[Elecraft] K3 QSK Timing

[Victor Rosenthal 4X6GP]

I have finally achieved one of my life goals and am the owner of a real 
digital oscilloscope (Siglent SDS1202X-E).

What is great about this compared to my prior life with super-cheap 
n-hand scopes is that it can actually measure things. So the first job I 
gave it was to measure the time between KEY OUT going to ground and the 
start of RF from my K3, as well as the time between the end of RF and 
KEY OUT opening.

It is a K3 upgraded with the new synthesizers. So I would expect the K3S 
to be similar, but if anyone feels like measuring one, I would be 
interested. MCU Firmware is 5.64 (not quite the latest).

So here is what I determined: with 35 wpm dits (or dahs, it doesn't 
matter) from the internal keyer in Full QSK mode, CONFIG TX DLY = 8 
(normal), there is a minimum of 5.6 ms between KEY OUT low and the first 
whiff of RF. I say minimum, because there is a certain amount of jitter 
in the length of the element, and the interval is sometimes a bit 
longer. The interval after RF before KEY OUT goes high is similar.

In semi-QSK the delay before RF is the same.

Setting TX DLY to a greater value subtracts time from the keyed element 
and adds it to the delay. So setting it to 9 gives you a delay of 6.6 ms 
but the period that RF is on is 1 ms shorter.

It's interesting that the normal delay with QRQ mode on is still 5.6 ms. 
And the jitter is still noticeable. I didn't try changing TX DLY with 
QRQ mode on.

Results were not significantly different at speeds of 25 and 40 wpm.

So if you are using it to drive an amplifier without special timing 
circuitry (i.e., where the key line passes through the amplifier before 
going to the transceiver), make sure the relays in the amplifier are 
fast enough!
-- 
73,
Victor, 4X6GP
Rehovot, Israel
Formerly K2VCO
CWops no. 5
http://www.qsl.net/k2vco/