Why K3 appears noisy?
Locked
 
|
Rather than debate whether K3 is noisy I suggest to debate why it appears so to many. Then it can lead to fixes. Below are a few hypotheses.
1. Highest AGC threshold is too low. 2. Gain distribution under RF gain. RF gain under AGC does not reduce noise as much as with AGC off. The reduction with AGC on seems to affect the first stages first. So for a large decrease of RF gain the noise reduction is not as strong. Try listening to the same station with AGC off and AGC on while adjusting RF Gain. On the same note, the audio sounds beautiful on strong stations with large AGC hold but there is lots of noise when the station goes off. 3. AF distortions as documented by UR5LAM in http://ur5lam.jimdo.com/elecraft-k3-mod-s-page/ These may fixed with the new DSP board. Partially as UR5LAM also applied fixes to mixers and IF amp. found a few other fixes necessary. The simplest fixes IMHO would be to extend the range of AGC THR, and then make RF gain behave similarly with AGC on and off. I noticed a large improvement after RF calib. For (3), a separate EQ setting per mode would help (if not already done). The noise is an issue important if K3 is marketed to everyone. Like most in this group, I find K3 highly superior to other radios in contesting and DX'ing. Ignacy, NO9E |
|
|
I believe I've written about this before, at least to a few, but perhaps it bears repeating. I have done a side-by-side comparison of my K3 with another K3 when "RF" (actually IF) gain was reduced with the AGC on. With my radio, tuned to a S9+50 dB BC station, even a slight decrease in gain was met by a large *decrease* in SNR. On the other this wasn't the case. IMHO, this results from the wide differences in the volts v. dB characteristic of the gain-controlled hardware, i.e. the FET/diode post crystal filter amplifier, from radio to radio. This is why one guy can write, "SNR is improved when I reduce RF Gain" and another can say, "SNR goes to pot when I reduce RF Gain." Both can be correct, depending on their hardware. It's not new knowledge that we want to delay the application of AGC to early stages in the radio so as to maintain SNR. But in this radio, one size of delay doesn't fit all. The RF gain "calibration" is an attempt to fix in firmware what is essentially a hardware problem and Lyle has done an excellent job doing this. If I understand correctly, (always a dubious proposition) he measures the gain v. volts of the FET amplifier and applies this new found correction to the gain control partitioning. This has made a remarkable difference in my radio, which must be a real outlier. Still, and this is conjecture, I have this nagging concern that there might be some latency in the process and under highly dynamic conditions, it "doesn't keep up." Wes N7WS --- On Thu, 12/17/09, Ignacy <[hidden email]> wrote: > 2. Gain distribution under RF gain. RF gain under AGC does > not reduce noise > as much as with AGC off. The reduction with AGC on seems to > affect the first > stages first. So for a large decrease of RF gain the noise > reduction is not > as strong. Try listening to the same station with AGC off > and AGC on while > adjusting RF Gain. On the same note, the audio sounds > beautiful on strong > stations with large AGC hold but there is lots of noise > when the station > goes off. > > 3. AF distortions as documented by UR5LAM in > http://ur5lam.jimdo.com/elecraft-k3-mod-s-page/ > These may fixed with the new DSP board. Partially as UR5LAM > also applied > fixes to mixers and IF amp. found a few other fixes > necessary. > > The simplest fixes IMHO would be to extend the range of AGC > THR, and then > make RF gain behave similarly with AGC on and off. I > noticed a large > improvement after RF calib. For (3), a separate EQ setting > per mode would > help (if not already done). > > The noise is an issue important if K3 is marketed to > everyone. Like most in > this group, I find K3 highly superior to other radios in > contesting and > DX'ing. > > Ignacy, NO9E > ______________________________________________________________ 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 Ignacy
Hi Wes:
