[K2] Is TX frequency susceptible to RF on chassis?

[David Woolley (E.L)]

This may be correlated with the balun thread.

 From the same station, I got a report that I was off frequency on 15m
SSB and a more quantitative one, of 120Hz low, for 10m.  This was during
a local contest, so I couldn't do a lot of investigation on air, and I
may not get many SSB opportunities on these bands in the near future.

I checked the transmit frequency relative to the receive one with a
dummy load, at 10 watts, and there is no problem.  It's possible that
this is simply a combination of my voice and the filter settings, but it
could also be that RF on the chassis is upsetting the BFO and/or
reference oscillator control voltages.  Is this a known issue?

The method I used to check with the dummy load isn't easily adapted to a
live antenna, but as I've observed that the K2 has EMC problems with UHF
transmitters, I tried holding a PMR446 transmitter close to various
chassis seams and managed to upset the frequency by a small amount (beat
frequency when SPOTted onto the 14MHz birdie), maybe no more than about
10Hz.  This happened well after the point where the RF pickup was
upsetting the received audio.

On 15m I was using the balun, but the actual antenna is approximately
resonant (disconnecting either side gives a strong drop in received
noise).  On 10m the antenna would be more like anti-resonant, so I would
expect much more signal on the chassis.  Unfortunately I didn't have a
quantative report for 15m.

If the frequency is shifting due to RF on the chassis, my next priority
really is going to have to be sorting out a good internal antenna
solution.  External antennas are out of the question for various
reasons, that includes flag poles - I don't have exclusive access to
anything at ground level!
--
David Woolley
Emails are not formal business letters, whatever businesses may want.
RFC1855 says there should be an address here, but, in a world of spam,
that is no longer good advice, as archive address hiding may not work.
_______________________________________________
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

[Don Wilhelm-4]

David,

I do not believe that RF on the chassis could easily cause a frequency
shift, but then who knows because each situation like that is different.

I suggest that you first take a good look at your SSB FL1 filter for
both LSB and USB.  If you have the curently normal SSB filter width of
2.4 kHz I recommend that you set up Spectrogram with markers at 300 and
2600 Hz, then adjust the BFO to center the passband between those
markers.  If you have the older 2.1 kHz filter use 300 and 2300 kHz markers.

With the 2.4 kHz filter aligned that way, you should be able to use it
for both transmit and receive with good fidelity and frequency
tracking.  If you have an extremely 'bassy' voice, then you may want to
set FL2 up as I have described and set FL1 for a passband about 50 Hz
lower (250 to 2550 Hz) and use FL2 for receive.

If you are using the K2 frequency indication to 'net' on a given
frequency, your dial calibration may be off a bit.  To get it right,
refer to the K2 Dial Calibration article on my website
http://w3fpr.qrpradio.com for techniques to accurately set the 4 MHz
reference oscillator and the following steps to achieve good dial
calibration.

Once you have assured yourself that the SSB filters are correct, try it
again and see if there appears to be any difference.

73,
Don W3FPR

David Woolley wrote:

> This may be correlated with the balun thread.
>
> From the same station, I got a report that I was off frequency on 15m
> SSB and a more quantitative one, of 120Hz low, for 10m.  This was
> during a local contest, so I couldn't do a lot of investigation on
> air, and I may not get many SSB opportunities on these bands in the
> near future.
>
> I checked the transmit frequency relative to the receive one with a
> dummy load, at 10 watts, and there is no problem.  It's possible that
> this is simply a combination of my voice and the filter settings, but
> it could also be that RF on the chassis is upsetting the BFO and/or
> reference oscillator control voltages.  Is this a known issue?
>
> The method I used to check with the dummy load isn't easily adapted to
> a live antenna, but as I've observed that the K2 has EMC problems with
> UHF transmitters, I tried holding a PMR446 transmitter close to
> various chassis seams and managed to upset the frequency by a small
> amount (beat frequency when SPOTted onto the 14MHz birdie), maybe no
> more than about 10Hz.  This happened well after the point where the RF
> pickup was upsetting the received audio.
>
> On 15m I was using the balun, but the actual antenna is approximately
> resonant (disconnecting either side gives a strong drop in received
> noise).  On 10m the antenna would be more like anti-resonant, so I
> would expect much more signal on the chassis.  Unfortunately I didn't
> have a quantative report for 15m.
>
> If the frequency is shifting due to RF on the chassis, my next
> priority really is going to have to be sorting out a good internal
> antenna solution.  External antennas are out of the question for
> various reasons, that includes flag poles - I don't have exclusive
> access to anything at ground level!

_______________________________________________
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

[David Woolley (E.L)]

Don Wilhelm wrote:
>
> I do not believe that RF on the chassis could easily cause a frequency

I hope that's the case, although UHF RF can cause a small shift, and I
assume the MHz birdies are less affected than the varactor controlled
oscillators.

> I suggest that you first take a good look at your SSB FL1 filter for
> both LSB and USB.  If you have the curently normal SSB filter width of

I'll have another look at that.

