Joe,
I said:
> ... I think it does show that the risks of AFSK are not as universal
> as some would have you believe.
Reading that back makes it look like it might have been directed at you,
personally. I didn't mean it that way, please forgive me if that is how it
came across. I want to make it clear that I'm not trying to steer people away
from good quality FSK keying schemes (yours or anyone else's), especially when
there are good reasons for those people to use them. You have good arguements
with which I am in agreement and I don't think you are trying to mislead
anyone.
What started this discussion was complaints about wide signals and some on this
list made claims that this was an FSK/AFSK issue. I and others have
demonstrated that this is a waveshaping/filtering issue at the transmitter and
makes no difference how that modulation is done. It was a new realization for
many people that their FSK generators "click" and have very real consequences
that generate complaints. Now we know that we can look in the mirror and stop
blaming mythical AFSK operators for that portion of our woes.
(For the record, MMTTY filters its output by default so I suspect a large
number of AFSK users are waveshaping without knowing it.)
VE3KI's observation that not all radios have the "idiot-resistant" nature that
the K3 has is what made me undertake that little exercise. I imagine it is a
rather extreme example in the grand scheme of things. Even so, I was rather
taken back by just how hard it was to mess things up even without the added
protection of the AFSK TX filter in the DSP chain. It's too bad that FSK
generator is significantly worse from the QRM point of view, at least for the
moment. That may be a notable exception that is worth something to people who
own K3s. There are probably quite a few connected to amplifiers, so if there
is a easy, low-risk way for them to declick we all benefit.
To take your other extreme, I actually have an FT-840 as luck would have it.
It now serves as a signal generator for various experiments rather than a
communication device. As I'm sure you know, AFSK through the (dynamic) mic
jack is the only option on this radio. Without any kind of feedback that seems
to me to be like tightrope walking blindfolded without a net--it can be done,
but that doesn't make it a good idea. It would be an interesting exercise to
see just how, eh... clean.... it can be under ideal circumstances. If there is
enough sincere (morbid?) interest in that I'd be happy to abuse it and take
pictures, but it doesn't matter for the current issue since those aren't the
radio largely hooked up to kilowatts on the contest weekend. At least the MP
has a "Packet" port in the back, but I will take it on faith that it is not
forgiving with gain levels if one decides to wander off script. Joe, you would
certainly know more about
that than I.
I emailed about 20 people who had noticeably narrow signals during the roundup
to figure out what they were doing right (many are on this list). I had to
choose CQing stations (its a software thing) that were >50dB SNR to be able to
get a good picture of what the signals looked like. All but two of them were
K3's running AFSK with MMTTY, an only a few were using the TX filter that does
all the magic (there was a Flex 5K and there was an Omni VII in there, also
MMTTY/AFSK). I also found one AFSK signal with some low-level buzz and a spur
at -40dBc, and it was probably a little worse than the clicks would have been
in FSK mode--it was a common contesting radio, modulated through the mic jack.
I'm quite certain that problem has already been rectified.
I think that little sample says more about how lazy people are in hooking up
their radios for RTTY than it says about Elecraft's "contest market share",
so don't too excited about the numbers. (It does make me wonder about the
unintended consequences of burying the line-in port in a 50-pin connector that
only elves can solder, but I'll let that be.)
I am quite sure that same attitude when applied to some other radios speaks
straight to your points, Joe.
Regards,
Andy K0SM/2
_______________________________________________
RTTY mailing list
RTTY@contesting.com
http://lists.contesting.com/mailman/listinfo/rtty
|