The Microset Timer has a multi-turn pot to adjust the TCXO frequency. In this experiment the output frequency of the TCXO was accurately measured about every second while the pot was adjusted from full CCW to full CW in a series of quick one-turn CW adjustments made once a minute. Near the end the pot was turned back CCW until the TCXO was on-frequency (exactly 16 MHz). The total run lasted 30 minutes (1800 seconds).
In the plot, the staircase effect clearly shows the discrete frequency increments that were made once a minute. We can see that the pot has a total of 22 turns. And interestingly we also see the large non-linearity of the TCXO adjustment (due either to the pot or the crystal itself).
Using the same data, the scale for the plot below was changed to relative PPM (parts per million). The range is about -6 to +14 ppm, for a total adjustment of about 20 ppm.
Below is a zoom of the five minutes (5 adjustments) right around where the TCXO went from being set too low to being set too high. The size of the full-turn steps suggests setting the TCXO to an accuracy of 1 ppm is trivial and even 0.1 ppm is not difficult.
Source: first half-hour of log52712.txt, selected seconds ##:##:[1-4][0-9].###
/tvb