(Was "Flex Power Genius XL 2KW amplifier")
On Thu, Oct 18, 2018 at 5:43 AM Jim Garland <4cx250b@miamioh.edu> wrote:
> (1) Most modern transceivers have provisions for delaying RF to an
> amplifier, usually through a menu setting, to give the amplifier T/R
> relays a chance to close before zapping them with RF. In modern
> transceivers, there is more to this delay than meets the eye.
>
Yes, and in the case of the Elecraft K3 and K3S, it is important to
understand that it does *not* work as documented.
> Consider, for example, the K3 and Flex 6000 series transceivers. The K3
> defaults to an 8msec delay, through menu setting TX DLY, and is
> adjustable up to 20msec.
Yes, that is what the Elecraft menu says, but the actual measured TX DLY is
somewhat different and it has a small variable jitter. N7YK discovered
this when it was apparent that his K3 was hot switching his homebrew QSK
amplifier. I repeated his measurements using a scope on my K3. Both
radios had the KSYN3A synthesizer upgrade, which apparently changed
things. I also measured a friend's K3 that did NOT have the KSYN3A
upgrade, and the results were quite different.
Given:
CONFIG:*TX DLY* *nor 008*
CONFIG:*CW QRQ OFF*
VOX ON
QSK ON
PWR OUT: 5W
FIRMWARE: 5.64
*Actual Measured TX Delay:*
K3 with KSYN3A, or K3S: 6 ms (+/- about 1 ms)
K3 without KSYN3A: 14 ms (+/- about 1 ms)
If you enable CW QRQ mode to improve QSK performance and reduce CW jitter
(at speeds above 37 WPM), things change drastically:
CONFIG:*CW QRQ ON*
*Actual Measured TX Delay with CW QRQ ON:*
K3 with KSYN3A, or K3S: 4.5 ms (+/- about 0.3 ms)
K3 without KSYN3A: 4.5 ms (+/- about 0.3 ms)
In both cases, the measured TX delay is the same in both Semi Break-in (QSK
OFF) and QSK modes.
When using VOX on Phone, the delay between relay closure and first RF is
considerably longer, about 30 ms as I recall, so there is no real concern
on phone.
Conclusion: for CW, don't go by what the K3 TX DLY menu says. Note that
the TX DLY setting has ZERO effect if CW QRQ mode is ON. Otherwise, with
CW QRQ OFF, CONFIG:TX DLY will increase the actually TX DLY by about 1 ms,
for every increment over 8 ms.
So, if you have a K3 with KSYN3A, or a K3S, and your amp. needs an 8 ms
delay between relay closure and first RF, set CONFIG:TX DLY to 010 or 011,
not 008, or use a logging program or Winkey to close the PTT line well
before any CW is sent to the radio.
> The Flex is adjustable from 0msec up to more
> than 1000msec.This delay does NOT mean that any CW characters sent by
> the operator will be lost if they occur before the delay is reached.
>
That's a nice feature of the Flex, though given the approximate 100ms
latency inherent in the receiver, it could make you quite late in the
pileup if you also set TX DLY too high.
In contrast, the K3 will truncate the first CW element (semi break-in) or
*all* CW elements (QSK ) by the measured TX DLY. If your callsign starts
with a dot, it may get truncated considerably at high speed when using VOX,
especially with TX DLY above 008. You can compensate for that using the
WinKey "Keying Compensation" or "1st extension" settings (not "weight").
Some logging programs (Win-Test) provide keying compensation without
needing a WinKey. The K3 internal keyer compensates automatically (no
element truncation).
With the K3, the entire delay range is brief enough that an operator
> probably
> wouldn't notice what's happening.
With external keying at 30 WPM or below, where a dot is 40 ms or longer,
losing 6-8 ms isn't very noticeable. But at 40 WPM, with a dot only 30 ms
long, an 8 ms truncation starts to make the keying sound pretty light, so
keying compensation (not weight) is recommended. The K3's internal keyer
compensates automatically, but external keying needs compensation at high
speeds.
73,
Bob, N6TV
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|