Can i measuring frequency vs voltage curve of VCO with constant voltage points?

[Terminator3]

What if i measure frequency using this way:
- put 1.1v on varactor - measure freq for 1 sec, save frequency value for point 1
- put 1.2v on varactor - measure freq for 1 sec, save frequency value for point 2
- put 1.3v on varactor - measure freq for 1 sec, save frequency value for point 3
- put 1.4v on varactor - .... , save frequency value for point N
Make this once, then using saved data. Can i interpolate and use this data for "fast" modulation - for example, make FMCW voltage curve from this data points to linearize frequency sweep? Why i ask this: frequency measuring time is so long (1 second). But modulation speed is very fast (for example, voltage sweep from 1 to 2v can be less than 10ms). So, is this approach correct?

 

[Borber]

This approach can be called calibrating voltmeter in frequency. This is like in old times when a new voltmeter scale is drawn to show frequency on pointer instrument. Such measurement is not accurate or accurate to few percents. It is more like a frequency indication then a frequency measurement.

 

[Terminator3]

why i ask this, usually fmcw transmitters measure average frequency, for example, using zero cross counter. then adjust voltage curve to linearize. also another approach with pll. i think the first approach with 1ms measuring interval not so precise. 30 times per 30ms. measurements can be stopped at any time and latest voltage curve used. so i cant understand can the same results be obtained with 1second constant frequency measurement points, and using them to produce linearization voltage curve.

 

[biff44]

yes, you can use either an analog linearizer (with breakpoints) or a DAC output that has a distorted output ramp that compensates for the VCO tuning non-linearities.

One thing u MUST do if trying to sweep at 10 mS rate is to make sure the tuning port of the VCO has sufficient bandwidth. You would want the tuning port to have at least a 5 KHz bandwidth to be able to sweep at a 10 mS rate (i.e. 1/10 mS * ~50).