> tracking.  If you have an extremely 'bassy' voice, then you may want to

Low on USB would suggest the opposite.

> If you are using the K2 frequency indication to 'net' on a given
> frequency, your dial calibration may be off a bit.  To get it right,

As the frequency appeared to be round, I may have been influenced by
that as well as the correct sound.

> refer to the K2 Dial Calibration article on my website
> http://w3fpr.qrpradio.com for techniques to accurately set the 4 MHz
> reference oscillator and the following steps to achieve good dial
> calibration.

Unfortunately, the sun doesn't seem to favour picking up WWV.  The
original calibration was against the MHz birdies on an FRG7, which was,
I think, calibrated against WWV in the past.  However I just tried on
some 41m broadcast stations, and it looks like the calibration might be
30 to 40 Hz high on that band, which probably equates to 50 to 70 Hz, in
the same sense, on 10m, so I may try a calibration against those
broadcast stations.  (Does anyone know a relatively accurate broadcast
station, receivable in a noisy location in the UK, and near the high end
of the VFO range?)
>


--
David Woolley
Emails are not formal business letters, whatever businesses may want.
RFC1855 says there should be an address here, but, in a world of spam,
that is no longer good advice, as archive address hiding may not work.
_______________________________________________
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

[David Woolley (E.L)]

Don Wilhelm wrote:

 > the same sense, on 10m, so I may try a calibration against those

Oops, I meant 20m.  The same applies if I mentioned 10m in the original.

--
David Woolley
Emails are not formal business letters, whatever businesses may want.
RFC1855 says there should be an address here, but, in a world of spam,
that is no longer good advice, as archive address hiding may not work.

_______________________________________________
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

[Julian, G4ILO]

David.

Your situation as described in another email seems very similar to
mine. I am operating a K2/10 with an indoor antenna, a closed loop
approximately 20m in circumference with the feedpoint about 2m
directly above the K2. The loop is fed via the KAT2, some 2.5m of
RG-213 and an Elecraft BL-1 balun at the feed point. I have no RF
ground, being in an upstairs bedroom. When connected to the power
supply, the K2 is grounded through the electrical supply earth, but I
have never noticed any difference when operating it from its internal
battery, completely isolated.

In over 6 years of operating, including frequency-sensitive digimodes,
I have never noticed any frequency shift of the K2 TX frequency.

I just attempted, by holding the FT-817 with whip antenna right next
to my K2, to test for sensitivity to UHF, while monitoring a PSK31
station on 20m, and there was no perceptible change in frequency.

Therefore I wonder if you should look elsewhere for the cause of the
problem. Before investigating inside the K2, is your power supply
immune to RF? Does the voltage change in the presence of RF?

--
Julian, G4ILO K2 s/n: 392  K3 s/n: ???
G4ILO's Shack: www.g4ilo.com
Zerobeat Ham Forums: www.zerobeat.net/smf


On 10/14/07, David Woolley <[hidden email]> wrote:

> I checked the transmit frequency relative to the receive one with a
> dummy load, at 10 watts, and there is no problem.  It's possible that
> this is simply a combination of my voice and the filter settings, but it
> could also be that RF on the chassis is upsetting the BFO and/or
> reference oscillator control voltages.  Is this a known issue?
_______________________________________________
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

[Don Wilhelm-4]

David,

The same principles work fine for any known frequency, it is just that
the subtraction of the BFO frequency from the VFO frequency is not as
straightforward.

An AM station broadcasting standard and constant tones (such as WWV) is
easiest because one can observe the demodulated audio tones with
Spectrogram assuring accurate tuning.  If you have a good ear, you may
be able to tune to a station broadcasting music to within 5 or 10 Hz.

If you do use an AM broadcast station without standard tones as your
standard, I would suggest that you tune it by ear in LSB or USB first,
then switch to CW while observing the audio with Spectrogram - the
carrier should appear 'straight and tall' at the frequency you have set
in the K2 for the sidetone pitch - thus assuring accurate tuning of the
station.  Then go back to SSB mode to set C22 while doing the
subtraction between the VFO and BFO frequencies.  A note to help - any
adjustment of C22 will move the VFO frequency reading about 5 times
faster than the BFO reading, so do the adjustment with the counter probe
plugged into TP1 and then check the result in TP2.  A few trials should
yield an accurate setting.

73,
Don W3FPR


David Woolley wrote:

> Don Wilhelm wrote:
>
>
>> refer to the K2 Dial Calibration article on my website
>> http://w3fpr.qrpradio.com for techniques to accurately set the 4 MHz
>> reference oscillator and the following steps to achieve good dial
>> calibration.
>
> Unfortunately, the sun doesn't seem to favour picking up WWV.  The
> original calibration was against the MHz birdies on an FRG7, which
> was, I think, calibrated against WWV in the past.  However I just
> tried on some 41m broadcast stations, and it looks like the
> calibration might be 30 to 40 Hz high on that band, which probably
> equates to 50 to 70 Hz, in the same sense, on 10m, so I may try a
> calibration against those broadcast stations.  (Does anyone know a
> relatively accurate broadcast station, receivable in a noisy location
> in the UK, and near the high end of the VFO range?)
>>
>
>

_______________________________________________
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

[Alexander Ponomarenko-5]

Hi David!

