SFDR Calculation -vs- Measurement

Forum for discussing PicoScope version 6 (non-automotive version)
Post Reply
Posts: 0
Joined: Fri Feb 28, 2014 1:45 am

SFDR Calculation -vs- Measurement

Post by mescalero » Fri Feb 28, 2014 2:14 am

I believe that I am having trouble getting accurate measurements of SFDR. I assume that picoscope calculates SFDR with harmonics, as opposed to without harmonics. In default settings, with a "1 kHz," 0.707V sine wave looped back from the generator to channel A, it measures an average SFDR of 65.06 dBc. I can clearly see that the highest harmonic is ~3 KHz and it is closer to (almost exactly) 100 dB down from the fundamental. Within C weighting, this region is flat and I would expect a(n) SFDR measurement of 100 dBc. Please correct me if I am mistaken on this.

OK, so, I switch the generator to a square wave and the software measures the SFDR average as being 9.158 dBc. Now, on the screen, the peak at ~3 kHz is "exactly" 8 dB down from the fundamental. So, I would expect a measurement of 8 dB.

Why the discrepancy? Is there an error in my assumptions regarding "C" weighting?

PicoScope® 6 - PC Oscilloscope software version:
Copyright © 1995-2014, Pico Technology Ltd

Model: PicoScope 4262
Serial Number: ARxxxxx
USB Version: 2.0
Calibration Date: Tuesday, January 7, 2014
Hardware Version: 1
Driver Version:
Firmware Version: /

Advanced User
Advanced User
Posts: 194
Joined: Thu Nov 21, 2013 9:22 am

Re: SFDR Calculation -vs- Measurement

Post by Karunen » Wed Mar 12, 2014 8:20 am


There may be confused between dBc and dB(C). The former is dB below carrier, and simply means that the SFDR is measured as a ratio of peak spur to carrier (fundamental). The latter refers to a "C" weighting (band-pass filter) function. PS6 does not perform any type of weighting of spectrum data.

Could you please send in a psdata file regarding the Square wave so that we can look?

You can either attach it here or mail it to support@picotech.com


Technical Specialist
Pico Technology

Post Reply