-----Original Message----- I wouldn't mind a few more words explaining that as it applies to the K2.
Eric KE6US --------------------------- The beat note you hear is determined by the exact local oscillator (l.o.) and beat frequency oscillator (bfo) frequencies. The local oscillator tuning is what you are adjusting by spinning the main tuning knob. The beat frequency oscillator frequency is set when you did the filter alignment. As you select various filter bandwidths, the bfo frequency must change to keep the signal centered within the filter bandpass. These oscillators are tuned by analog voltages that are applied to voltage-variable capacitors in the l.o. and bfo. Changing the voltage changes the oscillator frequency. When you ran CAL PLL you caused the K2 to measure and record the l.o. frequencies across its tuning range and record them in memory. It's called CAL PLL, not CAL L.O., because the local oscillator uses a phase-locked loop (PLL). Recall that you put the test probe in a test point on the RF board before you ran CAL PLL. That test point allowed the logic circuits in the K2 to measure the l.o. frequency. CAL PLL then applied tuning voltages to the local oscillator and measured the frequency produced for various tuning voltages applied. The process created a table of voltages and frequencies in the K2's memory that can be referenced whenever needed to tune the K2 to a specific frequency. That is, if you spin the knob to 14,036.96 kHz, the logic can look up the data and determine what tuning voltage must be applied to the l.o. to tune in that frequency. The same thing happens with the bfo. As you change filter bandwidths, the bfo frequency needed to produce the exact audio tone (or SSB quality) you want changes. So you run CAL FIL with the probe in a test point that measures the bfo frequencies. When you center up each filter bandpass using Spectrogram, you are adjusting the BFO frequency. When you switch away from that setting in CAL FIL, the logic records the voltage needed to produce that bfo frequency again. When you select that filter position in operation, the logic supplies the necessary voltage to put the bfo on the frequency you determined was correct when you ran CAL FIL. The shift in tones you sometimes hear is caused by the accuracy with which these frequencies, and the voltages needed to produce them, are stored and recalled. The analog tuning voltage is measured by the logic and converted into a digital number. That digital number is what is stored in memory. When the tuning voltage is needed again, the logic looks up the proper number, and converts this back into a tuning voltage level needed by either the local oscillator or the beat frequency oscillator. The conversion between the digital number and the analog tuning voltage needed is by logic circuits called Analog-to-Digital Converters or DACs. The accuracy of the conversion is determined by the DACs. Greater accuracy requires longer digits. For example, if you measured a dry cell with a DMM having one digit it might report 1-volt. If your DMM has two digits, it might report 1.2v. If it has three digits, it might report 1.21v and so on. Greater accuracy requires a greater number of digits and a greater number of digits requires more memory to store all the numbers needed. Elecraft chose an accuracy that can allow a small error in the tuning voltage when it is re-created in the interest of avoiding huge demands on memory. Ten or 20 Hz errors are common, although various ops will tell you different ways that you can reduce this to, apparently, zero. People often ask, "Why, if my dial says 14036.96 kHz, doesn't the logic put it exactly on that frequency?" It's important to realize that the frequency you see on the LCD is *not* the measured frequency. It is the *requested* frequency. You select a filter and then spin the dial to, say 14036.96 kHz. The logic circuits look up the digital values for the tuning voltages needed for the l.o. and bfo to tune in the K2 to that frequency using the filter you selected. These digital values go to the DACs who then produce analog voltages that are sent to the oscillators. The accuracy of the dial, therefore, is determined by the accuracy of the DACs and the stability of the oscillators since you ran CAL FIL or CAL PLL. Another point of confusion often comes up when ops try to "tweak" their bfo frequencies to compensate for this error while listening to a signal. The bfo and l.o. frequencies interact in a way that makes the bandpass appear to shift instead of the beat frequency change. That's why it's necessary to do CAL FIL using one of the techniques described in the manual or on the Elecraft web site. If you're interested in the nitty-gritty behind the clever way the l.o. and bfo frequencies interact, there's a paper on the Elecraft web site describing this process called "The Mechanics Behind CAL FIL". Ron AC7AC _______________________________________________ Elecraft mailing list Post to: Elecraft@mailman.qth.net You must be a subscriber to post to the list. Subscriber Info (Addr. Change, sub, unsub etc.): http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/subscribers.htm Elecraft web page: http://www.elecraft.com
