CW Crystal Filter Signal Loss

[Lemkes]

My K2 S/N 3202 exhibits what seems to be undue loss of sensitivity as the CW
crystal filter is narrowed.  After calibrating and recalibrating with
Spectrogram, and signal tracing through the receiver IF section, I purchased
a new set of seven CW crystals.

I installed the crystals today and recalibrated with Spectrogram.  I get
maybe 1 dBm of sensitivity improvement, probably due to my increased
calibration proficiency.  I am using an Elecraft XG1, 1 microvolt, 7040 kHz,
signal generator to measure sensitivity as minimum detectable signal, MDS.

I measure a sensitivity of -137 dBm with the 2.4 kHz SSB filter,
                                    -132 dBm with a 1.2 kHz CW filter,
                                    -129.5 dBm with a 800 Hz CW filter, and
                                    -126.5 dBm with a 400 Hz CW filter.

That's over 10 dB loss of sensitivity as I narrow the filters which can
easily lose a weak signal as I try to isolate it from nearby QRM.  I can't
believe that such loss of performance is normal.  Has anyone else measured
sensitivity with an XG1 1 microvolt signal source?

The filter varactor bias is working evenly across all seven crystals.  If a
varactor was shorted or installed backwards an abnormal bias level would
show up.  Spectrogram shows the filter widths to be proper and also centered
properly on the sidetone offset.  At each filter setting the XG1 signal is
also loudest at the sidetone frequency, 590 Hz.  So, the calibration seems
as good as it can get.

My next steps might be to remove the SSB option, and the Noise Blanker.  Any
other suggestions?  Has anyone had a similar experience?

Thanks in advance for your help es 73, Dick, AD7AF


Dick and Marty Lemke
PO Box 1038
Ocean Park, WA 98640
Tel. [phone]
www.lemkestudios.com

_______________________________________________
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

[Vic K2VCO]

Lemkes wrote:

 > That's over 10 dB loss of sensitivity as I narrow the filters which
can easily lose a weak signal as I try to isolate it from nearby QRM.  I
can't believe that such loss of performance is normal.

I don't think so.  One thing that could cause this is a problem with T7.
  Check that it is wound on an FT37-43 core and has a 5:20 turns ratio.
--
73,
Vic, K2VCO
Fresno CA
http://www.qsl.net/k2vco
_______________________________________________
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-3]

Dick,

So you replaced the crystals without much (if any) change!  If I recall your
original problem it did sound like a possible case of a bad crystal, but
that can be ruled out now.  Your filter may not be causing attenuation after
all and you may have another problem.

It is normal for the MDS to measure higher for more narrow bandwidths, but
yours gets lower.
I believe it is time to settle down and study the situation and the test
setup carefully.  If you are getting higher than nomal noise  after the IF
filter, that could manifest itself as a reduced S+N/N ratio and give you
readings that are based on that broadband noise rather than noise energy
that is reduced by reducing the bandwidth.

Try using Spectrogram to observe the level of the 1 uV signal with each of
the filters - do not touch the receiver or soundcard volume controls after
the first one.  If the peak of the signal remains at the same vertical
deflection point on the Spectrogram screen, the filter is not attenuating
the signal, but something else is 'strange'.  OTOH, if the signal peak level
does decrease as you reduce the filter width, we can continue to look at the
filter itself.

If you can send me some Spectrogram screen capture shots, perhaps I can
compare them with the ambient levels in my K2.  To do that, use the 1 uV
signal input and send screen captures of the passband for each of the filter
widths that you have listed below (2.4, 1.2, 800 and 400).  Also tell me if
the Preamp was on or off.


73,
Don W3FPR

> -----Original Message-----
>
> My K2 S/N 3202 exhibits what seems to be undue loss of
> sensitivity as the CW
> crystal filter is narrowed.  After calibrating and recalibrating with
> Spectrogram, and signal tracing through the receiver IF section,
> I purchased
> a new set of seven CW crystals.
>
> I installed the crystals today and recalibrated with Spectrogram.  I get
> maybe 1 dBm of sensitivity improvement, probably due to my increased
> calibration proficiency.  I am using an Elecraft XG1, 1
> microvolt, 7040 kHz,
> signal generator to measure sensitivity as minimum detectable signal, MDS.
>
> I measure a sensitivity of -137 dBm with the 2.4 kHz SSB filter,
>                                     -132 dBm with a 1.2 kHz CW filter,
>                                     -129.5 dBm with a 800 Hz CW
> filter, and
>                                     -126.5 dBm with a 400 Hz CW filter.
>
> That's over 10 dB loss of sensitivity as I narrow the filters which can
> easily lose a weak signal as I try to isolate it from nearby QRM.
>  I can't
> believe that such loss of performance is normal.  Has anyone else
> measured
> sensitivity with an XG1 1 microvolt signal source?
>
> The filter varactor bias is working evenly across all seven
> crystals.  If a
> varactor was shorted or installed backwards an abnormal bias level would
> show up.  Spectrogram shows the filter widths to be proper and
> also centered
> properly on the sidetone offset.  At each filter setting the XG1
> signal is
> also loudest at the sidetone frequency, 590 Hz.  So, the
> calibration seems
> as good as it can get.
>
> My next steps might be to remove the SSB option, and the Noise
> Blanker.  Any
> other suggestions?  Has anyone had a similar experience?
>
> Thanks in advance for your help es 73, Dick, AD7AF
>

_______________________________________________
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