Got it. Thank you. De WA6NHC -----Original Message----- From: Don Wilhelm [mailto:w3...@embarqmail.com] Sent: Monday, December 05, 2011 4:09 PM To: lrom...@ij.net Cc: 'Rick Bates'; 'Elecraft' Subject: Re: [Elecraft] Mush and Manufacturers
Lu, Rick and others interested --- Your question seems to relate to SSB operation - as far as I know, the "mush" problems that have been reported are for CW signals. It may be that many SSB signals of similar levels will create a similar effect, but when one adds the passband, (hi-cut, lo-cut) and the RX EQ settings to the mix, things begin to get very complex fast. Since the reported problem are on the receiving end, I cannot conceive any situation where changing the TX EQ would make any difference in the signal that is being *received* by the K3. This is an important issue to those who are experiencing it (and that is a lot of the DX-peditioners and contesters that the K3 was designed to assist the most), and the parameters of the problem are not yet well understood. So, my request is that everyone read the conditions carefully so that if you can add data to the search, please do that, but if what you are referring to is different than the main focus, please change the subject line so we do not confuse two different things. This is an important topic for Elecraft, and as far as I know, there are no consistent and reproducible parameters to this situation. In my long experience of designing and conducting tests on large system computing systems, if a "problem" could not be described in parameters that make it reproducible, then it is not considered a "problem" because it cannot be solved. OTOH, I once had a manager whose favorite expression was "perception is 90% of reality" - so while this "problem" is not really something that can be defined sufficiently to provide a fix (at this point in time), it is really encouraging that Elecraft is interested and pro-active in attempting to define the problem sufficiently to fix it. Bottom line - if you have perceived the problem, please provide all the relevant information that you can so this problem can be well described, defined, and made reproducible. Until the problem can be reproduced in the lab environment, one will never be able to say whether it is fixed or not. If you can help, please do. 73, Don W3FPR On 12/5/2011 6:35 PM, Luis V. Romero wrote: > Rick: > >> Did you try chopping all the lows and boosting the highs in >> the TX equalizer? It may not be broadcast quality audio >> then, but that provides a lot more 'punch' in the audio, >> which helps in DXing but not so much for the local chat nets >> (then turn on ESSB too). Depending on your voice >> characteristics, you might also want to boost the midrange. >> > Yes, ad infinitum. Its not a frequency response thing, its a modulation > density thing. Since the DSP "Processor" works akin to an RF Clipper (Or so > Im told), input leveling is critical to raise the average power (decrease > the ratio between peaks and valleys PRE-LIMITING) because this particular > audio chain has a somewhat narrow band between clarity and distortion. See > what you sound like if you increase the COMP setting beyond 25... Fuzzy, eh? > With AGC leveling pre-limiter, I dont need to push the COMP beyond 20-22 at > most. Denser, but less fuzzy, as I clip lesser amounts of the audio > waveform. The COMP software doesnt have to hack off large chunks of > overshooting levels to bring up the valleys, because the more you clip, the > more distortion you add (I know its a digital process, but it SOUNDS like > overruning the ADC to me, and digital audio is unforgiving when you > overshoot the ADC threshold). > > AGC helps here to keep the signal in the sweet spot for the COMP algorythms. > There seems to be a little AGC action in the rig's audio chain, but we dont > have a handle for it and every voice and mic combination is different. Also, > a little compression goes a long way in pre-conditioning a signal for later > clipping, you dont want to get carried away by turning the knobs to 11. > Compressing the voice waveform a bit, (I use 2:1) adding AGC to "level" the > waveform then limiting the result of the compression increases modulation > density and keeps the audio peaks and valleys within the DSP "window" for > best performance (IMO, others will undoubtedly disagree!). > > Listen to your local AM Talk broadcaster critically (If you are in a larger > market... Some of the small market guys sound like crap!) to understand the > process. Or listen to a Kenny Chesney song, his engineers really do an > outstanding job of making his voice stand out with both punch and clarity. > Some of the best voice processing I have heard lately is on the Kenny > Chesney song "I Go Back". > > Am I asking a lot from a Ham Rig's audio chain? Probably, but it never > hurts to ask! > ______________________________________________________________ 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
