Phil
You need to look at the character error rate and signal to noise ratio
curves. I carefully said with noise and without propagation effects like
selective fading, flutter etc. In this case between 2 and 2.5dB will take
one from 10% error to 1%. It is also dependent on the decoder being used.
The problem with taking into account selective QSB the error rate is very
dependent on the decoder you are using. With deep frequency selective QSB
MMTTY will be totally in error while decoders like 2Tone will carry on with
good copy. In these circumstances increasing power will have no effect on a
MMTTY decoder while 2Tone will give better copy with more power up to a
certain point as selective QSB also comes with multi-path propagation. So
trying to define XdB will result in Y improvement in error rate becomes
difficult to quantify as it is dependent on the exact propagation at any one
time.
As for radio based FSK have you actually measured your occupied bandwidth? I
have done this for 2Tone DOOK by transmitting on a K3 and receiving it on
another K3 in my shack. It is very narrow without any other artefacts about
350Hz at -80dB. The FT1000MP I used to have before I got the 2nd K3 was
wider at this level due to close in phase noise but otherwise was of similar
bandwidth at -60dB. Try doing the same with a rig based FSK system and see
how wide it is.
Another aspect of radio FSK is timing jitter this is significant when a
windows PC is "bit banging" a serial port to key the radio's FSK line. This
can and does cause decode errors. Audio has no timing jitter so avoids this
problem.
I agree a few people use low tones and overdrive their radio so it produces
audio harmonics or have RF feedback. My on air experience is very few use a
poorly set up radio as I can go for days and not hear one. A few may get it
wrong does not mean it is a poorer system for the vast majority. I have read
similar statements before about audio from NA amateurs. This causes me to
speculate whether NA amateurs are not so good at setting up their radios
compared to this side of the Atlantic ocean.
73 David G3YYD
-----Original Message-----
From: RTTY [mailto:rtty-bounces@xxxxxxxxxxxxxx] On Behalf Of Phil Sussman
Sent: 27 May 2017 07:58
To: David G3YYD
Cc: rtty@xxxxxxxxxxxxxx
Subject: Re: [RTTY] Power
My thoughts on RTTY and RF power are below:
1. A 'clean' signal will provide significantly better results than an
overdriven one.
The ability to decode an RTTY signal, first and foremost, benefits from FSK
rather than AFSK when setting up a station. Yes, AFSK, when properly
adjusted, will mirror FSK, but ask yourself how many times a poorly shaped
signal is the results of misadjusted AFSK.
2. I agree with David that power can have an effect; however, it is an order
of magnitude. I disagree that a 2db increase in power will significantly
decrease the error rate. Depending upon propagation and general background
(e.g. a contest going on or a misadjusted adjacent signal) modestly
increasing transmitter power will have little, if any, effect.
3. Antenna efficiency and feedline loss are far more significant that
transmitter power. Improving copy at the other end, in my opinion, requires
perhaps a 6 db improvement. As an added benefit reducing losses also
improves your receive sensitivity.
4. There is no doubt that running 100w (e.g. full power on a 100w rig) is
better than running 50w. But only slightly better, 3bd. Running full power
reduces the duty cycle and while you may not "burn out" your finals in the
short run, you will induce more heat and shorten overall life. Oddly enough,
reducing power by 50% increases life and reduces duty cycle (considering
RTTY to be 100% duty cycle) by more than 50%.
5. Propagation conditions are far more important than a 3db increase in RF
power. When conditions are optimum I have run 1w (or less) with the same
results as 50w. Under less than optimum propagation conditions, I have tried
using 1Kw when 50w was marginal and I was advised that my signal went from
barely copyable to readable but weak.
6. Some RTTY operators run lots of power which (unfortunately) causes other
stations to also have to increase power, just to compete. When that does is
drive the lower power signals away.
So, my suggestion is to improve efficiency of feedlines and antennas.
Raising an antenna
by only a few feet or replacing a lossy feed line will accomplish far more
than 3db.
'End of soap box'
73 de Phil - N8PS
Quoting David G3YYD <g3yyd@xxxxxxxxxxxxxx>:
> There has been recent discussion about power and the difference it
> will make. Unless you have studied this before you may be amazed at
> what sort of difference power can make to RTTY decode error rate.
>
> A received signal with decode errors cause by noise and not by
> propagation effects like selective fading, flutter etc has a character
> error rate of 10%. By increasing power by 2dB, e.g. 100 watts to 160
> watts the error rate will drop to about 1% or for most amateur purposes
perfect copy.
>
> So changing that long run of RG58 coax to RG213 or putting the antenna
> up a bit higher or turning the power up from 50 watts to 100 watts can
> significantly improve copy at the far end. Better copy at the far end
> = more QSOs.
>
> I note many run their 100 watt rigs at around 50 watts because they
> think the rig cannot take the full power on RTTY. I have news for you
> all the rigs I have ever had I have run at full power on RTTY without
> ever having blown the PA devices. The typical 13.8v 100 watt AB class
> PA has similar device dissipation at 50 watts out as at 100 watts out.
> Some PAs will have less dissipation at 100 watts than 50 due to the PA
> being in gain compression(saturation). The PSU will be supplying more
> power but again all of the commercial 13.8v SMPSU I have used have
> worked perfectly at 100 watts output.
>
> 73 David G3YYD
> _______________________________________________
> RTTY mailing list
> RTTY@xxxxxxxxxxxxxx
> http://lists.contesting.com/mailman/listinfo/rtty
_______________________________________________
RTTY mailing list
RTTY@xxxxxxxxxxxxxx
http://lists.contesting.com/mailman/listinfo/rtty
_______________________________________________
RTTY mailing list
RTTY@xxxxxxxxxxxxxx
http://lists.contesting.com/mailman/listinfo/rtty
|