I just took a look at the keying waveforms on my 1000D (new) and 1000MP (<
2yrs old). Here are the results for 38WPM, 50W output, VOX and QSK:
1000D VOX 1000D QSK 1000MP VOX 1000MP
QSK
total waveform length 32ms 24ms 32ms 35ms
rise time 4ms 4ms 2ms 1ms
fall time 2ms 2ms 2ms 1ms
In all cases, the rising waveform gently curves near the peak -- i.e., it's
not a triangle.
Note the element clipping by the 1000D in QSK mode. This isn't real
surprising to me since the 1000D is a decendent of the 990, which had
horrible element clipping in QSK (and brain-damaged T/R timing in general.)
My scope reveals a disturbing characteristic of the 1000D's QSK
performance -- there is a power spike approximately 50% greater than the
peak waveform approximately 4ms *after* the waveform has fallen to zero. The
spike is less than 1ms in duration. I'll have to check the T/R timing, but I
suspect that this spike occurs after the amp returns to RX bypass so it is
probably not getting amplified. Still, it shouldn't be there. Can anybody
speculate about the cause?
Also note that the 1000MP uses element stretching to compensate for the PTT
delays needed for proper keying of amp relays. However, it appears that this
circuit is overcompensating slighty. For those in the know about T/R timing
in the 1000MP, I have it set to the stock value of 5ms. What you may not
know is that there is a 7ms minimum delay no matter what you set the delay
in 7-4 to (above 8ms, the delay is equal to the value of 7-4.)
The leading edges in both radios seem reasonable to me, but I'd be
interested in comments from the experts as to whether they should be slowed
down. If the 1000D needs a slower rise time, as W8JI says, then the 1000MP
is in even more need (which could explain the complaints about clicks.) One
thing I liked about the TS950SDX was the ability to tailor the rise time of
the leading edge. Unfortunately, that radio put out a massive initial spike
(like 250W!)
73, Dick WC1M
--
CQ-Contest on WWW: http://lists.contesting.com/_cq-contest/
Administrative requests: cq-contest-REQUEST@contesting.com
|