[Elecraft] Transverter calibration

[Mike Harris]

G'day,

Another data point to add. Apologies for it being a bit long.

My K3 is fitted with the line up REFLOCK, EXTREF (10MHz GPSDO) and 
KTCXO-3 +/-1ppm. In the past I have done several warm up runs from cold 
start to to T+4 hours where for all practical purposes KTCXO-3 drift as 
measured by the EXTREF module stopped. The results are repeatable with 
small variations, cold start temperature and room temperature were a 
reasonably consistent, +23C. Front panel temperature change gives a 
reasonable indication of the warm up environment of the KTCXO-3.

My KTCXO-3 at power up starts at 49.379.836 on average and after warm up 
stops drifting at 49.379.890. It goes through two plateaux phases the 
first at T+5 minutes (49.379.863) until T+26 minutes (49.379.865). It 
then starts a slow rise to T+120 where after it enters the second 
plateaux (49.379.884) where after there is little change. This is 
repeatable. However, it must be noted that this is my KTCXO-3 other 
units will have their own characteristic, even drifting in the opposite 
direction.

I was unable to programme my GPSDO for both 10MHz and 145.5MHz outputs, 
maybe a forbidden combination, must get to grips with this. So I decided 
to do a run with a sniff of the 145.5MHz signal fed into the 2m input 
and the K3 not locked to an external source.

The manual REFCAL setting was at the top end of the drift range 
(49.379.886).

After cold start and a quick tune to the 145.500.000 signal the initial 
dial reading had to be set 176Hz high. There after retune took place for 
the first hour at one minute intervals and the dial error recorded. 
Retune was accomplished by offsetting the dial slightly LF and netting 
with TAP SPOT. There is a small 3Hz difference between netting from LF 
or HF using this method so for consistency LF was always used.

The change in dial reading was consistent with the previously measured 
warm up drift characteristic of my KTCXO-3:

T+1   +176Hz
T+5   +98Hz
T+28  +86Hz
T+60  +17Hz
T+80  +11Hz
T+100 +9Hz
T+120 +8Hz
T+180 +4Hz
T+240 +1Hz

At T+240 the manually set REFCAL frequency and the true drifting 
frequency of the KTCXO-3 will have converged. The dial frequency and the 
signal frequency has also essentially converged.

When I figure out how to programme two frequencies out of my GPSDO or 
find a suitably stable 2m source it will be interesting to run the same 
test with my K3 EXTREF locked.

There is great faith placed in digital displays to tell the truth, 
however, there are factors at play which can render this truth to be a 
little coloured. Any radio with a free running reference oscillator for 
example will be in error at some point.

I standby to be shot down.

Regards,

Mike VP8NO

On 23/06/2018 19:41, W1GO (Joe) wrote:
> Alan,
> 
> I have the same issue; mine is off about 50Hz.  Not a huge deal but just enough to be annoying on VHF SSB and CW.  I do have the REFLOCK, EXTREF (10MHz GPSDO) and KTCXO-1; nothing left to add.  You are correct that you can’t adjust the offset once you’re in this configuration.  As I say, for me its “only” about 50Hz but its consistent.  I was gonna have to Elecraft team look at it in their lab the next time I send the rig in (if I send it in).  Aside from that I’m living with the 50Hz delta.
> 
> Joe
> W1GO