwise, it's quite well done and shielded. Unfortunately, Comcast
is using more and more outside contractors, with minimal oversight,
resulting in some rather lousy installations.
<http://802.11junk.com/jeffl/pics/Comcast/slides/Comcast-cabling.html>
I have more photos that are worse, but I think you can see the
problem.

Also, there are problems with the roadside amplifier and pull boxes.
These are targets for vandals, copper thieves, and sloppy installers.
I've found them broken open and reported them to Comcast. I usually
get an instant reaction, not because of the potential loss of customer
service, but because there's high voltage on the big cable (to run the
amplifiers) which could easily kill some idiot trying to steal the
copper.

Incidentally, if you run into the test frequency that Comcast is using
for leakage testing, I'm interested. It changed a few months ago when
they went all digital and I haven't been able to pry it out of anyone.


--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

 

[dave]

On 02/14/2014 09:03 AM, Jeff Liebermann wrote:

On Fri, 14 Feb 2014 06:38:54 -0800, dave <ricketzz@earthlink.net
wrote:

On 02/13/2014 06:07 PM, Jeff Liebermann wrote:
On Thu, 13 Feb 2014 17:41:28 -0800, Bennett <bjprice@cal.berkeley.edu


Not good enough. If you just crammed a 68 ohm resistor into the
F-connector jack, you probably have a fairly good approximation of an
antenna. It's going to radiate junk, and pickup more junk (ingres).
If you insist on doing it like that, wrap the resistor with some
insulating tape, and then cover it with aluminum foil.

Do you have any idea what the wavelength is at 600MHz? It's half a meter
(Meters times megaHertz equals 300). That makes the length of a quarter
wave is 1/8 meter or about 5". A tiny resistor is not going to radiate
efficiently enough.

Whether it radiates efficiently or inefficiently, it still radiates.
Assuming a 1/2 watt resistor, with one lead wrapped around the
grounding screw, and the other shoved into the F-connector center
wire, there will be about 1/4" of exposed wire between the center pin
and the 60 ohm resistor. That's not much, can probably be tolerated,
and will be sufficient to slightly increase base line noise level in
the area. I would hate to think what might happen if EVERYONE decided
to use leaky terminators.[1]

I like to play ham radio (VHF/UHF) in my vehicle when driving. I
constantly hear the digital hash as I drive repeatedly by specific
locations. At first, I assumed that these were leaks from the
overhead cable plant. Nope. When I started banging on doors, I found
some really creative wiring and connector crimping. Most were caused
by distribution amps with unterminated cables, where the coax cable
becomes somewhat of an antenna. I also found either a defective drop
or tap on the pole (Comcast didn't tell me what they found, but they
did fix it).

Obviously, a leaky cable is a much longer length of wire than 1/4" of
exposed center conductor. Yet, after the gross leaks were fixed, I
still could still hear the test tones on the borrowed Trilithic
something leakage detector. I had to install terminators on literally
everything to reduce it to zero. Admittedly, I was working well below
the FCC Part 76.605 leakage limits:
http://transition.fcc.gov/mb/engineering/605.html
http://www.cablefax.com/tech/sections/columns/broadband/Broadband-Signal-Leakage-in-an-All-Digital-Network_33882.html
http://www.cablefax.com/tech/deployment/bestpractices/A-Leak-Is-A-Leak&hellip%20Or-Is-It_52272.html
but I needed to get the radiated junk down to a level where I could
use my radios in the neighborhood. Overkill yes, but necessary.

