[mythtv-users] Semi-OT: OTA, Comcast business HSI and the single coax

Allen Edwards allen.edwards at oldpaloalto.com
Fri Jul 25 17:43:03 UTC 2008


On Fri, Jul 25, 2008 at 9:39 AM, mw <meatwad2021 at gmail.com> wrote:
> Allen Edwards wrote:
>> Justin The Cynical wrote:
>>> Greetings.  This is semi-off topic IMO, but not fully as I'm sure
>>> someone on the list may have dealt with a similar situation.
>>>
>>> I am currently set-up for OTA digital only and my current connection to
>>> the world is via DSL.
>>>
>>> I am switching to Comcast business for my pipe, and would like to make
>>> use of a single coax for both.
>>>
>>> I know that OTA and Comcast television do not mix over the same coax,
>>> but what about just data?  Is there a splitter or filter that I can get
>>> to allow this to work, or am I going to have to run a new line just for
>>> the data?
>>>
>>> I am in Washington county, Oregon in the U.S. if that makes any difference.
>  >>
>> Cable modems use a normal TV channel for the data down and use
>> frequencies below channel 2 (around 30 MHz if I remember correctly) for
>> the return data.  That is one reason the return data is slower.
>>
>> Obviously you can run another coax to your computer to bring in the data
>> feed.  I have more than a dozen coax cables running around my house.  I
>> just run them under the eves or along the bottom of the siding and drill
>> a hole in the wall to get them in the house.
>>
>> Back to your question.  If I had to combine the signals I would first
>> put an amplifier on the antenna to keep the cable signals from going out
>> over the antenna to the rest of the world.
>
> Not a good idea. You would need an incredibly wide notch filter (or a
> combination of low and high-pass filters re-combined) from just below
> 52MHz upward to 550MHz. Shortly (2009), the cablecos will begin removing
> the analog allocations and replace them with digitally modulated content
> (still 6MHz in width) but will eagerly begin to use those channels for
> DOCSIS 3.0 which introduced channel bonding. Even prohibitively
> expensive frequency agile filters would not solve the problem.
>
>> If the cable company has
>> filtered out all the TV channels you have not paid for, you should be
>> able to just use a splitter to combine the signals at that point (after
>> the amplifier).  This assumes that the cable modem is at a frequency
>> above all the OTA stations, and I think that is a good assumption.  If
>> there are still TV signals on the Coax, you would need to filter them
>> out.  I don't know where you would get such a filter, except perhaps
>> through the cable company themselves.  You should check the assumptions
>> before trying it.
>
> The simple cableco filters used are enough to deteriorate the signal,
> not completely wipe it out. Once you overlay the standard OTA, the
> amplified OTA signals mixed into the existing cable spectrum will more
> than likely turn your inside wiring and the cableco's into unintentional
> radiators of unpredictable spurious RF energy at various harmonics in
> various strengths. Refer back to Brian Wood's example of unintentional
> radiation.
>
> Anything that mucks up 108-137 or many chunks of 225-400 MHz will
> endanger any private, commercial and military air operations in your
> vicinity and earn you an unpleasant visit from the authorities such as
> Brian's station did.
>
> IF you can run a CAT-5/6 from the cableco's Point Of Entry to the
> current location of the modem, simply install the modem at the POE and
> use a switch at the desired location. Problem solved.
>
> --
> mw
> _______________________________________________
> mythtv-users mailing list
> mythtv-users at mythtv.org
> http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-users
>

I think he has this solved with the addition of a coax line but to be
clear on what I was saying it was to filter out the cable co's main
bands leaving only the low and high bands.  If the cable co's don't do
it, he would need to.  However, if they are sufficiently attenuated,
it might still be OK as the other sources of isolation will take care
of it.

On the inside wiring being a radiator, this should not be a problem as
these were installed by the cable co and the levels of the non OTA
100-400 MHz signals are going to be less than they would have been had
he had cable TV.  Remember, the amplifier is only on the OTA signals.
Cable co's use (our should) RG-6U, which won't radiate.  The concern
is the radiation out the antenna.  For that, the signal has three
sources of isolation before it could get to the antenna.

If perfect any one of these would prevent the signal from reaching the
antenna 1) The filter on the cable signal.  2) The splitter isolation
between the inputs 3) the reverse isolation of the amplifier.  They
are not ideal, which is why I suggested that they all be there.

If you want more isolation, use two amplifiers with an attenuator
between them.  The OTA signal will be amplified, attenuated and
amplified.  The Cable signal will be attenuated by the second
amplifier, attenuated by the attenuator, and attenuated by the first
amplifier.  I really don't  think this would be necessary, however.

Also, the RFI example Brian gave was a clock signal, which is huge
compared to these low level cable signals.

You made a good point about coax being a radiator.  Cheap Radio Shack
coax might as well be an antenna.  Don't use it.  If you have cable
and use this kind of coax, one problem you will have it that the OTA
signals will get into the coax and mess up the cable channels.  I only
use RG-6U and professional grade connectors.

Anyway, I just wanted to make clear that I am not proposing something
that is going to cause RFI.

Allen


More information about the mythtv-users mailing list