I continue to be amazed at how much difficulty the overseas manufacturers have
in attaining good, click-free and thump-less QSK
keying where character elements are not truncated....requiring the use of added
weighting through a keyer or other similar means.
No doubt that recent PLL/synthesize designs have contributed to the QSK
problems due to PLL setting times and complex frequency
management schemes with RIT/XIT & wide splits don't help matters either. But
really....how difficult can it be for the three
pre-eminent manufactures to produce acceptable keying? We already know that
Tom has improved the keying on the top-end Yaesu radios
with "less than buck" in parts. I achieved similar results with the Ten Tec
Omni Six with two capacitor and resistor changes. Just
a little bit of attention to the CW keying design in these radios would go a
long way.
I suspect that the keying in the Orion has taken on much of the same ALC and
other control keying circuitry as used on the Jupiter
and Pegasus....which is absolutely exemplary: nearly perfect rise/fall symmetry
and nice rounded edges, notwithstanding the leading
edge of the very first leading edge in a series of characters...and this is
generally due to a slight compromise in the ALC design.
But it is entirely possible to design a ALC circuit which produces zero ALC
power overshoot while retaining a nice initial rise
time.
What does it take for the overseas companies to listen to us on this issue?
Perhaps the FCC should only "type accept" amateur
equipment contingent on the manufacturers' ability to produce evidence that the
occupied CW bandwidth falls within some
pre-determined and limited amount.
-Paul, W9AC
----- Original Message -----
From: "John Rippey" <w3uls@3n.net>
To: <topband@contesting.com>
Sent: Thursday, May 08, 2003 1:59 PM
Subject: Topband: Key clicks
> Earl, K6SE, has posted results of his listening for key clicks with a K-2
> transceiver as compared to his modified/unmodified 1000MP. His conclusion:
> somebody needs to do a mod for the K-2.
>
> Another conclusion: all currently available ham transceivers are equally
> sinful regarding key clicks and it should be something that is cured by the
> manufacturers in the product design phase rather than as post-purchase
> mods. W8JI has made this point many times.
>
> For example, the ARRL Lab in its extended report on the IC-746PRO included
> something like eight separate keying sideband plots, using QSK and
> semi-break-in and various rise times. Interestingly, there is no
> significant difference in the sidebands (down by -35dB @ 1 kHz) no matter
> what the variables. The 756PROII tested by ARRL showed a similar result,
> as did the FT-1000MP Mark V Field, etc., etc. In fact, in an e-mail to me
> some time back, Mike Tracy said that he had yet to find any transceiver he
> had tested that he could point to as having clearly superior keying sidebands.
>
> The ARRL lab's results seem to belie the reports of some hams who have
> adjusted the CW rise times on their DSP-based Icoms up to 6 or 8 ms. If the
> ARRL's keying sideband plots are to be believed, changing the rise times
> has no impact on the sidebands, hence key clicks. What is it? A Gaussian
> curve that is the ideal?
>
> The only significant rig not yet tested is the Ten-Tec Orion and it will be
> interesting to see if that rig does better at the ARRL lab than the -35dB
> down @ 1 kHz that now seems to be the standard among ham transceivers.
>
> 73,
> John, W3ULS
>
> _______________________________________________
> Topband mailing list
> Topband@contesting.com
> http://lists.contesting.com/mailman/listinfo/topband
>
|