[mythtv-users] us-cable users who specified finetune values, please help...

Brian Wood beww at beww.org
Mon Jul 24 20:45:42 UTC 2006



On Jul 24, 2006, at 1:45 PM, Michael T. Dean wrote:

> Daniel is currently working on fixing the frequency tables defined by
> MythTV to make it easier for users to configure their channels by
> ensuring the frequencies are correct "out of the box."  In doing this,
> we need additional information to determine whether a theoretically
> possible situation is forcing some users to specify finetune values to
> receive some of their channels.  Note that this only applies to users
> who are using the us-cable frequency table (and not us-bcast or
> us-cable-hrc or us-cable-irc) for any of their video sources.

<snip>

I've seen many posts from folks who have apparently had to input fine- 
tuning values, and I have never understood why.

Perhaps you need to take a step back and try and find out "why".

Over-the-air broadcasters obviously control their carrier frequencies  
extremely well, so errors in their transmissions can be eliminated.

Cable systems use either a modulator or a "hetrodyne processor" to  
create the individual channel signals emanating from their headends.

In the case of RF modulators the frequency is initially generated by  
an RF oscillator operating at an "IF" frequency of 45.75 Mhz. This  
signal is then converted to the final output frequency by mixing it  
with the output of a local oscillator, resulting in the final channel  
frequency.

So there are two possible causes of frequency error, the initial  
45.75 modulator and the local oscillator used to hetrodyne the signal  
to its final value.

Both of these oscillators are quite stable. Although temperature- 
controlled ovens are not used, the stability and accuracy of the  
crystal oscillators is usually much better than 1 Khz., for a worst  
case error of +- 2Khz. This is nowhere near enough to cause a visible  
problem in the received video signal.

Some off-air  channels are processed with "hetrodyne processors", in  
which the incoming signal is first mixed with a local oscillator and  
converted to an IF of 45.75 Mhz. So in the case of off-channel  
conversion the same potential error as for a modulator exists.

When converting "on-channel", in other words when using a hetrodyne  
processor to process a channel back onto its original frequency, the  
same oscillator is used for both conversions, so the result will be  
precisely the same as the original incoming frequency, any error will  
be cancelled out.

So, there really shouldn't be any channels on a modern cable system  
that are much more than 1 Khz. off from their "correct" value, with  
one exception:

The FCC sometimes assigns carrier frequencies with a plus or minus  
10Khz. offset. This is done to minimize co-channel interference but  
we need not go into those reasons.

The result is that if an off-air channel is being processed "on- 
channel" by a hetrodyne processor it may wind up 10 Khz. off from its  
"correct" value. This is still not really enough to cause reception  
problems, but perhaps in unique situations it might require the  
merest tweak to fine-tuning.

So if there is really no excuse for any off-frequency signals on a  
cable system why are folks having to fine-tune? Is it perhaps errors  
in the tuners? This really shouldn't be, but I guess it is more  
likely than errors in the original transmission.

The simplest possible sort of AFC (automatic frequency control)  
circuit should be able to deal with much wider errors than should be  
encountered with any cable TV system. I've never had to tweak any  
fine-tuning on my 8 different capture devices.

Before you look at possible changes to frequency tables I would want  
to know why there is a problem in the first place, there really  
shouldn't be.

(I have greatly simplified things in the above discussion, I've left  
out the fact of an aural carrier 4.5 Mhz. above the visual, I've not  
mentioned frequency control systems for HRC and IRC, and a few other  
matters, but none of them should change the basic facts of the problem).


More information about the mythtv-users mailing list