[Elecraft] KX3-2M module?

[Edward R Cole]

OK, there has been a lot speculation but nobody has one yet to test.

I both run JT65 on eme and WSPR on VHF and 600 meters, so I am fairly 
experienced in using these modes.  I plan to rigorously test the KX3 
+ 2M for specs and test running these modes with real eme signals.  I 
can run my K3 + 2m transverter (from another mfr) in parallel to 
compare performance receiving signals.  I will test the frequency 
drift transmitting 3w.

I have a EIP-538 mw frequency counter good to 26-GHz which uses an 
internal TCXO or external 10-MHz reference.  I have a Rubidium 
reference oscillator with short-term stability of +/- 5 E-11 (that is 
0.0000000005 or 0.0005 ppm) to use as reference source.  That will 
definitely measure the frequency accuracy and stability of the KX3 + 
2M combination.

I do not have a set up for measuring phase noise, but the parallel 
test described above will compare sensitivity using my K3 with 
TCXO-3/EXREF using an OCXO reference (be my phase noise 
reference).  The OCXO is +/- 5 E-12 with
Phase noise = -86 dBc/Hz, -118 dBc/Hz (100-Hz), -119 dBc/Hz (1KHz).
I will publish my testing results on my website (I'll post the link 
when this is done).
I also operate microwaves up to 10-GHz: currently have 1296-28, 
3400-144, and 10,368-144 MHz transverters.  All operate with an LO 
frequency about 1100-MHz and multiply to the frequency needed to mix 
with the IF radio.

Example: the 10,368 MHz transverter uses an xtal LO = 189.33 MHz 
which is multiplied to 10,224 MHz and mixes with 144-MHz from the IF 
radio.  the stability of the 189.33 MHz xtal oscillator is multiplied 
x54 and added to that of the 144-MHz radio.  It should be obvious 
that the stability of the transverter LO is way more important since 
it gets multiplied.

More and more PLL LO's are being used for this so the resulting LO 
frequency of 10,224 MHz is < +/- 9-Hz.  At worst case this is added 
to the +/- 10-Hz of the KX3+2M to provide a frequency error of up to 
19-Hz.  Wow!

I think folks need to settle down and wait a bit until we can provide 
some independent testing before judging the merits of the KX3+2M.

73, Ed - KL7UW
-------------
Date: Tue, 8 Apr 2014 17:08:29 -0700
From: Wayne Burdick <n6kr at elecraft.com>
To: Walter Underwood <wunder at wunderwood.org>
Cc: "Elecraft at mailman.qth.net Reflector" <elecraft at mailman.qth.net>
Subject: Re: [Elecraft] KX3-2M module?
Message-ID: <854564BD-B009-4BE0-903A-1FC38A3E72B8 at elecraft.com>
Content-Type: text/plain; charset=iso-8859-1

Actually, with extended temp compensation applied, the KX3-2M 
module's long-term frequency stability is typically better than +/- 
0.1 ppm (+/- 10 Hz). But it's the short-term excursions within this 
range that would probably not work with JT65. Again, I haven't tried 
it. These excursions are due to the FLL (frequency-locked-loop) 
algorithm, which obtains data from an EEPROM table having increments 
of about 0.1 degree C (after calibration is complete).

73,
Wayne
N6KR


73, Ed - KL7UW
http://www.kl7uw.com
     "Kits made by KL7UW"
Dubus Mag business:
     dubususa at gmail.com