We purchased couple AD737 in order to measure the RMS of a high frequency sine wave (e.g. 500 kHz), but could not make it work. So we would like to see an example of the chip connection measuring the RMS of a sine wave precisely.
Thank you for the providing the RF frequency. Please pardon my unfamiliarity with your acronym but are you referring to a 'Variable Frequency Drive (VFD) as a type of motor controller?
If so, you can use the AD737 but I recommend that you use the AD736 instead. The reason is that the AD736 provides a positive-polarity voltage that is more usable than the negative polarity output of the AD737. The negative output is very confusing to most people, especially if their technical skills are less consistent with analog design concepts.
Jonathan provided you with several very useful responses above, however I sense you were unable to access them.
Therefor I've inserted a schematic of an AD736 eval board below. I've also inserted some application information, however our EZ structure doesn't provide much in the way of a communication resource for in-depth training.
A good way to overcome this would be to find someone local who can help you with the information provided below.
For the type of 20kHz modulated 60Hz. 2Vp-p the rms output should be 0.35VDC. If you don't measure this voltage look for waveform non-sinusoidal elements in the 20kHz carrier. If that is the case, then you can use a high quality DMM as a reference. The error in the AD736 should still be very small.
I hope this helps,