Hi Nathan,
The low frequency method counts an internal timebase of known frequency during one period of your input signal. We can then tell the period of your input signal with a resolution of ±1 period of the internal timebase. This result is inverted to obtain the frequency of the signal.
The reason that the method is called "low frequency" is that it is not suitable for higher frequency signals where 1 tick of the internal timebase would represent a significant percentage of the signal's period.
Depending on your hardware, you would have different timebases available. For example:
E Series: 20 MHz, 100 kHz
M Series: 80 MHz, 20 MHz, 100 kHz
X Sereis: 100 MHz, 20 MHz, 100 kHz
Let's assume an E Series board using the 20 MHz timebase.
Input signal: 220 Hz
Period of input signal: 4.5454545 ... ms (1 / 220 Hz)
Period of E Series Timebase: 50 ns (1 / 20 MHz)
So, for every period of your input signal you will have about 90k timebase ticks (4.5 ms / 50 ns). Since the resolution of the measurement is 1 timebase tick, this gives about 11 ppm of measurement error (1/90k) at frequencies around 220 Hz. For most applications 11 ppm shouldn't be an issue (e.g. 219.9978 Hz instead of 220 Hz). Many newer DAQ cards (M or X Series) have a faster timebase that would give an even lower error.
Given that the accuracy of the timebase of most multifunction DAQ cards is on the order of 50 - 100 ppm, the 11 ppm of measurement resolution will likely be suitable.
For more information, the X Series User Manual has a section describing the accuracy of the different methods starting on page 7-18. The "Sample Clocked" method only applies to newer hardware like X Series, but the other information should relate to most other DAQ cards.
In short, 220 Hz is typically not considered a high enough frequency to warrant the use of one of the other methods. The exception is if you are using a board with a very accurate oscillator (e.g. 6608) and are looking for the most precise measurement possible (e.g. for calibration purposes).
Best Regards,
John Passiak
