[mythtv-users] Interference lines in all recordings

[Brian Wood]

On Feb 13, 2007, at 11:03 AM, Tim Gray wrote:

> I found a solution to my problem.
>
> After lots of tests, adding in aluminum shields, soldering the shield
> cans, and not getting rid of the interference I discovered the problem
> that may help others.  Regular TV's and the ReplayTV have no  
> sensitivity
> to this so it has to be a PVR-350 specific problem...
>
> My signal coming in the house is too high. Specifically on the lower
> channel numbers.  I had a 18+DB difference between channel 2 and  
> channel
> 96  adding in a tilt compensator where the cable comes in the house
> eliminated the problem by reducing the signal of the lower channel
> numbers and not affecting the higher channel numbers signal strength.
> Granted I have tools at my disposal that most of you do not such as a
> NTSC/QAM signal strength meter/ cable plant analyzer but adding in a
> $8.95 tilt compensator to your cable may solve some interference
> issues.  It certianly did for me.  I am actually planning on adding a
> overall attenuator to my cable line as well to reduce the signal  
> another
> 3 DB across all channels as it is still higher than reccomended.
>

So what's you're saying is that your cable company was giving you  
"out-of-spec" signals.

The FCC-specified specs state that there shall be no more than a  
12db. difference between the highest-level channel and the lowest, so  
you were out of the "tilt" spec by 6db. This might sound minor, but  
to put it another way the highest-level channel had four times the  
power it was "supposed" to have relative to the lowest-level channel.

"Tilt" is caused by the fact that coax cable has more attenuation at  
high frequencies than at low ones. A piece of coax that has 6db. of  
loss at channel 2 (54 Mhz.)  will have 12db. of loss at channel 13  
(216 Mhz.). This is usually compensated by equalizers and  
compensating amplifiers.

So if the lowest-level channel was at the FCC minimum of 0DbMV then  
the highest would have been at +18DbMV., high enough to cause serious  
beats in the picture due to overload in the front-end of the TV.

In case you're interested, the rest of the specs for signal level are:
--------------
A minimum of 0.0DbMV on any one channel

No more than a 12DbMV difference between any two channels on the system

No more than a 3DbMV difference between any two adjacent channels

No more that 5% "Hum" or low-frequency AM noise on the system

No more than a 12 DbMV variation in any 24-hour period.

Composite Triple Beat or "Cross-Mod" components at least 48Db. below  
signal level.
---------------

As usual it's more complicated than it seems. Technically the FCC  
speca apply only to "Broadcast Signals", so non-broadcast channels  
like CNN and HBO do not count, also, any gain or AGC action from a  
STB is allowed to be taken into consideration as the level is  
specified as "at the TV terminals". All this is a gross over- 
simplification but you get the point.

As always the regulations are somewhat out of date with reality, as  
is probably my own information since it's been a while since I had to  
worry about the regs.

As you point out, most folks do not have the ability t measure these  
values in any case. I wonder how many of the "strange" problems  
reported here are actually caused by similar problems.