Here, Brad, WB9FIP describes how he measures frequency with a IC-706MKIIG and a tuning fork.
Using a borrowed IC-706MKIIG (getting it ready for Field Day) I let it warm up for four hours, which is a time previously determined to achieve good stability. A half hour prior to the test, receiving WWV on 15 MHz LSB, computer software determined the radio dial error to be +28.55 Hz. From here on no computers, just the calibrated radio, a 440 Hz tuning fork (previously calibrated), and a stop watch. I hope the tuning fork and stop watch do not violate the "no test equipment" rule. All bands were tuned in the LSB mode to keep the math the same. After finding Connie's signal, I quickly tuned the radio for an approximate 440 Hz note (I have pretty good pitch), or you could find zero beat and then tune up 440 Hz on the dial. Tuning a little lower than 440 Hz is desirable to begin the next step. Now with your ear facing the radio speaker and an eye on the radio dial, strike the tuning fork, placing it near the speaker so that you can here a beat note between the two. Tune the radio higher in frequency so as to minimize the beat frequency. This assumes a radio like the IC-706 that tunes in 1 Hz increments. Now tune lower one, or at most two Hz to increase the beat frequency to a rate that can be counted. Use the stop watch to measure the time elapsed in ten beat cycles. Ten (beats) divided by the stop watch time (seconds) gives frequency in Hz to be added to the (corrected) dial readout. This should be good to about 0.2 Hz or better. Finally subtract off the 440 Hz tuning fork frequency (LSB) to give the measured carrier frequency. I hope this is right, or my measurements are going to be way off. If so, ignore the above wasted bandwidth!
Of course this is viable only if you can hear the signal well above the noise. Try as I might, I could not detect the 80 meter signal at all.
73 - Brad WB9FIP