[Elecraft] A new Elecraft product?

[Edward R Cole]

I assume anyone interested in this topic is reading all the 
post...so, I randomly chose Brian's post to reply to:

First off, I read a lot of theorizing about the K3 TCXO's and The 
K3EXREF.  I was a field tester of the EXREF and have had it in 
operation continuously since installing it...what follows is not theory:

I have a mw frequency counter that is good to 26-GHz reading down to 
1-Hz accuracy (that is 25 parts per billion).  The counter has an 
internal TCXO which may run whenever the unit is plugged in (I 
haven't checked so that is a guess).  What isn't is that the counter 
reads my rubidium reference oscillator as exactly 10-MHz.  The Rb 
stability (after warm up) is +/-5 E-11 = 5/100,000,000,000 Hz.  This 
is not as good a standard as using a GPS receiver but certainly good 
enough for most of us (assuming you are not active on >10-GHz).

My stations 10-MHz reference is a $45 e-bay surplus Russian OCXO with 
+/-5 E-12 short-term stability (after warm up).  I run it 
continuously on a battery that is float charged by my station 12v PS, 
thus no wait when I turn on my K3/10.

You want to install the TCXO-3 (option) if you are going to install 
the EXREF.  The EXREF provides a reference frequency for the TCXO to 
check against.    The K3 uses the TCXO as reference osc for its DDS 
VFO's and other LO's.  The stability of the K3 is that of the model 
TCXO you have installed, so if you are running the TCXO-3 that is 
listed (pg73 of the manual) as having +/- 0.5 ppm or we could write 
that +/- 5 E-7.  Using 28-MHz as an example, that means the radio 
keeps to +/- 14 Hz if 0.5 ppm is being realized.

The EXREF improves the stability by periodically checking the 
frequency of the TCXO against that of an external 10-MHz reference 
and generating a correction for the TCXO to bring it back to 
48.380.000 MHz.  This happens about every 4-seconds so the TCXO 
drifts normally at 0.5 ppm in between times.  That results in about 
2-Hz drift, so the K3 is never more than +/-2 Hz off frequency when 
using the EXREF.  2-Hz is approximately 0.1 ppm.

There is a momentary frequency step when the EXREF does its check and 
compensation.  This has been a criticism by many "purist".  But I 
have not had this cause any issues when I am running JT65C on 1296-MHz eme.

If +/- 28 Hz is acceptable for your ham radio use, then you will 
probably be happy just running with the TCXO-3.  If you need better 
stability the EXREF will probably suffice (it does for me).

Oh some suggest that the K3 will still drift at power-up because the 
TCXO-3 is warming up.  If you do not use the EXREF that is true.  I 
see about 80-Hz movement after warm up.  But the K3 is kept to +/- 2 
Hz from start by the EXREF so it does not matter that the TCXO has 
drift because that is compensated immediately at start up (if your 
10-MHz reference is stable).  And it continues to be corrected every 
4-seconds after that.

OK, sorry for the long tutorial, but this should remove all 
specializing on this topic.

73, Ed - KL7UW
600meters to 3cm (on 17 bands)
WD2XSH/45 experimental license for 600m
144-eme
1296-eme
--------------
From: Brian <alsopb at nc.rr.com>
To: elecraft at mailman.qth.net
Subject: Re: [Elecraft] A new Elecraft product?
Message-ID: <543FBC64.9080705 at nc.rr.com>
Content-Type: text/plain; charset=utf-8; format=flowed

Don,

The issue isn't whether the TCXO is an improvement over the standard.

The issue is whether the TCXO is good enough for some of the wide range
of uses the K3 is put.

What you are hearing here is that it isn't.  Elecraft has responded with
the external locking option.  That option, being a discontinuous
correction, one has some significant disadvantages to some.

My original posting on this issue was related to the locking option.  It
only seemed reasonable to have it do a sanity check on the difference
between the internal and reference oscillator implied correction before
doing a correction.  Why do the correction is the external oscillator is
warming up and 250 Hz off initially?  The K3 also drifts during warmup
with both the standard and TCXO.  This amounts to between 20 and 50 Hz
(@ 10 MHz) for the first 15 minutes.  Why do the correction until the K3
has is relatively stable.  Perhaps the external locking should not be
done for say the first 15 minutes after K3 turn on?

73 de Brian/K3KO


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