This webpage is to introduce the Elecraft KX3 which is now my secondary radio and primary
for portable use with microwave transverters. At home it mainly monitors either 2 meters
or 6 meters (50.125).
Here is the review article I wrote for inclusion in the proceedings of the 2012 Microwave
Update held in Santa Clara, CA:
Review of the KX3 (2.8 Mb pdf)
I ordered a custom aluminum heat sink from Fred Meier, VE7FMN, on March 30, 2014. It arrived on
April 10. I chose the plain aluminum finish as paint might interfere with good heat transfer.
The basic price is $60.00 plus $5.50 for stainless 4-40 screws and allen wrench for the machine
screws. Special matte-black power-coating is available at $93.00. You can e-mail him at
fsmeier (at) telus (dot) net, (substitute the proper symbols for "at" and "dot"!).
It took about 20-minutes to open the radio, remove existing heat plate, and install the new heat
Testing results will be added, but testing in simulated JT65 operation, where the KX3 transmits
continuously for 50-seconds, followed by a 70-second rest period (Rx mode), there is nearly no
discernable temperature rise on the heat sink vs the stock radio getting very hot running it in
Initial Frequency Drift measurements were made without running the Reference Frequency Calibration so
radio starting frequency is offset about 20-Hz below dial frequency. Drift is upward changing 9-Hz
in the first 50-second transmission. Temperature does not cool significantly during the 70-second Rx
period so frequency offset accumulates with each sequential Tx period. Curve in period-7 appears to be
bad data due to retrograd frequency change and period-8 was thrown out because of operator error.
In the last two periods drift seems to be the same indicating heating has topped out. Total drift over
the twenty minute test was 103 Hz.
At this point I ran the standard frequency calibration. Frequency calibration brought receiving frequency
to zero error (within +/- 1 Hz). Running six Tx/Rx sequences resulted in upward drift 10,12, 12, 22, 26,
and 28 Hz during each Tx period. Total frequency moved from -4 Hz to +50 Hz, for a 54-Hz drift over
12-minutes of testing.
I performed the Temperature Compensation Procedure, cooling the radio in the refrigerator for 30-minutes:
The radio read 25c when I began the routine:
T=26c, F=887 T=42c, F=526
T=29c, F=784 T=44c, F=494
T=30c, F=764 T=46c, F=454
T=32c, F=724 T=48c, F=432
T=35c, F=662 T=50c, F=418
T=38c, F=582 T=52c, F=406
*Note: F refers to the CW pitch frequency which was set to 550-Hz at the start of the procedure.
I ran another reference frequency calibration after the temperature compensation and REF*CAL moved form
reading 114.272.28 to 114.275.12. Receive frequency is within +/- 1Hz.
I re-ran the JT65 testing sequences to measure frequency drift after temperature compensation was done.
Initial sequence shows the most drift with a peak +13 Hz upwards drift in the first sequence. The remainder
Showed under 10-Hz with the delta trending to +5 Hz in later sequences.
Please check back as add more content to this page! Please read of testing peformed on the new 2M transverter
module from Elecraft.
Return to Home Page