I was reading W8JI's article http://www.w8ji.com/what_causes_clicks.htm but
see that the advice in it (use a narrower filter than an SSB filter for transmitting CW) conflicts with the K3 manual, which says: "Rotate VFO A to select a CW transmit filter (2.7 or 2.8 kHz). Note: Key clicks may result if a narrower filter is selected for CW transmit." Of course I will take the K3 manual's advice, but I am wondering why a narrow filter can cause key clicks? 73 de Graham G3ZOD FISTS #8385 http://www.fists.co.uk ______________________________________________________________ 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 |
Graham,
Tom W8JI was dealing with analog transceivers when that was written. The K3 generates the signal digitally in the DSP portion of the transceiver, so many of those old "rules of thumb" for analog radios do not apply to the K3. As Guy has said many times on this reflector, "This is not your grandaddy's analog radio". A properly operating K3 does not have key clicks and its transmit IMD is very good. 73, Don W3FPR On 1/17/2011 1:09 PM, Graham Smith wrote: > I was reading W8JI's article http://www.w8ji.com/what_causes_clicks.htm but > see that the advice in it (use a narrower filter than an SSB filter for > transmitting CW) conflicts with the K3 manual, which says: > "Rotate VFO A to select a CW transmit filter (2.7 or 2.8 kHz). Note: Key > clicks may result if a narrower filter is selected for CW transmit." > Of course I will take the K3 manual's advice, but I am wondering why a > narrow filter can cause key clicks? > > 73 de Graham G3ZOD FISTS #8385 > http://www.fists.co.uk > > ______________________________________________________________ > 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 Graham Smith-2
> I am wondering why a narrow filter can cause key clicks?
One cause is related to the placement of the CW frequency in relation to the filter passband. If skewed to the edge of a sharp filter, a severely distorted waveform will develop. A CW waveform is a quasi square wave and replication accuracy is dependent on a linear-phase response. This topic was recently covered: http://mailman.qth.net/pipermail/elecraft/2011-January/143163.html 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 |
Probably the biggest variant in how to think about this is that in an analog
transmitter you start with a badly flawed signal which needs to be cleaned up. The filter attenuates sideband energy (clicks) from a waveform which was already distorted at the point where the RF was keyed. In the K3 and any other digital radio using this scheme, a "perfect" keying form is created in firmware by numeric means, and the TX digital to analog converter creates a pristine CW signal from that stream of numbers, which can only be messed up, not improved. Therefore the skirts of the filter are set away from the CW, using the SSB TX filter, to prevent any distortion of what STARTED OUT clean. This numeric waveform has the advantage of being able to have its "purity" improved in firmware. In the analog radio this is a design and manufacturing change. W8JI's celebrated fix for the ghastly clicks in the FT1000MP changed keying constants and delays in each INDIVIDUAL MP across a string of circuits so that what emerged had as little clicks as possible. It was not so much a design change as a value change that affected the TX string switching waveforms, and it varied by rig because the values changed were set in the MP with plus and minus fifty percent cheep components. The K2's elegant click fix in early years addressed a single keying waveform problem in a single stage. But neither the JI-fixed MP or the current K2, however good as they are, can compete with the digital solution. 73, Guy. On Mon, Jan 17, 2011 at 1:39 PM, Paul Christensen <[hidden email]> wrote: > > I am wondering why a narrow filter can cause key clicks? > > One cause is related to the placement of the CW frequency in relation to > the > filter passband. If skewed to the edge of a sharp filter, a severely > distorted waveform will develop. A CW waveform is a quasi square wave and > replication accuracy is dependent on a linear-phase response. > > This topic was recently covered: > > http://mailman.qth.net/pipermail/elecraft/2011-January/143163.html > > 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 > 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 |
Free forum by Nabble | Edit this page |