Of course Elecraft owes us an explanation. Thankfully Eric just attempted to do exactly that, but why on earth would you think otherwise?? The rig distorts signals in normal operation, and it is definitely a problem for many competent contesters even if you don't fall into that category. The list of top scoring contesters and serious DXpeditioners who have stepped forward on this issue is getting pretty large ... AND THEY ALL SAY THE SAME THING. And here's a bit of enlightenment for you ... the guy who started this thread (KE7X) is the guy who writes and sells the most comprehensive manual on the K3.
There are good reasons why simply turning off the AGC is not a good solution when running a pileup, not the least of which is the possibility to blow the output IC as some users have experienced when a strong caller suddenly shows up. Besides, I don't recall anyone saying that turning off the AGC was a definitive fix for the problem, and some have openly said that it isn't. If it WAS a fix, I suspect that Elecraft would be able to address the issue with software and I happen to know that they have already said they haven't been able to. As far as more clearly defining the problem is concerned, here's what is probably needed: a. At least five or six independent high purity signals within a narrow frequency range. I typically use a 250 Hz 8-pole filter with DSP set to 300 Hz when contesting, but as Eric pointed out many callers are almost zero beat with each other. The signals need to be such that sum and difference products overlap with other signals within the same passband. b. The signals need to be injected into the K3 front end with variable attenuation and lots of isolation between them so that the test isn't colored by the possibility of one signal source pulling the others. c. The signals need to be keyed, preferably at different speeds from each other. Constant tones won't cut it. d. Assuming that the audio from the K3 was fed into a high quality computer sound card, it probably would require some piece of software to analyze the recorded wav file to quantify the relative level of the undesired products relative to the pure tone inputs. I don't think simply viewing on a spectrum analyzer would show anything because of the timing dynamics of keying. Now then, how many hams do you think have the necessary resources to recreate those kinds of conditions? At one point I considered using a program like CW Player to record several different CW streams close in frequency, merge them, and then feed them into the mic input of a second rig nearby driving a dummy load. I didn't follow through on it because the was no way I could quantify the possible mixing caused by the second rig before signals ever got to the K3, but does that sound like someone who is simply too lazy to turn off the AGC??? I'm glad to hear that Elecraft recognizes this as a problem and I am anxiously looking forward to whatever they find. For the record, my parameters that I use for CW are: AGC DCY = SOFT AGC HLD = 0.20 AGC PLS = NOR AGC SLP = 000 AGC THR = 008 AGC-F = 140 AGC-S = 030 I mostly use AGC FAST for contests but change as needed. I never use NR or NB at all ... ever. Dave AB7E On 12/4/2011 1:08 PM, Grant Youngman wrote: > What I find most bothersome about these rants, is that somehow certain of the > owner class seems to think that Elecraft OWES them a solution or an > explanation. As though "mush" is enough of a description for a DSP wizard to > have an epiphany and write a few lines of code and fix the problem, which is > ill defined and probably difficult to reproduce in the lab. > > More telling is that it has been pointed out that turning off AGC eliminates > the effects. I don't know about these guys, but turning off AGC when the > issue is recognized -- since they can clearly recognize it -- sounds like a > far more sane and reasonable solution than blaming their contest scores and > apparent 'embarrassment' on 'mush' and trying to get some other radio vendor > to fix whatever problem shows up in THAT radio. > > Isn't it reasonable to assume that a 'top contester' should know how to > utilize his equipment beyond turning it on and twisting the dial? Geez. > > Grant/NQ5T > > Sent from my iPhone > > On Dec 4, 2011, at 1:44 PM, Mike Fatchett W0MU<w...@w0mu.com> wrote: > >> >> I am glad Elecraft is attempting to isolate this with more scientific >> data instead of anecdotal comments. Did anyone bother to record any of >> the mushiness they are describing? Use the line out and an audio >> capture program. I would love to hear what you are hearing. >> >> ______________________________________________________________ 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
