Wayne, Thank you for your detailed answer.
I was following the conversation (or thought I was) that the external reference and transverter lock was another way to get high stability/accuracy at UHF and higher frequencies. I guess the issue then becomes, which is more annoying and/or more effective? Maybe I'm missing something obvious, but wouldn't a .25 (or better) ppm-calibrated TXCO be less expensive than all the other stuff, and take up a lot less room (and weight)? I'm really curious about this (and the RTC question I posed) mainly because I'm working with a supplier right now on a fix for a drifting time-of-day clock issue, and the proposed solution involves a Seiko TXCO oscillator that operates over -40C to +85C (automotive interior) and supposedly gives .5 ppm. 73, -- Dave, N8SBE > -------- Original Message -------- > Subject: Re: [Elecraft] [K3] K3EXREF K144REFLK and K144XV > From: Wayne Burdick <n...@elecraft.com> > Date: Thu, June 09, 2011 2:08 pm > To: "Dave New, N8SBE" <n8...@arrl.net> > Cc: elecraft@mailman.qth.net, "Tim Taylor" <tim-tay...@comcast.net>, > "David Pratt" <da...@g4dmp.fsnet.co.uk> > > > Dave, > > The 49 MHz "1-ppm" oscillator is a TXCO, and it works well (0.5 ppm > from 0 to 50 C). Originally we thought we would not be able to hit > this target without adding firmware-based correction. > > I did implement the firmware correction, but I wasn't happy with it, > because it didn't provide much of an improvement beyond 0.5 ppm with > all oscillators tested. What I found is that I had to create data for > every 1 degree C, rather than every 5 degrees, in order to get to hit > a more aggressive 0.25 ppm target. But the manufacturer of the TCXO > had predicted that 5-degree data points would be sufficient for the > algorithm, and that is what they supply with the oscillators. We could > ask the for 1-degree data, but that would increase the cost (not to > mention create five times as much data). It would also not be of much > help to those who already had 1-ppm TCXOs installed. > > What we could do in a future release is provide a way for customers to > manually generate and use 1-degree data. Call it "Method 4." It seems > a bit onerous, but here's the idea: First, you'd put your K3 in the > refrigerator long enough to get to 0C, or the desired low operating > temp. Next, set the rig up with an accurate frequency counter > connected to the output of the oscillator, and put the cover back on, > or at least part of the way on (the freq counter cable has to come out > at the back or side). Then, as the radio heats up, write down the > oscillator frequencies for each 1-degree point. When it gets as hot as > it's going to get due to ambient temperature, turn on a hot lamp or > hair dryer and aim it at the top cover, about where the oscillator is. > Continue heating and logging data until the desired max operating > temperature is reached. > > You could then enter this info into a revised oscillator calibration > data screen in K3 Utility. I'd have to add the firmware necessary to > allow the oscillator to be updated continuously based on 1-degre data. > > This is a long answer to your question. It begs the further question, > Would anyone bother? > > 73, > Wayne > N6KR > > > On Jun 9, 2011, at 10:36 AM, Dave New, N8SBE wrote: > > > Wayne, > > > > What I guess I don't understand is why the TXCO functionality was > > never > > finished. Seems like that would help a number of accuracy/calibration > > issues, and I bought the .5 ppm option expecting that it would be > > useful. > > > > I'm disappointed in what appears to be a decision to abandon finishing > > that project. > > > > On a related note, it wouldn't hurt to do something to fix the real > > time > > clock, so I don't have to keep syncing it with my laptop to keep it > > reasonably on time. > > > > Thanks, > > > > -- Dave, N8SBE > > > >> -------- Original Message -------- > >> Subject: Re: [Elecraft] [K3] K3EXREF K144REFLK and K144XV > >> From: Wayne Burdick <n...@elecraft.com> > >> Date: Thu, June 09, 2011 1:02 pm > >> To: David Pratt <da...@g4dmp.fsnet.co.uk> > >> Cc: elecraft@mailman.qth.net, Tim Taylor <tim-tay...@comcast.net> > >> > >> > >> I'm on the hook for this. I'll send out test code when it's ready. > >> > >> 73, > >> Wayne > >> N6KR > >> > >> On Jun 9, 2011, at 9:36 AM, David Pratt wrote: > >> > >>> Tim - My understanding is that the K3EXREF does not stabilise the > >>> transverter offset. I believe this is planned in a future K3 > >>> firmware > >>> upgrade, but as far as I am aware this has not yet happened. > >>> > >>> 73 de David G4DMP > >>> > >>> In a recent message, Tim Taylor <tim-tay...@comcast.net> writes > >>>> So right now, the K3EXREF will stabilize the frequency of the basic > >>>> radio; > >>>> however, it will not stabilize the frequency of the K144XV. It is > >>>> my > >>>> understanding that the firmware for the K144XV is "in the que". > >>> -- > >>> + - - - - - - - - - - - - - - - - + > >>> | David M Pratt, Kippax, Leeds. | > >>> | Website: http://www.g4dmp.co.uk | > >>> + - - - - - - - - - - - - - - - - + > >>> > >>> > >>> > >>> ______________________________________________________________ > >>> 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 > >> > >> ______________________________________________________________ > >> 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 ______________________________________________________________ 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