It's also a problem in the other direction, which is called ingress.
Nearby transmitters, such as mine, can easily get into the cable
system, and interfere with reception. Once it gets in, there's no
easy way to get rid of the signal. The best approach is proper
shielding. It isn't just my transmitters, but broadcast stations,
intermod with cordless phones, RF remotes, wireless security cameras,
and other RF devices. Mostly, they cause problems with cable modems,
which use the higher RF frequencies, but in extreme cases, can affect
TV reception. It's also difficult to see with a spectrum analyzer:
http://www.jdsu.com/ProductLiterature/DSAM-D3-Spectrum-Analyzer-App-Note.pdf
Can you see it? Spoiler:
http://blogs.jdsu.com/hfc/Lists/Photos/dd6cc9ceb7e6a11e1006e1ad654edd7f.jpg

More:
"Characterizing Signal Leakage from an All-Digital Cable Network"
http://www.incospec.com/resources/webinars/files/Characterizing%20Signal%20Leakage%20from%20an%20All-Digital%20Cable%20NetworkMS.pdf

If the "splitter" is a hybrid the unterminated port
will cause higher attenuation through the splitter. Maybe a single dB.
Not important when receiving.

Without a termination, the splitter begins to resonate with the coax
cable which can be made to radiate. I'm not sure of the unterminated
losses, but I suspect it's more than 1dB. I'll try it on the bench
when I have time (mostly to clean off the accumulated junk from the
bench).
http://802.11junk.com/jeffl/pics/home/slides/test-equip-mess.html

You'd be better off getting a DC block. The splitter loses at least 3 dB.

A DC block won't do anything useful except block DC. I think you
might have meant a straight through barrel connector.



[1] Incidentally, you might be amused to look at my study of monopole
antenna gains versus length.
http://802.11junk.com/jeffl/antennas/Monopole/index.html
The title includes the wavelength. For example, 0_750 is 3/4
wavelength long and 5_250 is 5.25 wavelengths long. If you look at
the antenna gains, you'll notice that the gain doesn't really drop
very much for shortened antennas. From the antenna patterns:
Wavelength Gain dBi
0.250 5.19
0.125 4.85
0.050 4.75
That's a 0.45dB gain loss in an antenna that's 1/5th the length of a
1/4 wave radiator. That might explain why I'm concerned about even
small amounts of exposed center conductor.

Here's one kind of splitter:

http://www.e-meca.com/tech_papers/hybrid_basic.php

And another:

http://www.minicircuits.com/pdfs/SEPS-4-272+.pdf

Only the hybrid type leaks a bunch. The port to port isolation of the
second type keeps unterminated ports from radiating.

 

[Jeff Liebermann]

On Sun, 16 Feb 2014 06:16:31 -0800, dave <ricketzz@earthlink.net>
wrote:

Here's one kind of splitter:
http://www.e-meca.com/tech_papers/hybrid_basic.php
And another:
http://www.minicircuits.com/pdfs/SEPS-4-272+.pdf

There are several types of 2-way splitters and combiners. Resistive,
0 degree, 90 degree, 180 degree, Wilkinson, DH-P, and whatever else I
forgot, with and without DC pass.

The typical commodity CATV 2-way splitter is a simple bifilar wound,
center tapped, ferrite transformer. There is no 100 ohm resistor
between output ports which is needed to improve isolation. They're
easy enough to open and look inside:
<http://www.cliftonlaboratories.com/userimages/6_db_h22.jpg>
<http://img198.imageshack.us/img198/6017/twoway.jpg>
<http://1.bp.blogspot.com/_8yDjtDOx9ws/TFt0DqJJP8I/AAAAAAAAFCw/z6nSWUh7aX4/s1600/3-way.JPG>
<http://www.waltzking.org/potd/20070124.jpg>
Not every complexicated or impressive.

More:
<http://www.cliftonlaboratories.com/6_db_hybrid_combiner.htm>
Scroll down to "RCA VH476 CATV Splitter" section:

Only the hybrid type leaks a bunch. The port to port isolation of the
second type keeps unterminated ports from radiating.

Nope. You misinterpreted the meaning of the port to port isolation
specification. What it means is that if you have all the ports
properly terminated, and you inject a signal into one of the 2
splitter ports, how much of that signal will appear on the other
splitter port.


--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558