K2: VFO Drift and Temperature-Compensated PLL Reference Upgrade

[Michael J. Linden]

  I'm trying to understand a bit more about temperature-related VFO
drift in the K2. My K2 sits in the basement where it is currently seeing
stable ambient temperatures around 67F. When using 10 MHz WWV as a
reference, I see approximately 40 Hz of drift (10,000.03 to 9,999.99)
over a 30 to 45 minute period from a cold start. I am not transmitting
during this period, I assume this is just due to a slight heating (can't
be much of a temperature change) of various resistors, capacitors,
inductors, and varactors. This is clearly in-specification as it is much
less than the 100 Hz that is specified in the manual. ;^)
 
  My K2 is serial number 4137, so it has had the Temperature-Compensated
PLL Reference Upgrade from the start. The Temperature-Compensated PLL
Reference Upgrade documentation says that you can check the VFO drift by
transmitting for a while to raise the temperature of the radio 15 or 30
degrees F _AFTER_ you have allowed a warm-up period of 30 or more
minutes. This appears to infer that there is a difference in the nature
between warm-up drift and post warm-up drift. If so, can someone explain
what that difference is?
 
  I haven't performed the formal post warm-up drift testing, but I plan
to give it a try to see how stable the VFO is after warm-up. This raises
another question... Will making any changes to resistor RA on the PLL
thermistor board have any affect on the 40 Hz of warm-up drift (as
opposed to the post warm-up drift), or is this something that can't
really be addressed by the Temperature-Compensated PLL Reference
circuit?
 
  Thanks, Michael N9BDF
 
PS  It was nice to see that resistor RA was placed at the top of the PLL
thermistor board -- makes it much easier to fine tune the temperature
compensation!  :^)
 
_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Ron D'Eau Claire-2]

Michael, N9BDF, wrote:

My K2 sits in the basement where it is currently seeing stable ambient
temperatures around 67F. When using 10 MHz WWV as a reference, I see
approximately 40 Hz of drift (10,000.03 to 9,999.99) over a 30 to 45 minute
period from a cold start. I am not transmitting during this period, I assume
this is just due to a slight heating (can't be much of a temperature change)
of various resistors, capacitors, inductors, and varactors. This is clearly
in-specification as it is much less than the 100 Hz that is specified in the
manual. ;^)...

Temperature-Compensated PLL Reference Upgrade documentation says that you
can check the VFO drift by transmitting for a while to raise the temperature
of the radio 15 or 30 degrees F _AFTER_ you have allowed a warm-up period of
30 or more minutes. This appears to infer that there is a difference in the
nature between warm-up drift and post warm-up drift. If so, can someone
explain what that difference is?

-------------------------------------------------

Yes, the additional heating caused by the RF power amplifier running
(especially if you have a K2/100) causes additional drift. Also, even with
the mods installed, there is a little drift in the BFO and PLL reference
oscillators and they tend to drift such that their errors add on the higher
frequency bands and cancel on the lower frequency bands.

My K2 (S/N 1289) showed a drift on 15 meters (where the drift of the
oscillators add to make it worst-case) of up to 240 Hz from a cold start
before the mods.

After the mods, I see about 50 Hz drift from a cold start for about 15
minutes or so. Then, worst case, I see up to 50 Hz additional drift if I
really "hammer" the K2 at 100 watts out for a long-winded CW transmission
(20 minutes or so). That's on 15 meters. On  the lower bands, the drift is
very close to zero, thanks to the canceling effect of the BFO and PLL
reference oscillator.

I believe my Reference Oscillator mod is typical. It is a "breadboard"
version built out of discrete components mounted on the bottom of the main
board that was used in testing the design for the production version of the
circuit. It worked FB, so I never bothered to replace it with the
"production" mod.

Ron AC7AC


_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[John, KI6WX]

Michael;
Let me make a few comments about this circuit.


>   I'm trying to understand a bit more about temperature-related VFO
> drift in the K2. My K2 sits in the basement where it is currently seeing
> stable ambient temperatures around 67F. When using 10 MHz WWV as a
> reference, I see approximately 40 Hz of drift (10,000.03 to 9,999.99)
> over a 30 to 45 minute period from a cold start. I am not transmitting
> during this period, I assume this is just due to a slight heating (can't
> be much of a temperature change) of various resistors, capacitors,
> inductors, and varactors. This is clearly in-specification as it is much
> less than the 100 Hz that is specified in the manual. ;^)

A K2 or K2/100 will warm up about 5-10F above the local ambient temperature
due to the internal power dissipation; more options installed will push this
toward the higher number.
>
>   My K2 is serial number 4137, so it has had the Temperature-Compensated
> PLL Reference Upgrade from the start. The Temperature-Compensated PLL
> Reference Upgrade documentation says that you can check the VFO drift by
> transmitting for a while to raise the temperature of the radio 15 or 30
> degrees F _AFTER_ you have allowed a warm-up period of 30 or more
> minutes. This appears to infer that there is a difference in the nature
> between warm-up drift and post warm-up drift. If so, can someone explain
> what that difference is?

There is no difference in the root cause of the drift.  This adjustment
procedure was designed so that you could make a consistent measurement of
drift as you adjusted the value of RA.  It is possible to reduce the 20
meter drift to less than 50 Hz with a 50F temperature swing.  My K2 has been
tested with this mod over a temperature range of 40F to 130F.  The drift is
not as well compensated below 50F, so you will see somewhat more drift if
you operate outside in the winter.
>
>   I haven't performed the formal post warm-up drift testing, but I plan
> to give it a try to see how stable the VFO is after warm-up. This raises
> another question... Will making any changes to resistor RA on the PLL
> thermistor board have any affect on the 40 Hz of warm-up drift (as
> opposed to the post warm-up drift), or is this something that can't
> really be addressed by the Temperature-Compensated PLL Reference
> circuit?

Reducing the drift following the test procedure will also reduce the warmup
drift.  The warmup drift on my K2 is less than 20 Hertz as determined with
the front panel frequency display and WWV.
>
>   Thanks, Michael N9BDF
>
> PS  It was nice to see that resistor RA was placed at the top of the PLL
> thermistor board -- makes it much easier to fine tune the temperature
> compensation!  :^)
>
This was intentionally done in the design so that any K2 user could improve
the performance with a little additional work.  That is also why RA is a 1/4
watt resistor and all of the other ones on the circuit board are 1/8 watt
resistors.

-John
 KI6WX


_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com