[Elecraft] KIO3B
iw1ayd - Salvatore Irato
iw1ayd at gmail.com
Sun May 17 07:42:36 EDT 2015
Hi Ed and all.
That's interesting, but not resolutive, nor a full solution for
everyone. It will introduces changes that needs to be addressed.
But even if addressed, those changes, will not give such a great
solution redly usable by for everyone. I.E. not contesting with a medium
or big SO setup.
First, Ed, as you pointed out. there will be the need of another setting
in a new programmed set of TTY programs.
Second, this will be moving the TXing/translating of 7 bits to 5 bits
tasks from the PC or the interface /controller just into the radio. I
don't see it as a solution, but just as integration of another
virtualized items. Any single physical item that fail will trash down
several virtual items.
Third, I strongly hope that the USB onto the KIO3B interface is on an
insulated USB - inside the radio. If it isn't electrical noise noise
will come from each side to the other. Some RFI and EMI may arise in
particular installations. The Liquid Music approach is showing rules
that apply here. The right quality doesn't stay on high ends but in
simple and effective solutions, since the layout of power rails.
The first point is a wait if any programmers is interested it. This will
be a another cable, USB this time, to be connected to the radio (1).
Where is the PTT now, a CAT command on USB? Does anybody still need
sequencing or other signals? Quick and dirty? Add a cable not
subctracting all the others.
The second point also introduce the need of a controller to tie two
radios together. This will not superseded by a USB add on for the radio.
Why a second controller that tie radios to a PC: SO2R. Or, more simply,
control via a single points different commands to different I/O even
with single radio ... not having to use other interfaces and boxes to do it.
USB by itself couldn't seem to be a problem, but it will have some
pitfalls: the radio processor capacity in any season, the needs of a
controller interface for other tasks (i.e. SO2R related), the still
needed capacity of a PC (i.e. its processor and OS configuration) to
pack 7 bits data into the USB stream. This last will not solve the CPU
capabilities issues, but thankfully move all those to better timings
realm. That's the only good thing adding an USB to communicate in
between the radio and the PC. It moves the strict timing problem of a 5
bits UART emulations into the not so strict problems of a simple ASCII
communication - program side. But still leaving all the code or
libraries intervention to communicate on a USB bus with its drivers
deeply run by the same CPU. This is one of the why a multi/multi CPU is
anyway preferred when used with latest OSs.
The third consideration USB insulation may require someone more expert
than me. But out of the noise propagation quite granted with Galvanic,
metallic, continuity there is some more prosaic aspects. Digital signals
doesn't give insulation by itself, that's a job for the wire transport.
Until yesterday we have used insulated interface and now what?
The new KIO3B doesn't seems to me such interesting out of casual RTTY.
It's giving, maybe, a solution that will only aggravate others problems.
Maybe, as others already and rightly pointed out, it's too early to
discuss about. For this reason I am just on the surface here, not going
to deep with other, bot corollaries or mains, considerations
73 de iw1ayd Salvo
PS real changes are to be made on faster and sharable interfaces, data
transports and presentations. Moving a serial in or out a PC will not
solve anything, it will just move the problems. Problems that already
have workable, effective and high available, solutions made within
Microham interfaces. Use a knife as a knife and a fork as a fork.
On 17/05/2015 02:30, elecraft-request at mailman.qth.net wrote:
> Message: 7
> Date: Sat, 16 May 2015 15:45:28 -0400
> From: "Ed Muns"<ed at w0yk.com>
> To:<elecraft at mailman.qth.net>
> Subject: Re: [Elecraft] KIO3B
> Message-ID: <93A7AE215A1F405ABE58128F1D34DCFC at X2201>
> Content-Type: text/plain; charset="us-ascii"
>
> "True" 45.45 baud FSK (with no bit timing jitter and near perfect overall
> bit timing) is available via the KIO3B USB port,*IF* the logger or other
> application program sends ASCII text via KY commands. This is really the
> best way to do FSK with the K3, or any radio. Minimal cabling, no PC Serial
> port, no K3 ACC port connection, highest performance. Only a SMOP for the
> application folks. Elecraft has done their part of the system design quite
> well, including properly filtered FSK for minimum bandwidth.
>
> Ed W0YK
More information about the Elecraft
mailing list