Date: Mon, 05 Mar 2001 00:38:14 -0800 From: Wayne Burdick Subject: Re: [Elecraft] K2 Dial accuracy Hi Mike, The dial accuracy will vary from band to band, and the absolute error will increase towards the high end. The high end of 80 meters is the worst case by far because it represents the largest extreme of PLL reference oscillator movement. On the other bands the error is quite small across the entire ham band. So the short answer to your question is no, your K2 is not unusual. In fact the next revision of the K2 manual goes into this in more detail and will include a revised spec. There are two levels of VCO linearization in the firmware (I can give you a lot more detail if you're interested). Our goal was to achieve smooth tuning, i.e. no gaps or overlaps as you tune from one 5 kHz segment to the next. That we did. But because of firmware space and EEPROM size limitations we had to accept a certain amount of nonlinearity within the low end of 5 kHz segments. It's only noticeable when you use an external counter, not when tuning. We have a scheme in mind to further improve accuracy which may make it into the next firmware revision, now that we're moving to an MCU with more ROM. 73, Wayne N6KR Mike WA8BXN wrote: > > I have noted on 80 M that the displayed frequency is about 200 Hz low near 4 > MHz while it is right on at 3.5 MHz. Similar errors to lesser extent on the > narrower bands. I have checked C22 settings several times and done CAL PLL > with still no improvements. I have used various methods to set C22, > including WWV reception at 10 MHz and a just recalibrated freq counter. > Bottom end no problem, high end not quite right. Am I expecting too much, is > this typical, or is there something wrong with my particular unit? Thanks > for any input on this! ++++++++++++++++++++++++++++++++++++++++++++++++++ Date: Mon, 05 Mar 2001 08:57:56 -0800 From: Wayne Burdick Subject: Re: [Elecraft] K2 Dial accuracy Mike WA8BXN wrote: > > Is there any way to do the calibration in the middle or high end of given > bands? How does the dial frequency when CAL PLL is run affect things? The present algorithm requires the use of the frequencies specified, i.e. the low end of each band. Moving up will not increase accuracy at the high end, and it may cause problems at the low end. Wayne ++++++++++++++++++++++++++++++++++++++++++++++++++