Posted by
Joe Subich, W4TV-4 on
Dec 20, 2009; 5:17pm
URL: http://elecraft.85.s1.nabble.com/Re-K3-receiver-noise-questionable-test-results-tp4192141p4195231.html
> I note that all except the K3 show the taper from 1600 Hz to
> the upper passband edge that follows something similar to a
> "pink noise" spectrum.
That's fairly clear but since 1600 Hz to the top edge of the
passband is less than one octave it is difficult to determine
if the roll-off is "pink" (-3dB/octave) or "brown" (-6db/octave).
> It is interesting that the K3 Norm does show some falloff, but
> not nearly as much as either the FT-2000 or the MK V or IC-706.
I should not that the roll-off in the "Norm" setting for this K3
(S/N 622) is due entirely to the roofing filter. If I make the
same measurements in LSB, the LF end of the spectrum will be
similarly effected and if I run the tests on my other K3 there
will be a slight (<1 dB) valley in the midrange. Roofing filters
are never absolutely flat but a 1 dB ripple is nothing to be
concerned about. Even the "wide" setting has some ripple but not
enough to be numerically significant given the resolution of
Spectrogram.
> I am ready to conclude that the K3's ultra-flat in-passband
> response is the cause of several of the "noisy" responses.
I don't know if it is the cause of the "noisy" responses but it
is probably to blame (along with the headphone response) for the
perception that the K3 is "harsh" sounding. Since the human ear
is most sensitive in the 2000 - 4000 Hz octave, the flat response
of the K3 is perceived as "more noise" by someone using SSB or
CW with wide filters.
> That can be easily handled by shaping with the RX EQ if one
> desires.
The most critical RX EQ band is 2400 Hz. Even setting it to
-4 or -5 dB without any change at 1600 and 3200 Hz does a lot
to "warm up" the audio and make it sound more like other rigs.
> I do data modes frequently, and prefer the flat response of
> the K3.
RX EQ is disabled in DATA mode. If you use DATA A instead of
USB you can adjust RX EQ at will for SSB and not effect data
mode response.
> The audio response above the filter passband apparently has less
> influence than I had originally expected.
I don't know about that. I can clearly hear several of the
ADC clock products even though they are 40 dB or more below
audio in the voice passband. They stand out clearly ... as
much as 20 dB above the "out of band" noise level. There is
also considerable dynamic "program related" noise in the
stopband that I'm looking to clean up with the lowpass filter.
However, I suspect the audio response is only marginally related
to the "noisy receiver" complaints.
73,
... Joe, W4TV
> -----Original Message-----
> From: Don Wilhelm [mailto:
[hidden email]]
> Sent: Sunday, December 20, 2009 12:42 AM
> To:
[hidden email]
> Cc:
[hidden email]
> Subject: Re: [Elecraft] : K3 receiver noise - my test results (long)
>
>
> Joe,
>
> Fine business on those measurements, and thank you for providing the
> data. I note that all except the K3 show the taper from 1600
> Hz to the
> upper passband edge that follows something similar to a "pink noise"
> spectrum. The K3 is flat without EQ. It is interesting that the K3
> Norm does show some falloff, but not nearly as much as either the
> FT-2000 or the MK V or IC-706.
>
> I am ready to conclude that the K3's ultra-flat in-passband
> response is
> the cause of several of the "noisy" responses. That can be easily
> handled by shaping with the RX EQ if one desires. I do data modes
> frequently, and prefer the flat response of the K3.
>
> The audio response above the filter passband apparently has less
> influence than I had originally expected.
>
> 73,
> Don W3FPR
>
> Joe Subich, W4TV wrote:
> >> It would be interesting (at least to me) for someone to
> make similar
> >> measurements on an FT-1000 - perhaps with the INRAD filter. I am
> >> especially interested to know if the in-passband response shows a
> >> similar taper at 2 kHz and above, and what is the response for the
> >> FT-1000 above the filter passband.
> >>
> >
> > Here are some measurements using Spectrogram and broadband noise:
> >
> > FT-2000 --- MK V --- IC-706 ------ K3 ------
> > Analog DSP MKIIG Norm BW=4 EQ
> >
> > 50 -34 -44 -49 -46 -35 -7 -35
> > 100 -31 -24 -27 -40 -18 -6 -18
> > 200 -17 -10 -9 -21 -5 -2 -5
> > 300 -6 -4 -4 -11 -1 0 -1
> > 400 -3 -2 -1 -5 0 0 0
> > 600 -2 -1 0 -1 0 0 2
> > 800 -1 -1 0 0 0 0 1
> > 1000 -1 -1 0 -1 0 0 0
> > 1200 0 -1 0 -2 0 0 0
> > 1400 0 0 0 -1 0 0 -2
> > 1600 -1 -2 -1 -2 0 0 -3
> > 1800 -2 -3 -2 -3 0 0 -4
> > 2000 -5 -5 -2 -4 -1 0 -5
> > 2200 -7 -6 -3 -6 -2 0 -7
> > 2400 -8 -8 -4 -8 -2 0 -8
> > 2600 -10 -10 -6 -15 -3 0 -9
> > 2700 -23 -13 -7 -27 -4 0 -10
> > 2800 -34 -19 -11 -31 -8 0 -14
> > 2900 -41 -32 -24 -33 -18 0 -24
> > 3000 -50 -43 -36 -35 -58 0 -60
> > 3100 -42 -67 0
> > 3200 -58 0
> > 3300 0
> > 3400 0
> > 3500 0
> > 3600 -1
> > 3700 -2
> > 3800 -4
> > 3900 -5
> > 4000 -13
> > 4100 -34
> > 4150 -66
> >
> > All measurements were made at the speaker output for consistency.
> > All measurements were made in USB mode with the default filter
> > settings. FT-1000MP Mark V was measured with both the analog
> > and DSP (100-3100 Hz setting) detectors.
> >
> > The two additional K3 measurements are 1) FM filter, BW=4.00
> > and FC=2.00 and 2) "NORM" using the "pink EQ" settings (1.60=-3,
> > 2.40=-5, 3.20=-6).
> >
> > Note the K3 audio amplifier is the cleanest hands down. The other
> > transceivers had noise floors at about -120 dBV (IC-706 mkIIg was
> > -100 dBV). Except for what appear to be artifacts of the ADC
> > clock at 4, 8, 12, 16, 18 and 20 KHz the K3 out of band audio
> > noise is <140 dBV.
> >
> > 73,
> >
> > ... Joe, W4TV
> >
______________________________________________________________
Elecraft mailing list
Home:
http://mailman.qth.net/mailman/listinfo/elecraftHelp:
http://mailman.qth.net/mmfaq.htmPost: mailto:
[hidden email]
This list hosted by:
http://www.qsl.netPlease help support this email list:
http://www.qsl.net/donate.html