On Tuesday 05 December 2006 16:07, Andrew Dodd wrote:
> Comment: I haven't been following the PVR-500 issues as much as I
> should have. Somehow my Samsung-based 500 started working perfectly
> after a few weeks with no intervention (I'm still running the
> 0.5.something or 0.6.something - I'm not home at the moment to
> check.) I honestly can't figure out why, but I've been afraid to
> touch it. :)
>
> I think I need to browse through the archives this weekend to see
> where things are normally, as it sounds like a lot more is known
> about the card now.
>
> Quoting Haavar Valeur <[EMAIL PROTECTED]>:
> > Do you mean there was no clear criteria when to turn on or off the
> > amp, or do you mean that it was not clear if turning it on or off
> > would improve the picture?
>
> Unfortunately, when you're dealing with receiver nonlinearity and
> frontend overloading, things are never cut and dry in the RF world.
>
> :( Nonlinearity pays my bills (I'm an RF engineer by profession.)
> :
> > What I think is that the lines people see is when the signal gets
> > clipped. That is when the signal strength is too high. This will
> > occur when the amp is on when it should not have been. I think the
> > picture is snowy when the signal is too weak. That is the case when
> > the amp is not on, but it should be on.
> >
> > When I had the amp on, some channels where clear, but others had
> > the lines. When I turned the amp off the most of the channels that
> > was clear now turned snowy, but the channels that had the lines
> > where generally a lot better.
> >
> > Looking at the picture over time I get the impression that the
> > signal for each channel changes. Channels sometimes has the lines
> > and sometimes not. The same with the channels that are prune to
> > snow. I don't think there is a static list of channels that should
> > use the amp and not, but there are some channels that are more
> > likely to need the amp (<130Mhz).
>
> Unfortunately, depending on the receiver design (Hans, does the
> datasheet happen to have a block diagram showing where in the signal
> path the LNA is located?),
Alas, no. I've read through it again but no mention is made where the
LNA is exactly.
> moving the LNA in/out of the signal path
> on a channel by channel basis will likely not help much. Usually
> LNAs are placed as early in the signal chain as possible (It can be
> mathematically shown that the noise figure of the first amplifier in
> your receiver chain is the most important, thus LNAs are usually the
> first thing in the signal path, with a few exceptions.)
> Unfortunately, this also means the LNA does not have too much
> filtering in front of it, which will make it more prone to getting
> overloaded by a strong signal. This strong signal can be anywhere in
> the receivers' frontend passband, which is usually pretty wide. Thus
> a strong signal on one channel can foul up reception of every channel
> in a given band, as tuning of the channels and selectivity *usually*
> occurs after the LNA.
The LNA is in the bandswitch byte (in the datasheet they call the LNA
bit 'bandswitch port3'). This suggests that the LNA is early in the
signal chain.
Hans
>
> I'd really need to know more about the signal chain to be sure
> though.
>
> Given that you are indicating significant differences in signal
> strength between channels, I'm guessing you are using OTA reception
> and not cable? Signal strength that varies over time is to be
> expected with OTA reception, since both multipath and weather can
> affect VHF and UHF propagation significantly.
>
>
> _______________________________________________
> ivtv-users mailing list
> [email protected]
> http://ivtvdriver.org/mailman/listinfo/ivtv-users
_______________________________________________
ivtv-users mailing list
[email protected]
http://ivtvdriver.org/mailman/listinfo/ivtv-users
