[Elecraft] K3 V4.61 CW decode broken?
Was checking out RTTY decode which worked fine. Decided to look at CW decode. Here are my observations. Not working well here or op here inept at using it. My perception is that it did much better in older versions. However, I didn't use it much then. ZL9HM was 599 @35 wpm this AM. WPM check always showing between 48 and 82 wpm. Decode with WPM check off was bunch of E's and T's. Not due to noise. Using 400 Hz filter in CW mode. Tried auto, both choices for speed, auto, different thresholds. Closest to working is auto. Similar problem with C6AKQ at 32 wpm when decoding some characters right. Both stations generating machine generated CW. I'm also thinking that these stations are using "contest" spacing of characters which may be confounding. In WINKEY and N1MM the word spacing for contest spacing is 6 dits/wordspace instead of normal 7. Admittedly these are really short exchanges and the time to lock on is really short. However, my guess is that these are the exact conditions some guys are using the decoder for. A longer rag chewing K9 at 28wpm did decode OK but was not spectacular. Although his signal was strong, characters were dropped. He seemed to be using normal spaced CW. I'm wondering is "contest spacing" is confounding for the decoder. 73 de Brian/K3KO - No virus found in this message. Checked by AVG - www.avg.com Version: 2012.0.2221 / Virus Database: 2634/5443 - Release Date: 12/07/12 __ Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:Elecraft@mailman.qth.net This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html
Re: [Elecraft] K3 V4.61 CW decode broken?
Brian, Virtually all other reports on this release have said the CW decoder was working very well. I just retested this morning, too. If you're seeing E's and T's, try using a much narrower bandwidth, 50-150 Hz, and turn on the noise blanker to reduce the level of impulse noise. I recommend using THR 2 or 3 for most signals, and THR 1 for very weak signals. These fixed settings usually work better than AUTO. When there is no CW signal above the required threshold, you'll see random characters decoded. 73, Wayne N6KR On Dec 7, 2012, at 5:41 AM, Brian Alsop wrote: Was checking out RTTY decode which worked fine. Decided to look at CW decode. Here are my observations. Not working well here or op here inept at using it. My perception is that it did much better in older versions. However, I didn't use it much then. ZL9HM was 599 @35 wpm this AM. WPM check always showing between 48 and 82 wpm. Decode with WPM check off was bunch of E's and T's. Not due to noise. Using 400 Hz filter in CW mode. Tried auto, both choices for speed, auto, different thresholds. Closest to working is auto. Similar problem with C6AKQ at 32 wpm when decoding some characters right. Both stations generating machine generated CW. I'm also thinking that these stations are using "contest" spacing of characters which may be confounding. In WINKEY and N1MM the word spacing for contest spacing is 6 dits/wordspace instead of normal 7. Admittedly these are really short exchanges and the time to lock on is really short. However, my guess is that these are the exact conditions some guys are using the decoder for. A longer rag chewing K9 at 28wpm did decode OK but was not spectacular. Although his signal was strong, characters were dropped. He seemed to be using normal spaced CW. I'm wondering is "contest spacing" is confounding for the decoder. 73 de Brian/K3KO - No virus found in this message. Checked by AVG - www.avg.com Version: 2012.0.2221 / Virus Database: 2634/5443 - Release Date: 12/07/12 __ Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:Elecraft@mailman.qth.net 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:Elecraft@mailman.qth.net This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html
Re: [Elecraft] K3 V4.61 CW decode broken?
Wayne, This did the trick for me as well. I had the bandwidth way too wide and was just seeing ET in the decoder. Been enjoying the new firmware ever since. Works great!!! Thanks, Keith AK6ZZ Sent from my iPhone please excuse typos On Dec 7, 2012, at 7:47 AM, "Wayne Burdick" wrote: > Brian, > > Virtually all other reports on this release have said the CW decoder was > working very well. I just retested this morning, too. > > If you're seeing E's and T's, try using a much narrower bandwidth, 50-150 Hz, > and turn on the noise blanker to reduce the level of impulse noise. I > recommend using THR 2 or 3 for most signals, and THR 1 for very weak signals. > These fixed settings usually work better than AUTO. > > When there is no CW signal above the required threshold, you'll see random > characters decoded. > > 73, > Wayne > N6KR > > > On Dec 7, 2012, at 5:41 AM, Brian Alsop wrote: > >> Was checking out RTTY decode which worked fine. Decided to look at CW >> decode. >> >> Here are my observations. >> >> Not working well here or op here inept at using it. >> >> My perception is that it did much better in older versions. However, I >> didn't use it much then. >> >> ZL9HM was 599 @35 wpm this AM. WPM check always showing between 48 and 82 >> wpm. Decode with WPM check off was bunch of E's and T's. Not due to noise. >> >> Using 400 Hz filter in CW mode. Tried auto, both choices for speed, auto, >> different thresholds. Closest to working is auto. >> >> Similar problem with C6AKQ at 32 wpm when decoding some characters right. >> >> Both stations generating machine generated CW. >> >> I'm also thinking that these stations are using "contest" spacing of >> characters which may be confounding. In WINKEY and N1MM the word spacing for >> contest spacing is 6 dits/wordspace instead of normal 7. >> >> Admittedly these are really short exchanges and the time to lock on is >> really short. However, my guess is that these are the exact conditions some >> guys are using the decoder for. >> >> A longer rag chewing K9 at 28wpm did decode OK but was not spectacular. >> Although his signal was strong, characters were dropped. He seemed to be >> using normal spaced CW. >> >> I'm wondering is "contest spacing" is confounding for the decoder. >> >> 73 de Brian/K3KO >> >> >> - >> No virus found in this message. >> Checked by AVG - www.avg.com >> Version: 2012.0.2221 / Virus Database: 2634/5443 - Release Date: 12/07/12 >> >> __ >> Elecraft mailing list >> Home: http://mailman.qth.net/mailman/listinfo/elecraft >> Help: http://mailman.qth.net/mmfaq.htm >> Post: mailto:Elecraft@mailman.qth.net >> >> 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:Elecraft@mailman.qth.net > > 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:Elecraft@mailman.qth.net This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html