Three Recordings
Locked
 
|
Delete this message now if you are sick and tired of hearing about noisy receivers.
I have made three recordings of three different receivers tuned to 40 meter band noise, all with CW bandwidth of (nominally) 500 Hz centered near 550 Hz sidetone, all with the RF gain full up. One of the three is a K3. If you care to listen to them and then report back with which one you think is the K3, I would be most interested in hearing from you. The files are available at Hector's web site at the following link: http://www.ad4c.us/Elecraft%20K3/shared%20files%20from%20K3%20users/W6LX%20recordings_/ and are called, "#1.wav", "#2.wav" and "#3.wav". (Thanks very much, Hector, for giving up some temporary space to make these available to everybody.) I have asked Hector to delete them after several days. This is not a 'trick' test... I firmly believe all of you that have reported listener's fatigue from the K3... I am trying to determine qualitatively how much noisier a K3 is perceived compared to some other common ham receivers. I wish I had more receivers to sample, but these were all I could immediately get my hands on. In light of the recent announcement that a new DSP board for the K3 with low pass filter is now available, and the fact that many folks were wondering publicly if they really needed the upgrade, perhaps this test will also help you decide. If it is not obvious from these recordings which of the three is the (offensive) K3, then perhaps you are immune to the artifacts and don't need the upgrade. Since these recordings were made, two additional theories have been put forth on this reflector. The first is that even if you can't hear the high frequency artifacts, they may still cause listener's fatigue-- which is quite a provocative idea; the second is that the artifacts are really only a problem under "modulated conditions" -- I took this to mean when a signal present within the receiver bandwidth. I may make new recordings in the future to test these hypotheses. Because it was important to preserve the full 20 kHz audio bandwidth-- not just the CW bandwidth-- these are 44.1 kHz .wav files that I deliberately did not convert to MP3-- and so to keep the file sizes reasonable they are only 15 seconds each... not a whole lot to go by, but I hope enough to make a decent assessment. (Certainly long enough for someone who hears the artifacts to identify them.) They were all recorded straight from the headphone output of each of the three receivers. All recorded within a few minutes of each other on 7057.0 kHz in the evening. (Incidentally, the K3's DSP board is a Rev. B board.) To save reflector bandwidth, why don't you reply to me privately and I will compile and report the group's results. Thanks for your consideration, and I'm looking forward to receiving your reports! Regards, Al W6LX ______________________________________________________________ Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:[hidden email] This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html |
|
|
The spectra can be seen at
http://homepage.mac.com/chen/temp/audio1.png http://homepage.mac.com/chen/temp/audio2.png http://homepage.mac.com/chen/temp/audio2.png (Estimated using cocoaModem's spectrum display with 1.5 second exponential averaging of each FFT bin. Please ignore the red markers, those are the default RTTY tone pairs :-)) Each horizontal dotted green line is a 20 dB step. Labels in the horizontal axes are in Hz. #1 is a typical flat passband and "brick wall" transition band, with very low noise floor in the stop band. Probably from DSP filtering of some kind. But since sharp skirts like this does not occur in nature, it might actually be more fatiguing to the human ear/brain in the long term. #2 has a couple of spurs and minor bumps, but not as weird as #3. #3 is the weirdest looking one. Even though the passband is narrower than #1 and #2, the noise immediately outside the passband is quite high and slowly tapers off to about 2500 Hz, where it takes another step lower. Definitely not something a human is accustomed to hearing. 73 Chen, W7AY ______________________________________________________________ Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:[hidden email] This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html |
|
|
On Jan 6, 2010, at 9:08 PM, Kok Chen wrote: > The spectra can be seen at > > http://homepage.mac.com/chen/temp/audio1.png > > http://homepage.mac.com/chen/temp/audio2.png > > http://homepage.mac.com/chen/temp/audio2.png Oops, the last one should be http://homepage.mac.com/chen/temp/audio3.png This is what cut-and-paste does to you :-) :-) 73 Chen, W7AY ______________________________________________________________ Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:[hidden email] This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html |
|
|
In reply to this post by alorona
My votes in this contest, can we win a P3 :-)
wav #1 is K3. recognized by: -cleanest overall passband (the others obviously have substantial wideband gain *after* the CW filter,resulting in surrounding WB noise) -but K3 has most "bumpy-road" agc of the 3 73' Paul PD0PSB <quote author="Al Lorona"> Delete this message now if you are sick and tired of hearing about noisy receivers. I have made three recordings of three different receivers tuned to 40 meter band noise, all with CW bandwidth of (nominally) 500 Hz centered near 550 Hz sidetone, all with the RF gain full up. One of the three is a K3. |
|
|
Paul,
The K3 "bumpy road" AGC can be cured by use of the AGC menu Threshold and Slope parameters, there is no reason for it to be bumpy. Try Threshold 008 and Slope 002 for an AGC response that behaves much like the K2. The slope can be increased if you want all signals to sound closer to the same audio level, but reduce the Threshold carefully. The Threshold is the major contributor to the "bumpy road" observation. 73, Don W3FPR pd0psb wrote: > My votes in this contest, can we win a P3 :-) > wav #1 is K3. > recognized by: > > -cleanest overall passband (the others obviously have substantial wideband > gain *after* the CW filter,resulting in surrounding WB noise) > -but K3 has most "bumpy-road" agc of the 3 > > 73' > Paul > PD0PSB > > Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:[hidden email] This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html |
|
|
In reply to this post by alorona
I find that #1.wav is the most 'quiet' recording because it contains
only low frequency (in-passband) content. I believe that is the K3. The others have significant response outside the filter passband, and I would refer to that content outside the passband as "noise". I would not refer to the in-passband content as noise attributed to any receiver. 73, Don W3FPR Al Lorona wrote: > Delete this message now if you are sick and tired of hearing about noisy receivers. > > I have made three recordings of three different receivers tuned to 40 meter band noise, all with CW bandwidth of (nominally) 500 Hz centered near 550 Hz sidetone, all with the RF gain full up. One of the three is a K3. > > ______________________________________________________________ Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:[hidden email] This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html |
|
|
In reply to this post by Don Wilhelm-4
Clear Don,
I was referring to this particular wav file, where the AGC threshold is clearly set low and interacting quite strongly with the noisefloor. But ofcourse, not sure yet if I picked the right file :-) 73' Paul PD0PSB
|
|
|
In reply to this post by Kok Chen
Hi Chen,
I wrote you a while back about wanting direct frequency readout on CocoaModem. I have since then discovered a work-around that suits me fine. (1) I set the K3 transciever to 7034.5 kHz (and lock the VFO A) (2) I set the CocoaModem offset to 500 (3) The frequencies on the CocoaModem screen now read 0 for 7035.0, 100 for 7035.1, ..., 1000 for 7036.0, etc., etc. So I just have to remember to add 7035 to the numbers I see. Great program. Use it all the time. Thanks for putting it out for us to use. 73, Oliver W6ODJ ______________________________________________________________ Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:[hidden email] This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html |
«
Return to [K3]
|
1 view|%1 views
| Free forum by Nabble | Edit this page |