[RTTY] Getting rid of clicks--what's the risk?

[Joe Subich, W4TV]

Andy,

 > Keep in mind that this is an examination of the safeguards and
 > pitfalls in one particular model and does not represent any others,
 > but I think it does show that the risks of AFSK are not as universal
 > as some would have you believe.

The point is as you said, "this is an examinations of ... one
particular model and does not represent any others."

The "clean up" you see is due entirely to the Elecraft's unique AFSK
filter.  Before anyone can generalize those results to a wider range
of transceivers, please repeat the measurements on Yaesu FT-1000MP
and Icom 7600/765Pro-x *at a minimum* as examples of transceivers
most likely to be in a competitive contest station.

There is no need to also repeat the measurements on a 20 year old
FT-840 with a badly misaligned analog balanced modulator driven by
an "all knobs max" signal from an interface that hangs an unbuffered
diode detector directly across the output of the sound chip feeding
the dynamic microphone (<5 mV) input ... but it would be interesting
to show just what a segment of the RTTY community is *really* using.

An operator who knows what he's doing - and a great many RTTY DXers
and contesters fall in that category - can generate clean RTTY with
waveshaped AFSK.  However, many of them do not want to give up the
receive benefits of the RTTY mode on their transceivers.

In addition, there are many, many operators who simply do not know
any better.  They will run MMTTY without any AFSK output filtering;
they will not use good engineering practices when connecting an
AFSK interface.  To top it all off, many will purchase the cheapest
interface they can find, connect it to the mic and headphone jacks
of the transceiver and listen to the "experts" who say to run the
interface "all knobs to the right" making any adjustments only with
the transceiver's mic gain control.

73,

    ... Joe, W4TV


On 1/20/2013 11:03 PM, aflowers at frontiernet.net wrote:
>
> "Someone transmitted a bad signal using AFSK.  Therefore you
> shouldn't do it."
>
> Your decision whether or not to use waveshaped AFSK today shouldn't
> be based on the fact that its technically possible to do something
> bad.  The big guns in the RTTY world understand their stations and
> they know far above the average yahoo when it comes to the "wrong
> thing to do" with AFSK.  As of this moment, rtty clicks are an issue
> to many people who understand it and know why it is an issue for
> their situation.  I remember (barely) when the "key click mods" were
> making the rounds on a manufacturer's radios.  Not all of them got
> changed, but I'm willing to bet that those that continued to be used
> at the big contest stations by and large did.  Why?  Because their
> users understood the issue and were willing to do surgery on kilobuck
> radios to make things better for the people around them.  The people
> doing the modification didn't get one dB out of it.  Everything I
> have seen in the RTTY contesting world leads me to believe these are
> the same type of self-sacrificing people, if not in many cases the
> very same people.  For those who understand the size of their
> footprints, one-size-fits-all answer that we should wait for
> manufacturers to change things and click away in the meantime doesn't
> sit well.
>
> This is probably applies to a minority of people on this list, but it
> is also a very visible minority.  This isn't anywhere near as big of
> an issue for the occasional operatator with a 100W and a dipole as it
> is for the contest CQ machine with 1000W and a stack.
>
> Right now you many of these people do not have to wait for
> manufacturers to change their FSK keying circuits (or in many cases
> what is more likely to be a firmware revision).  AFSK with a
> waveshaped input may be an alternative.  Is that ideal?  Hardly.  In
> fact it's kinda silly when you think about it from an engineering
> point of view, but it is a very viable means to an end and is being
> done properly by the vast majority of AFSK users in contests.  The
> $64,000 question: it a good idea for you to do it?  Alas, the
> corollary to the above is also true:  just because you can *possibly*
> transmit a click-free doesn't mean you should do this either.
> Ah...economic utility rears its confounding head again.
>
> So, in an effort to quantify the risk-benefit tradeoff in my
> particular case, I took a very popular contesting radio and set out
> to transmit the most rotten signal I could.
>
> That's right--I tried to do all the bad things AFSK abusers are
> supposedly prone to doing and attempted to spread wreckage across the
> band with my K3.  I documented the wreckage for your entertainment
> and also so that other K3 owners could take a look and have some
> examples for what the risk/benefits are without having to attempt to
> break their own radios.   I've tried to make it as value-free as I
> can so that you can look at the spectra and decide for yourself what
> for you, if you are a K3 owner.  Even if you are not a K3 owner, you
> might still find it interesting.  I had fun write writing it and
> learned quite a bit about my radio.
>
> Here is where you can read my epic trail of destruction:
>
> http://www.frontiernet.net/~aflowers/k3_afsk/
>
> We can talk about the carnage if you find anything interesting.
> There are suprises in there for those willing to take the time.
> Enjoy!
>
> Keep in mind that this is an examination of the safeguards and
> pitfalls in one particular model and does not represent any others,
> but I think it does show that the risks of AFSK are not as universal
> as some would have you believe.  For the record I don't really care
> what you choose to do at your station so long as you do so
> intelligently.  If anyone wants to put together a plan to get the
> manufacturers' attention to start cleaning up FSK generators I am
> willing to be a part of it.  It seems that is a point of agreement
> here and I'm willing to work in the long term.  My inbox is open for
> ideas (and no Bob, I'm not calling my congressman).
>
> Oh, and please don't shoot the messenger.
>
> Regards,
>
> Andy K0SM/2