K3: killing RX key clicks
Locked
12
12
 
|
My take is that in the extreme cw is an AM signal modulated by a pure square wave and as such prduces infinite odd order harmonics which will appear as sidebands/clicks. Filtering reduces the extent of the sidebands and smooths the square wave shape particularly rise and fall times but you will still have sidebands equivalent to 7th or 9th harmonics..? This would represent the ideal and you would have to be very close to the signal to hear the sideband energy - a bit of work on the calculator will tell you where the 7th or 9th harmonics would appear for a string of dits depending on the keying speed etc.
But I'm convinced that most bad clicks are actually a result of spurious switching spikes at the 'corners', often at the trailing edge and can produce sideband energy *way* out from the fundamental. This is nothing to do with the inherent physics of the cw signal and as such is curable. 73, Stewart Rolfe, GW0ETF
|
Re: K3: killing RX key clicks
|
|
In reply to this post by Jan Erik Holm
> Are you sure about 2.5 ms?
Yes. > As far as I remember I measured my K3 > to about 5 ms and Elecraft also stated 5 ms. This was over a year > ago... Measure it again. The rise time changed roughly around the F/W 3.00 revision. I went back to a 2.xx version and it measured ~ 6 msec. Paul, W9AC ______________________________________________________________ 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 |
|
|
Paul Christensen wrote:
>> Are you sure about 2.5 ms? > > Yes. > >> As far as I remember I measured my K3 >> to about 5 ms and Elecraft also stated 5 ms. This was over a year >> ago... > > Measure it again. The rise time changed roughly around the F/W 3.00 > revision. I went back to a 2.xx version and it measured ~ 6 msec. > > Paul, W9AC > 3.00 I measured. / SM2EKM ______________________________________________________________ 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 Wed, 02 Dec 2009 16:40:20 +0100, Jan Erik Holm wrote:
>Ok Elecraft changed it, something I missed. One of the (many) benefits of Elecraft's use of DSP is the ability to shape keying in the more sophisticated ways that W4TV has described. This allows clicks to be minimized without excessively softening the CW. 73, Jim K9YC ______________________________________________________________ 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 P.B. Christensen
Paul Christensen wrote:
>> The 'RC rise/decay' wave shape that was in the handbooks for many years >> is actually a *bad* shape because it has a very sharp corner on key-up. > >Ian, *bad* may be a bit too harsh. The League's optimized envelope was >described during a time when only simple R/C values were used to >develop a keyed envelope. I'm not sure how one would have produced an >economical Blackman-Harris or raised cosine function until say...the >early to mid '90s. How would you have done it? Probably the only >solution at the time was to set leading and trailing edges produced by >the R/C network so soft that they're painful to copy. > available technology, it was wrongly promoted as being the optimum that designers should aim for. Given better information, I believe designers could have produced analog circuits to approximate the true optimum shape... but that didn't happen. -- 73 from Ian GM3SEK http://www.ifwtech.co.uk/g3sek ______________________________________________________________ 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: killing RX key clicks
|
|
In reply to this post by Steve Ellington
Steve Ellington wrote: > > While on 160m a few nights ago I heard loud key clicks and tracked it down > to a ham just 1 mile away who was chasing a DX station. When he > transmitted, > my K3 was rendered totally deft from any signal within 10 Khz of him! My > 200hz roofing filter, ATTN, preamp off and RFG down had no effect. I guess > even the K3 has it's limits. > Steve > N4LQ > [hidden email] > ----- Original Message ----- > From: "David Gilbert" <[hidden email]> > To: "Ron D'Eau Claire" <[hidden email]> > Cc: "'Elecraft Reflector'" <[hidden email]> > Sent: Tuesday, December 01, 2009 1:29 PM > Subject: Re: [Elecraft] K3: killing RX key clicks > > >> >> I'll accept all of that, but it still doesn't change the fact that there >> are some rigs out there that generate bad key clicks for no reason other >> than the fact that their users either don't realize it or don't care >> enough to fix them. >> >> When I first received my Icom 756Pro (now my backup rig) several years >> ago I discovered that the default rise/fall times (adjustable in one of >> the menus) was set to 2 msec (!). That's unconscionable, but I'll bet >> the majority of Icom users never bothered to check it. Many of the >> Yaesu rigs generate horrible key clicks unless their users have >> performed a simple hardware modification on them (see the info at >> W8JI.com). >> >> I operated about 36 hours in the contest this last weekend and made over >> a thousand contacts using the 8-pole 250Hz roofing filter. Many times >> I'd be running a frequency within 200 Hz of a S9+30db station without >> even knowing he was there. Other times I'd hear key clicks (loud enough >> to cover the dits in callsigns I was trying to copy) from stations I >> couldn't even find while tuning with the subreceiver! One large M/M >> operation had great sounding signals on all bands except 15m, where the >> key clicks were objectionable two KHz away even when their signal was >> S5-S7. At least in that case they have promised to find the problem and >> fix it. >> >> Actually, there is a third reason why some stations have bad key >> clicks. They admit they do it on purpose in a contest because it gives >> them more elbow room. >> >> In