[RTTY] rtty

[Kok Chen]

On Sep 28, 2008, at 2:55 PM, WS7I wrote:

> Truth is 1500 watts on PSK will rule the day!!!!!

There are more nuances than that.

PSK is phase shift modulation.  PSK31 also amplitudes modulate the  
envelope, otherwise raw PSK bandwidth would be about as wide as RTTY!   
Precise envelope shaping is what keeps a PSK31 signal nice and narrow,  
and thus it is important that the PA is very linear, or PSK signals  
can become very wide.  This is the "IMD" measure that PSK31 ops talk  
about.

Anyhow, because information is carrier using phase modulation,  
multipath will completely ruin reception.  Imagine a reference phase  
that takes two ionosphereic paths from the transmitter to the  
receiver: the received signal is a vector sum from the two paths and  
the phase will bounce all over the place as the path changes rapidly.

Now, imagine what happens when you raise the transmitter power.  The  
receiver will see the same relative power from each path -- thus the  
received phase will be wrong by exactly the same amount as if you are  
running low power!

With PSK31, the signal just needs to be a little above the noise and  
you are off to the races, more power from there will do you no good.   
Compare the Berr vs Eb/No plots of PSK vs FSK and you will find that,  
for the same error rate, PSK will get through with an average of 3 to  
4 dB less power.  Because of FEC, QPSK31 is potentially even better  
than BPSK31.

Very tiny bits of power is enough to carry on a PSK31 QSO (Barry has  
vouched for PSK31's weak signal capability from his PSK Roundup  
experience some years back :-).

Once the ionospheric path degenerates, however, it usually comes with  
extra paths (one of the causes of selective fading that we RTTY ops  
only know too well).  Unlike RTTY, pushing more power does nothing to  
help resolve the problem since the phase error is a constant no matter  
how much power you apply to PSK31.

Unless you are talking about overcoming QRM that is right on top of  
your signal, increasing power will not help get you get through any  
better.  And a QSY works better anyway when that happens.

This is why seasoned PSK31 ops use low power.  Very-low-power.   
Because there is no reason to do otherwise.  From practice, they have  
learned that increasing power won't do any good -- their real life  
experiences confirm theory.

You have to wonder why QRP ops with small verticals or attic dipoles  
in their CC&R homes are gravitating to PSK31 and not to RTTY?   I  
think the answer is very simple, they can make QSOs while using low  
power and inefficient antennas.

The interesting thing is that I saw many local "PSK" callsigns in  
yesterday's RTTY contest.  These are the folks you see on PSK all the  
time.   You can also tell by how they call stations and how they send  
exchanges.  But many of them can't get the DX to hear them.  These are  
the same guys who are used to working Europe from the Pacific  
Northwest on PSK31 with 25 watts and a vertical.  If they can't do the  
same using RTTY, I'm afraid that they'll all go back to calling CQ on  
PSK31 when the contest is over.  There is really only a 2x difference  
in power requirement between PSK31 and RTTY, but that might be enough  
advantage for some people.

MFSK modes such as MFSK16, Olivia and DominoEX do even better than  
BPSK31.

Take a look at Daniel HB9TPL's article "Signal Resilience to  
Ionospheric Distortion of HF Digital Chat Modes" in the November/ 
December 2007 issue of QEX.

The article shows that under good ionospheric conditions, RTTY has the  
worst performance.  When you add fading conditions (2 ms multipath  
delay), RTTY and Feld -Hell are about equal, but still worse than  
other modes.

When you add flutter (less than 1 ms multipath delay), PSK31 and  
DominoEX give up the ghost, while RTTY and Feld-Hell work if you dump  
enough power into it, while the MFSK modes such as MFSK16 and Olivia  
hums along just fine.

(While DominoEX is an MFSK mode, it suffers from having an interleaver  
that is is too short, and thus it not useful when ionospheric  
conditions degrade.)

Under fluttered or high latitude disturbed conditions, Olivia beats  
RTTY by some 24 dB (!) -- i.e., when Olivia can get through with 6  
watts, RTTY needs 1500 watts.

Wide band demodulators is not by itself the culprit either.  Poor rigs  
account for that.

The Flex 5000 has a 192 kc wide pass band (yep, not just 3 kc wide but  
192 kc wide) and yet it will copy a weak signal when a 90 dB stronger  
signal  (15 'S' Units) is next to it, or 2 kc away or 100 kc away --  
of course the rig has a good mixer and the built in "sound card" (an  
Asahi chip) has a dynamic range of better than 123 dB.

Taking the Elecraft K3 mixer output into a N8LP LP-PAN box and feeding  
the LP-PAN into a good sound card will do pretty much the same thing  
-- 190 kc passband, and today limited only by the dynamic range of the  
sound card that you use.

Of course you can't expect this kind of dynamic range from  
"traditional" receivers when used in wide band mode.  Even the Icom  
7800 only has 80 dB of 2kc dynamic range; the 756 Pro II has a dynamic  
range of just 75 dB (the numbers are taken from Rob Sherwood's web  
site) -- requiring their owners to use narrow filters for RTTY  
operation.  But the fact is that there are lots of people today who  
can safely receive very wide passbands and not be bothered by a couple  
of loud stations.

73
Chen, W7AY