K3 CWT decoding update please
Locked
 
|
running vs 4.67 | 1.115 | 2.81 | 2.81 | 1.03
I remember there was a discussion about improving the K3's cw decoding after some agc upgrade that took place a long time ago. I wonder if what we have now is the 'final' version that we are just going to have to settle for? I ask because I am still having a lot of trouble with cw decodes I run thr 2 and cw 5-40 but decoding is just about useless I have tried agc slow/fast and off I have played with rf gain as well as filter widths (I have 200 & 500hz filters as well as the 2.7 ssb filter installed) with my new external svga display for the P3, I was looking forward to doing some lazy cw monitoring on the external monitor at best I can copy strong stations like ARRL's code practice runs well, but trying to decode most other cw qso's is pretty bad. what was the older firmware version that did such a good job with decoding? also if you do have it working, what setting are you using that do work? TIA -- GB & 73 K5OAI Sam Morgan ______________________________________________________________ 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 |
|
|
Sam,
I have found that the first step is to make sure the signal is tuned as close as possible. Signals way off frequency will not decode well at all. Then by setting the RF gain fully clockwise, and NR off, set the threshold to a value that allows the CWT indicator to flash in time with the CW signal. If it is on too long, adjust the RF gain down a bit. I can usually find a setting that works very well for most decent CW. Guys sending with lllooonnnggg dahs and short dits are hard to decode. So is irregular rhythm or speed. Bad spacing is also a problem. But if adjusted well, it works fine. ...bill nr4c -----Original Message----- From: Sam Morgan [mailto:[hidden email]] Sent: Wednesday, June 19, 2013 12:45 PM To: [hidden email] Subject: [Elecraft] K3 CWT decoding update please running vs 4.67 | 1.115 | 2.81 | 2.81 | 1.03 I remember there was a discussion about improving the K3's cw decoding after some agc upgrade that took place a long time ago. I wonder if what we have now is the 'final' version that we are just going to have to settle for? I ask because I am still having a lot of trouble with cw decodes I run thr 2 and cw 5-40 but decoding is just about useless I have tried agc slow/fast and off I have played with rf gain as well as filter widths (I have 200 & 500hz filters as well as the 2.7 ssb filter installed) with my new external svga display for the P3, I was looking forward to doing some lazy cw monitoring on the external monitor at best I can copy strong stations like ARRL's code practice runs well, but trying to decode most other cw qso's is pretty bad. what was the older firmware version that did such a good job with decoding? also if you do have it working, what setting are you using that do work? TIA -- GB & 73 K5OAI Sam Morgan ______________________________________________________________ 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 |
|
|
Sam...
In my *limited* experience with the K3 decoder, I have found that the quality of the sent cw is a big factor in effective decoding. For example, leaving all radio settings constant, one signal will decode perfectly, while another of the same strength and general speed will be gibberish. Another test is to copy W1AW code practice; decodes perfectly every time here. Of course, that is machine sent code. Signals sent at speed using a bug/Vibroflex seem to give the decoder a fit. ...robert On 6/19/2013 19:51, Bill Conkling wrote: > Sam, > > I have found that the first step is to make sure the signal is tuned as > close as possible. Signals way off frequency will not decode well at all. > Then by setting the RF gain fully clockwise, and NR off, set the threshold > to a value that allows the CWT indicator to flash in time with the CW > signal. If it is on too long, adjust the RF gain down a bit. I can usually > find a setting that works very well for most decent CW. Guys sending with > lllooonnnggg dahs and short dits are hard to decode. So is irregular rhythm > or speed. Bad spacing is also a problem. But if adjusted well, it works > fine. > > ...bill nr4c > > > -----Original Message----- > From: Sam Morgan [mailto:[hidden email]] > Sent: Wednesday, June 19, 2013 12:45 PM > To: [hidden email] > Subject: [Elecraft] K3 CWT decoding update please > > running vs 4.67 | 1.115 | 2.81 | 2.81 | 1.03 > > I remember there was a discussion about improving the K3's cw decoding after > > some agc upgrade that took place a long time ago. I wonder if what we have > now > is the 'final' version that we are just going to have to settle for? > > I ask because I am still having a lot of trouble with cw decodes > I run thr 2 and cw 5-40 but decoding is just about useless > I have tried agc slow/fast and off > I have played with rf gain as well as filter widths > (I have 200 & 500hz filters as well as the 2.7 ssb filter installed) > > with my new external svga display for the P3, I was looking forward to doing > > some lazy cw monitoring on the external monitor > > at best I can copy strong stations like ARRL's code practice runs well, > but trying to decode most other cw qso's is pretty bad. > > what was the older firmware version that did such a good job with decoding? > > also if you do have it working, what setting are you using that do work? > > TIA > -- Robert G Strickland, PhD ABPH - KE2WY [hidden email] Syracuse, New York, USA ______________________________________________________________ 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 |
Re: K3 CWT decoding update please
|
|
And CW that is not perfectly sent will give most any CW decoder "fits".
