[Elecraft] KAT3 ATU Problem
Tom Boucher
tomg3olb at gmail.com
Wed Dec 15 13:35:49 EST 2021
Mike K8CN wrote
*< It wouldn't have occurred to me to look atthe LPA stage as a source of
trouble, but ifthe TX power is ramping up slowly, then I cansee where the
ATU control algorithm isconfused by the change in power levels bothforward
and reflected. My ATU seems tochange relay settings very fast, and the
controlalgorithm is looking at change direction (maybechange magnitude as
well) in SWR to decidewhich relays to flip next. The relative
adaptationspeed of the control algorithm and the TX powerramp-up could
certainly cause interactions thatmight slow the ATU convergence, but I
wouldthink that once the TX output power stabilized,the ATU control
algorithm would be receivingvalid SWR readings as it flips relays.When you
say it takes seconds to ramp upto full (TUNE) output power, does this
happenon every TUNE button press? Does it also occurwhen you send CW from
the paddle/key? Canyou watch the RF envelope on a scope to timethe power
ramp-up?73,Mike, K8CN >*
Mike,
Thanks for that and I do understand what you say about the possibilty of
the control algorithm getting confused by the slow ramp up of power. Last
week Rob at Elecraft Support had me using a 50 ohm dummy load with the KAT3
bypassed. Set PWR to 5.0 W and hold down XMIT to TUNE, this performs a Tx
gain calibration. Then set to 11.9 W and TUNE again. My results were 5W
immediately but rising from 8.4 to 9.8 W in about 6 seconds
Rob then had me checking WMTR LP setting in CONFIG menu (TECH MD ON). This
was set to the default value of 70 which he suggested I increase . (I'm not
at all sure what this is doing). Increasing this value to the maximum of
180 gave me 10 W increasing to 12.0 W in 2 seconds, faster but still quite
a delay.
This does happen on every TUNE button press but I don't have any problem
during my daily use of the rig on CW. I have not fired up my old Tektronix
scope yet to look at the keying envelope.
Elecraft Support have gone quiet this week so no further progress.
73 and thanks for your help,
Tom G3OLB
More information about the Elecraft
mailing list