| |
PF1F > TECH 29.03.05 22:43l 106 Lines 4594 Bytes #999 (0) @ WW
BID : 110088PF1F
Read: GUEST
Subj: Re: Transmitter offsets/co-channel
Path: DB0FHN<DB0NOE<DB0GAP<DB0FSG<DB0MRW<DB0WUE<DK0WUE<7M3TJZ<ON0AR<PI8WFL
Sent: 050329/2034Z @:PI8WFL.#NH1.NLD.EU #:33237 [Enkhuizen] $:110088PF1F
From: PF1F@PI8WFL.#NH1.NLD.EU
To : TECH@WW
G0FTD wrote:-
>
> I am aware of a technique with TV transmitter engineers that makes use
> of a specific multiple of co-channel tx offset to reduce patterning
> when enhanced propagation allows.
>
> I wonder if anyone can explain more about it ?
>
> Also can a similar technique be used with things such as sound
> transmitters,
> repeaters etc ?
>
Hello Andy
1) Co-channel offset working - terrestrial TV
Most of Holland is covered by UHF TV transmitters, but there are a few VHF
ones. Four of those VHF ones have a "precision offset" frequency.
(Presumably, all the others are also a bit off their channel but they
DON'T know exactly by how much ;-)
Under lift condtions, you will get patterning.
A fairly large carrier frequency offset (one Dutch one has an 8 kHz
offset) makes the raster so fine that it will, hopefully, not trouble too
much.
For this to work, the frequency difference between the two transmitters
should be reasonably stable, in the Hz or tens of Hz range, and the
practical way to do that is to mutually agree simply to make both
transmitters that frequency stable.
(The transmitters are quite independent. The other one is not even Dutch).
2) Co-channel offset working - VHF ground-to-air radiotelephone
Currently, channel spacing for most channels is 25 kHz. There are 8.33 kHz
spaced channels but they are not the subject of what follows.
A typical bandwidth for a 25kHz-spaced VHF AM comms receiver is 18 or 20
kHz. It is that wide, not to accomodate combined frequency drifts anymore,
but for other good reasons (receiving data is one).
Where a large volume of air is to be served with one radio channel, I have
two, three, or four transmitters on the same nominal frequency.
Of course you could trade in one 25 kHz channel for three 8.33 kHz
channels, but what we want here is a common channel, with no switching and
all users hearing each other, but over a larger - or more awkwardly shaped
- volume than one transmitter could economically serve.
In the case of two transmitters, one is offset -7.3 kHz, the other +7.3.
In the case of three, one is on the channel frequeny, the second is offset
by -5 kHz and the third by +5 kHz.
Where four transmitters serve a channel, the offsets are -7.5, -2.5, +2.5,
and +7.5 kHz.
Finally, I could (but don't) use -8, -4, 0, +4 and +8 kHz offsets to
accomodate as many as five transmitters on one nominal channel.
To the airplane, it hardly matters. When two transmitters come in at once,
the heterodyne will be inaudible on account of the LF lowpass filter in
the receiver which cuts off above 3.2 kHz. The modulation (which is
applied to all transmitters at once, of course) quite undistorted,
considering.
There is a small problem because the delay time of the lines is a little
unpredictable nowadays with all those AD and DA converters and packet
switching ( I compensate for the bulk, but a minor variation remains).
If the audio to one transmitter is delayed more than a few milliseconds
with respect to the other, sometimes pilots notice a "head in the bucket"
effect where the signals are received equally loudly.
3) So much for ground-to-air. The other way is as easy but different.
There are as many receivers as there are transmitters, but their audio is
not mixed. My radio switchboard just picks the first receiver that sees
over 20 uV. If no signal is over 20 uV, the switchboard picks the first
receiver that gets a signal over 2 uV.
(Under 2 uV nothing happens in the first place - remember this is AM)
Whenever a traffic controller feels he can do better, he can lock the
voting system out and choose for himself. This could happen if the loudest
signal is mainly interference, making the weaker one the better
signal-to-noise ratio. Fortunatly, those occasions are rare.
It is clear, then, that aircraft are not supposed to work offset, but it
frequently appears they do. That is because you are hearing not the
aircraft itself, but a retransmission (by an offset-working transmitter)
of the aircraft's transmission on another, at that time coupled, channel.
I have used the first person deliberately; this is definitely how it is
done at Schiphol.
Offset working is great for AM modulation, like television and AM phone.
For FM, things are quite different. There is no simple heterodyne to
filter out (visually on the TV screen, of in the LP filter of the
receiver) and things get rather messy. Fortunately, it is not that much of
a problem in wideband FM work, because the capture effect is so much
greater there.
- Rudolf, PF1F -
Read previous mail | Read next mail
| |