For fine tune C22 you can use the simplest method with PC and specrpogram.
There is the original UA6HJQ's method:
-connect K2's AF-out to PC mic-input, and phones to PC phone out;
-turn On K2 and PC, and wait for 50-60 minutes (for temperature stability);
-check the connection "TP1=>Control Board";
-run the Spectrogram, and set for it: Scale -60db, FFT 8192, Band 640 -  
1329Hz
  and set marker to 1000Hz;
-set LSB-mode, and tune VFO at 10001,00kHz (WWV), if ANT is connected
you have to hear WWV-signals in headphones,
-if we see the WWV-signal peaked on 1000Hz (our marker on Spectrogram),
it's the ideal point for C22.
-if not, turn VFO for peaking the WWV-signal at 1000Hz-marker.
-f.e. if we have 10001,20kHz we'll have to subtract 20 (in later steps),
  if we have 10000,80 we'll have to add 20 (in later steps)
- MENU => CAL => CAL FCTR (don't touch VFO-knob!)
- we can see 14914,70 (for example)
- now if we had 10001,20 we must to subtract: 14914,70-20=14914,50
  (if we had 10000,80 => 14914,70+20=14914,90)
-now slowly turn C22 for 14914,50 readings (or 14914,90)
-recall CAL PLL for TP1 on 40m band
-recall CAL FIL for TP2 on each filter/mode.
Now you can recheck your K2 on WWV at 10001,00 kHz, as a rule the error  
makes
few Hz.
Same pictures you can see if follow here (in russian):
http://www.hamradio.cmw.ru/elecraft/calibrated.htm

73! Alex
K2#5287

On Mon, 15 Oct 2007 15:44:31 +0300, Don Wilhelm <[hidden email]>  
wrote:

> David,
>
> The same principles work fine for any known frequency, it is just that  
> the subtraction of the BFO frequency from the VFO frequency is not as  
> straightforward.
>
> An AM station broadcasting standard and constant tones (such as WWV) is  
> easiest because one can observe the demodulated audio tones with  
> Spectrogram assuring accurate tuning.  If you have a good ear, you may  
> be able to tune to a station broadcasting music to within 5 or 10 Hz.
>
> If you do use an AM broadcast station without standard tones as your  
> standard, I would suggest that you tune it by ear in LSB or USB first,  
> then switch to CW while observing the audio with Spectrogram - the  
> carrier should appear 'straight and tall' at the frequency you have set  
> in the K2 for the sidetone pitch - thus assuring accurate tuning of the  
> station.  Then go back to SSB mode to set C22 while doing the  
> subtraction between the VFO and BFO frequencies.  A note to help - any  
> adjustment of C22 will move the VFO frequency reading about 5 times  
> faster than the BFO reading, so do the adjustment with the counter probe  
> plugged into TP1 and then check the result in TP2.  A few trials should  
> yield an accurate setting.
>
> 73,
> Don W3FPR
>
>
> David Woolley wrote:
>> Don Wilhelm wrote:
>>
>>
>>> refer to the K2 Dial Calibration article on my website  
>>> http://w3fpr.qrpradio.com for techniques to accurately set the 4 MHz  
>>> reference oscillator and the following steps to achieve good dial  
>>> calibration.
>>
>> Unfortunately, the sun doesn't seem to favour picking up WWV.  The  
>> original calibration was against the MHz birdies on an FRG7, which was,  
>> I think, calibrated against WWV in the past.  However I just tried on  
>> some 41m broadcast stations, and it looks like the calibration might be  
>> 30 to 40 Hz high on that band, which probably equates to 50 to 70 Hz,  
>> in the same sense, on 10m, so I may try a calibration against those  
>> broadcast stations.  (Does anyone know a relatively accurate broadcast  
>> station, receivable in a noisy location in the UK, and near the high  
>> end of the VFO range?)
>>>
>>
>>
> _______________________________________________
> 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

--
Отправлено M2, революционной почтовой программой Opera:  
http://www.opera.com/mail/
_______________________________________________
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

[michael taylor-3]

On 10/15/07, David Woolley <[hidden email]> wrote:
...
> broadcast stations.  (Does anyone know a relatively accurate broadcast
> station, receivable in a noisy location in the UK, and near the high end
> of the VFO range?)

I there is or was the LDS station from School of Electronic and
Electrical Engineering, Leeds University on 5.000 MHz that is accurate
to around 1x10^(-12). Otherwise there is RWM from Russia is at 4.996,
9.996, 14.996 MHz.

-Michael, VE3TIX
_______________________________________________
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