I am looking for ways to optimize CW performance. I am getting increased reports that some ops need to fill in missing information "in their heads" while copying my CW. I have now engaged auto-space as a way to help mitigate the problem.
The control site uses fast Comcast broadband Internet service. The remote site is limited to a marginal DSL connection. The remote site typically has about 8 Mbps download and only 600 kbps upload. Obviously, the upload speed is a severe limitation.
In reading p. 93 of the RRC manual, there's a discussion of the LF DELAY and KEY DELAY values. But I don't see a discussion on optimization or a direct relationship between those parameters, especially how to set KEY DELAY with changes in LF DELAY. I have copied the relevant portion of the RRC manual below.
When I ping the remote site, this is my response:
Time=47ms TTL=53
Approximate round trip times in milli-seconds:
Minimum = 46ms, Maximum = 48ms, Average = 47 ms.
So, based on this information, it appears LF DELAY should be 48ms, but what value for KEY DELAY? Is there a network utility to measure jitter to come up with an answer that is better than an educated guess, or based on trial and error? Tnx!
Paul, W9AC
"...The value of lf-delay depends on the latency of Internet. A good start value is
the same value as the ping time between control and radio QTH.
Another unique feature is a kind of adjustable jitter buffer at the radio end. If the Internet
connection has varying latencies, you can set a key-delay before the keying of the radio is
executed. This delay makes it possible to reconstruct the transmission on-the-fly and the
system will be much more tolerant against varying latencies (jitter). Longer key-delays give
higher tolerance against jitter but also add delay until the keying is executed at the radio.
Even if you have a good connection you could try with a low value. Remember that when you
increase this value you must also increase the lf-delay value."