The best CW decoder is the human brain, it is more tolerant of imperfections in timing and band conditions than any other decoder. I just tried the K3 decoder on a W1AW bulletin broadcast, and with the QRN on the 40 meter band, copy was not perfect, but it was reasonable. My head could fill in the blanks easily. 73, Don W3FPR On 6/19/2013 5:47 PM, Robert G Strickland wrote: > Sam... > In my *limited* experience with the K3 decoder, I have found that the > quality of the sent cw is a big factor in effective decoding. For > example, leaving all radio settings constant, one signal will decode > perfectly, while another of the same strength and general speed will > be gibberish. Another test is to copy W1AW code practice; decodes > perfectly every time here. Of course, that is machine sent code. > Signals sent at speed using a bug/Vibroflex seem to give the decoder a > fit. > ...robert > ______________________________________________________________ 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 |
|
|
So..."Just Learn Morse Code"
73, Deni F5VJC On 20 June 2013 00:00, Don Wilhelm <[hidden email]> wrote: > And CW that is not perfectly sent will give most any CW decoder "fits". > The best CW decoder is the human brain, it is more tolerant of > imperfections in timing and band conditions than any other decoder. > > I just tried the K3 decoder on a W1AW bulletin broadcast, and with the QRN > on the 40 meter band, copy was not perfect, but it was reasonable. My head > could fill in the blanks easily. > > 73, > Don W3FPR > > > On 6/19/2013 5:47 PM, Robert G Strickland wrote: > >> Sam... >> In my *limited* experience with the K3 decoder, I have found that the >> quality of the sent cw is a big factor in effective decoding. For example, >> leaving all radio settings constant, one signal will decode perfectly, >> while another of the same strength and general speed will be gibberish. >> Another test is to copy W1AW code practice; decodes perfectly every time >> here. Of course, that is machine sent code. Signals sent at speed using a >> bug/Vibroflex seem to give the decoder a fit. >> ...robert >> >> > ______________________________**______________________________**__ > Elecraft mailing list > Home: http://mailman.qth.net/**mailman/listinfo/elecraft<http://mailman.qth.net/mailman/listinfo/elecraft> > Help: http://mailman.qth.net/mmfaq.**htm<http://mailman.qth.net/mmfaq.htm> > Post: mailto:[hidden email].**net <[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 |
Re: K3 CWT decoding update please
|
|
In reply to this post by RobertG
Chuckle.....
Gary Vk1ZZ K3, KX3, KPA500-FT, KAT500-FT,P3. On 20/06/2013 7:47 AM, "Robert G Strickland" <[hidden email]> wrote: > Sam... > In my *limited* experience with the K3 decoder, I have found that the > quality of the sent cw is a big factor in effective decoding. For example, > leaving all radio settings constant, one signal will decode perfectly, > while another of the same strength and general speed will be gibberish. > Another test is to copy W1AW code practice; decodes perfectly every time > here. Of course, that is machine sent code. Signals sent at speed using a > bug/Vibroflex seem to give the decoder a fit. > ...robert > > On 6/19/2013 19:51, Bill Conkling wrote: > >> Sam, >> >> I have found that the first step is to make sure the signal is tuned as >> close as possible. Signals way off frequency will not decode well at all. >> Then by setting the RF gain fully clockwise, and NR off, set the threshold >> to a value that allows the CWT indicator to flash in time with the CW >> signal. If it is on too long, adjust the RF gain down a bit. I can >> usually >> find a setting that works very well for most decent CW. Guys sending with >> lllooonnnggg dahs and short dits are hard to decode. So is irregular >> rhythm >> or speed. Bad spacing is also a problem. But if adjusted well, it works >> fine. >> >> ...bill nr4c >> >> >> -----Original Message----- >> From: Sam Morgan [mailto:[hidden email]] >> Sent: Wednesday, June 19, 2013 12:45 PM >> To: [hidden email] >> Subject: [Elecraft] K3 CWT decoding update please >> >> running vs 4.67 | 1.115 | 2.81 | 2.81 | 1.03 >> >> I remember there was a discussion about improving the K3's cw decoding >> after >> >> some agc upgrade that took place a long time ago. I wonder if what we have >> now >> is the 'final' version that we are just going to have to settle for? >> >> I ask because I am still having a lot of trouble with cw decodes >> I run thr 2 and cw 5-40 but decoding is just about useless >> I have tried agc slow/fast and off >> I have played with rf gain as well as filter widths >> (I have 200 & 500hz filters as well as the 2.7 ssb filter installed) >> >> with my new external svga display for the P3, I was looking forward to >> doing >> >> some lazy cw monitoring on the external monitor >> >> at best I can copy strong stations like ARRL's code practice runs well, >> but trying to decode most other cw qso's is pretty bad. >> >> what was the older firmware version that did such a good job with >> decoding? >> >> also if you do have it working, what setting are you using that do work? >> >> TIA >> >> > -- > Robert G Strickland, PhD ABPH - KE2WY > [hidden email] > Syracuse, New York, USA > ______________________________**______________________________**__ > Elecraft mailing list > Home: http://mailman.qth.net/**mailman/listinfo/elecraft<http://mailman.qth.net/mailman/listinfo/elecraft> > Help: http://mailman.qth.net/mmfaq.**htm<http://mailman.qth.net/mmfaq.htm> > Post: mailto:[hidden email].**net <[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 Deni F5VJC