my experience with the K3 in several major contests since I bought it >> last January, the very great majority of bad signals I've heard on the >> air have been due to the TX on the other end, not the RX on my end. >> >> 73, >> Dave AB7E >> >> >> >> Ron D'Eau Claire wrote: >>> Key clicks can be generated by several different mechanisms within the >>> receiver itself. Paradoxically, it's easier to produce false clicks with >>> a >>> high performance receiver than it is with a lesser receiver. >>> >>> These have nothing to do with the transmitted signal. >>> >>> The first line of "defense" is to turn off the Preamp and turn on the >>> Attenuator to reduce the overall strength of the signals. Next is to set >>> the >>> AGC to "Slow" to ensure a strong signal isn't within the roofing filter >>> bandpass and triggering the AGC while the variable DSP filter is set >>> narrow >>> so the beat note isn't heard. In some cases using the RF gain to control >>> the >>> level helps a great deal. Another approach is to use a narrower roofing >>> filter or to shift your bandpass so the edge of the roofing filter is >>> very >>> close to the desired signal on the side toward the interfering station >>> and >>> as far from the interfering station as possible. Shifting the bandpass >>> may >>> require you open up your DSP bandwidth to continue to hear the desired >>> signal or shifting the DSP filter position within the roofing filter >>> bandpass. >>> >>> And, remember, "clicks" are an essential component of a CW signal. They >>> are >>> the sidebands created by modulating (keying) the signal. Very careful >>> shaping of the keying envelope, such as used in the K2 and K3, can >>> minimize >>> clicks while maintaining an easily readable signal, but the only way to >>> completely eliminate clicks is to not key the signal at all. With a >>> really >>> high-performance you can snuggle up very close to a signal, as many here >>> have noted. When you do that, you're much more likely to hear the >>> essential >>> clicks required for good keying. >>> >>> Ron AC7AC >>> >>> >>> ------------------------------------------------------------------------ >>> >>> ______________________________________________________________ >>> 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 > > > -------------------------------------------------------------------------------- > > > > No virus found in this incoming message. > Checked by AVG - www.avg.com > Version: 9.0.709 / Virus Database: 270.14.88/2538 - Release Date: 12/01/09 > 02:59:00 > > ______________________________________________________________ > 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 > > -- View this message in context: http://n2.nabble.com/K3-killing-RX-key-clicks-tp4093607p4116874.html Sent from the [K3] 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 |
Re: K3: killing RX key clicks
|
|
In reply to this post by Steve Ellington
Because of illness, I've been unable to complete my 160 transmit antenna in time for the contest so I decided to experiment with the k3 receive on 160. A well known local contester who shall be nameless parked himself on one frequency for the entire first seven hours of the contest and supplied a stable, less than clean strong signal,. Another contester on the east coast parked himself 300 Hz higher and supplied a stable weak signal for the other signal.
I tested variations of four things: enabling and disabling my 250 Hz roofing filter adjusting the bandwith adjusting the noise blanker to suppress the strong local station's key clicks adjusting the Noise Reduction switching between the soft agc and the default harware agc. A friend who works 160 is considering a k3 and asked whether I though the k3 could be used with just the ssb filter and what the difference was. Mostly, I used the 200 Hz bandwidth setting which is one that I've come to prefer as a default cw setting but i also used narrower settings. For some of this experiment, I placed the filter peak between the two test signals, moving the weak signal closer to the filter skirt, to make it a bit more challenging. At various times, I used the attenuator or the preamp and also the rf gain control. All of these were useful controls in optimizing reception. With the conditions above, I found: No discernable blocking occurred with the 250 Hz filter enabled whereas substantial, obvious, blocking gain reduction occurred without it along with extraneous crud. The soft agc was indeed an substantial improvement over the default hardware agc in allowing weak signals to be copied through the clicks. I settled on AGC F as the best setting. I found that, at least with this particular set of test signals, using the blanker with the right settings could reduce the clicks significantly - from S5 on the meter to S1. At best, the weak signal peaked S2. Interestingly, I could copy the weak signal through the clicks without the blanker enabled but not some of his callers. In any case, using the blanker to suppress the clicks was much more pleasant and probably would help in a different combination of signals. Using the Noise reduction made the reception more pleasant although for this particular pair of test signals, it didn't affect the ability to copy the weak signals. By experimenting without being distracted by actually transmitting, etc., I thiink I learned some useful things about using the K3. I found that with the best control settings, I could have operated happily within 100 Hz of the strong, local. Wayne and crew, I'm impressed. You've supplied a good set of useful, flexible tools for reception under adverse conditions. 73, Dunc, W5DC, Boulder, CO
|
«
Return to [K3]
|
1 view|%1 views
| Free forum by Nabble | Edit this page |