I took a look at the RF Gain Calibration on the K3 Utility and they offer two ways of doing it. One requires a signal generator and one uses factory default values. Is the factory default method satisfactory or is it necessary for us to purchase the XG2 to do the calibration properly. Thanks, Mike K2MK Wes Stewart Thu, 17 Dec 2009 09:34:56 -0800 I believe I've written about this before, at least to a few, but perhaps it bears repeating. I have done a side-by-side comparison of my K3 with another K3 when "RF" (actually IF) gain was reduced with the AGC on. With my radio, tuned to a S9+50 dB BC station, even a slight decrease in gain was met by a large *decrease* in SNR. On the other this wasn't the case. IMHO, this results from the wide differences in the volts v. dB characteristic of the gain-controlled hardware, i.e. the FET/diode post crystal filter amplifier, from radio to radio. This is why one guy can write, "SNR is improved when I reduce RF Gain" and another can say, "SNR goes to pot when I reduce RF Gain." Both can be correct, depending on their hardware. It's not new knowledge that we want to delay the application of AGC to early stages in the radio so as to maintain SNR. But in this radio, one size of delay doesn't fit all. The RF gain "calibration" is an attempt to fix in firmware what is essentially a hardware problem and Lyle has done an excellent job doing this. If I understand correctly, (always a dubious proposition) he measures the gain v. volts of the FET amplifier and applies this new found correction to the gain control partitioning. This has made a remarkable difference in my radio, which must be a real outlier. Still, and this is conjecture, I have this nagging concern that there might be some latency in the process and under highly dynamic conditions, it "doesn't keep up." Wes N7WS ______________________________________________________________ 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 Ignacy
Ignacy,
Tnx for the input. I am gong to order the Rev D DSP board with the low pass filter for one of my K3's so I can compare to the other one. Has anyone else applied all the fixes that UR5LAM has installed referring to the 1st and 2nd mixer, IF amp, and Audio Freq Amp? 73, N2TK, Tony -----Original Message----- From: [hidden email] [mailto:[hidden email]] On Behalf Of Ignacy Sent: Thursday, December 17, 2009 10:23 AM To: [hidden email] Subject: [Elecraft] Why K3 appears noisy? Rather than debate whether K3 is noisy I suggest to debate why it appears so to many. Then it can lead to fixes. Below are a few hypotheses. 1. Highest AGC threshold is too low. 2. Gain distribution under RF gain. RF gain under AGC does not reduce noise as much as with AGC off. The reduction with AGC on seems to affect the first stages first. So for a large decrease of RF gain the noise reduction is not as strong. Try listening to the same station with AGC off and AGC on while adjusting RF Gain. On the same note, the audio sounds beautiful on strong stations with large AGC hold but there is lots of noise when the station goes off. 3. AF distortions as documented by UR5LAM in http://ur5lam.jimdo.com/elecraft-k3-mod-s-page/ These may fixed with the new DSP board. Partially as UR5LAM also applied fixes to mixers and IF amp. found a few other fixes necessary. The simplest fixes IMHO would be to extend the range of AGC THR, and then make RF gain behave similarly with AGC on and off. I noticed a large improvement after RF calib. For (3), a separate EQ setting per mode would help (if not already done). The noise is an issue important if K3 is marketed to everyone. Like most in this group, I find K3 highly superior to other radios in contesting and DX'ing. Ignacy, NO9E -- View this message in context: http://n2.nabble.com/Why-K3-appears-noisy-tp4181768p4181768.html Sent from the Elecraft mailing list archive at Nabble.com. ______________________________________________________________ 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 ______________________________________________________________ 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 Mike K2MK
No "official" reply to this yet. I assumed that the factory default option was simply to restore the radio to its original state if the user wasn't happy after doing the calibration.