not very helpful to have garbage code displayed on the screen
when trying to demonstrate the ' CWT feature' of the K3 to someone new to the hobby and not yet proficient in cw :-( On 6/19/2013 5:12 PM, F5vjc wrote: > So..."Just Learn Morse Code" > > 73, Deni F5VJC -- GB & 73 K5OAI Sam Morgan ______________________________________________________________ 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 |
Re: K3 CWT decoding update please
|
|
If that is the case, tune to a machine sent CW signal under good
conditions (low QRN and no QRM) - the decode will be good. Even if there are errors in the decoded CW, that should be an indication that decoded CW is not perfect, and thus an encouragement to learn CW in order to "fill in the blanks". Your demo might rather be a decode of perfectly sent CW from a local transmitter operating into a dummy load rather than an "on the air" transmission with all the attendant QRN and QRM that is present with such signals. No CW decoder is perfect, but the K3 decoder is one of the better ones. I am not a great CW operator, but I can 'hold my own' at 20 WPM, and sometimes at 25. It does take practice, but I would advise anyone contemplating CW operation to learn CW rather than depending on a decoder. There are many operators out there who do not use machine sent code, and there are even some that use a bug or straight key with an 'accent' - CW decoders will have a hard time with that CW, but the human brain can figure it out. While copying W1AW to see what would happen, I noted that the Farnsworth spacing used at lower speeds did not decode as words - there was a space between each letter - and yes, that is what the Farnsworth method does, it sends each character at a higher rate than the set speed, but there is a delay between characters, which the decoder properly interprets as a space. 73, Don W3FPR On 6/19/2013 7:55 PM, Sam Morgan wrote: > not very helpful to have garbage code displayed on the screen > when trying to demonstrate the ' CWT feature' of the K3 > to someone new to the hobby and not yet proficient in cw :-( > ______________________________________________________________ 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 k5oai
I just loaded the latest firmware a few weeks ago and was impressed with how well CW decode now works. My best results were with NB and NR off, AGC-S, RF gain turned down so ambient noise just kicks the meter, THR=AUTO, speed 5-40. Use the CWT auto-spot to tune and narrow the passband to exclude other signals. YMMV
As an aside, I have used the TX-ONLY option of CW decode to improve my character and word spacing. I figure if the K3 can understand what I'm saying, most CW ops can too. 73, Brian, K0DTJ On Jun 19, 2013, at 9:45, Sam Morgan <[hidden email]> wrote: > running vs 4.67 | 1.115 | 2.81 | 2.81 | 1.03 > > I remember there was a discussion about improving the K3's cw decoding after some agc upgrade that took place a long time ago. I wonder if what we have now is the 'final' version that we are just going to have to settle for? > > I ask because I am still having a lot of trouble with cw decodes > I run thr 2 and cw 5-40 but decoding is just about useless > I have tried agc slow/fast and off > I have played with rf gain as well as filter widths > (I have 200 & 500hz filters as well as the 2.7 ssb filter installed) > > with my new external svga display for the P3, I was looking forward to doing some lazy cw monitoring on the external monitor > > at best I can copy strong stations like ARRL's code practice runs well, > but trying to decode most other cw qso's is pretty bad. > > what was the older firmware version that did such a good job with decoding? > > also if you do have it working, what setting are you using that do work? > > TIA > > -- > > GB & 73 > K5OAI > Sam Morgan > ______________________________________________________________ > 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 Don Wilhelm-4
Those of us who "head copy" CW find Farnsworth code to be very
aggravating because we are hearing words instead of characters. Military Operators don't like it either because they can't tell where 5 letter code groups start and end. 73, Tom Amateur Radio Operator N5GE ARRL Lifetime Member QCWA Lifetime Member On Wed, 19 Jun 2013 20:24:56 -0400, Don Wilhelm <[hidden email]> wrote: [snip] >While copying W1AW to see what would happen, I noted that the Farnsworth >spacing used at lower speeds did not decode as words - there was a space >between each letter - and yes, that is what the Farnsworth method does, >it sends each character at a higher rate than the set speed, but there >is a delay between characters, which the decoder properly interprets as >a space. [snip] ______________________________________________________________ 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
|
|
|
In reply to this post by Brian Hunt
Well said and a good way to get those words out there by themselves.
I've been doing that and was stunned at how many words I've been running together ;o) 73, Tom Amateur Radio Operator N5GE ARRL Lifetime Member QCWA Lifetime Member On Wed, 19 Jun 2013 17:41:18 -0700, Brian Hunt <[hidden email]> wrote: [snip] >As an aside, I have used the TX-ONLY option of CW decode to improve my character and word spacing. I figure if the K3 can understand what I'm saying, most CW ops can too. 73, > >Brian, K0DTJ [snip] ______________________________________________________________ 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
|
«
Return to Elecraft
|
1 view|%1 views
| Free forum by Nabble | Edit this page |