Julian, G4ILO. K2 #392 K3 #222 KX3 #110
* G4ILO's Shack - http://www.g4ilo.com * KComm - http://www.g4ilo.com/kcomm.html * KTune - http://www.g4ilo.com/ktune.html |
|
On Sun, 20 Dec 2009 07:53:45 -0800 (PST), "Julian, G4ILO"
<[hidden email]> wrote: In answer to K2MK's question... You need to do the RF Gain calibration with a signal generator to be most accurate, but you don't have to buy the XG2. You could use any signal generator. Tom, N5GE [hidden email] K3 #806 with SUB RX, K3 #1055, PR6, XV144, XV432, KRC2, W1, 2 W2's and other small kits 1 K144XV on order http://www.n5ge.com http://www.swotrc.net > > > >Mike K2MK wrote: >> >> I took a look at the RF Gain Calibration on the K3 Utility and they offer >> two ways of doing it. One requires a signal generator and one uses factory >> default values. Is the factory default method satisfactory or is it >> necessary for us to purchase the XG2 to do the calibration properly. >> > >No "official" reply to this yet. I assumed that the factory default option >was simply to restore the radio to its original state if the user wasn't >happy after doing the calibration. > >----- >Julian, G4ILO. K2 #392 K3 #222. >* G4ILO's Shack - http://www.g4ilo.com >* KComm - http://www.g4ilo.com/kcomm.html >* KTune - http://www.g4ilo.com/ktune.html ______________________________________________________________ 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
Amateur Radio Operator N5GE
|
|
|
I haven't read all the details but I have the impression that you need
to be able to set 1 and 50 microvolts which requires rather considerable attenuation for some common signal generators such as my Elenco SG 9500. Is my impression correct? On the Clifton Laboratories web site, there is a graph labeled Elecraft K3 Transceiver Incremental Incremental RF Gain Reduction 9dB) versus RF Gain Value. Is this effect along with basic calibration that the gain calibration is intended to correct? My K3 doesn't have the discontinuity or discontinuities that others have mentioned; I just curious. See: http://www.cliftonlaboratories.com/elecraft_k3_and_manual_rf_gain_control.htm Dunc, W5DC Radio Amateur N5GE wrote: > On Sun, 20 Dec 2009 07:53:45 -0800 (PST), "Julian, G4ILO" > <[hidden email]> wrote: > > In answer to K2MK's question... > > You need to do the RF Gain calibration with a signal generator to be > most accurate, but you don't have to buy the XG2. You could use any > signal generator. > > Tom, N5GE > > [hidden email] > K3 #806 with SUB RX, K3 #1055, PR6, > XV144, XV432, KRC2, > W1, 2 W2's and other small kits > > 1 K144XV on order > > http://www.n5ge.com > http://www.swotrc.net > > > > >> >> Mike K2MK wrote: >> >>> I took a look at the RF Gain Calibration on the K3 Utility and they offer >>> two ways of doing it. One requires a signal generator and one uses factory >>> default values. Is the factory default method satisfactory or is it >>> necessary for us to purchase the XG2 to do the calibration properly. >>> >>> >> No "official" reply to this yet. I assumed that the factory default option >> was simply to restore the radio to its original state if the user wasn't >> happy after doing the calibration. >> >> ----- >> Julian, G4ILO. K2 #392 K3 #222. >> * G4ILO's Shack - http://www.g4ilo.com >> * KComm - http://www.g4ilo.com/kcomm.html >> * KTune - http://www.g4ilo.com/ktune.html >> > > ______________________________________________________________ > 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 > > ______________________________________________________________ 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 |
|
|
For the RF Calibration Utility you only need a 50 uV source. For S-Meter Calibration you need 1 uV in addition to 50 uV. The factory defaults for both are for "average" units. To achieve custom calibration for variations in individual units, you need to do individual calibrations. The RF Calibration (in the K3 Utility) is totally automatic once you apply a 50 uV source. S-Meter calibration (in the operating manual) is a little more involved but still quite easy to do. 73, Bill |
|
|
In reply to this post by Dunc Carter - W5DC
For the RF Calibration Utility you only need a 50 uV source. For S-Meter Calibration you need 1 uV in addition to 50 uV. The factory defaults for both are for "average" units. To achieve custom calibration for variations in individual units, you need to do individual calibrations. The RF Calibration (in the K3 Utility) is totally automatic once you apply a 50 uV source. S-Meter calibration (in the operating manual) is a little more involved but still quite easy to do. 73, Bill |
«
Return to [K3]
|
1 view|%1 views
| Free forum by Nabble | Edit this page |