How to measure frequency response on a scope(3)

How to measure frequency response on a scope

The input signal, from an arbitrary waveform generator, is a 66.7MHz carrier phase modulated by a 2 radian step at the centre of the trace. The arbitrary function generator produces a phase step with a transition time which is within a single sample clock period of the generator yielding a high modulation bandwidth. Normal signal generators have limited modulation bandwidth and may not provide sufficient bandwidth. The input signal is shown in trace C1 in the upper left corner. Trace Z4 is a zoomed view of the phase step in the upper right corner. The TIE or time interval error (instantaneous phase) of this signal is measured by parameters P1 and P2. Parameter P2 has its internal PLL turned on using a nominal bandwidth of 667kHz (a cut-off factor of 100). It is the frequency response of this PLL that is being determined.

Math function F1, below C1, shows the track of the parameter P1. The track is a time-synchronous plot of the parameter value versus time. It shows the cycle-by-cycle change in the TIE measurement. It may be thought of as the instantaneous phase variation of the input signal. You can see the step function due to the phase modulation. The magnitude of the phase step is 4.958 ns, which at a carrier period of 15ns represents the 2 radian step amplitude.

The track of P1 is differentiated in trace F2 displayed below F1. The FFT of the differentiated step, in F3 (lower left) shows a flat frequency response out to 2MHz. This is the spectrum of the input to the PLL.

The right side of the display, starting with the Track of P2 in math trace F4, follows the same steps for the PLL output. F5 is the impulse response and F6 is the spectrum of the output. The PLL within the TIE measurement exhibits a high pass characteristic with an upper -3dB point at approximately 667kHz.

These examples show several methods and variants for measuring the frequency response of a device using an oscilloscope equipped with FFT, differentiation, sparse, averaging and extrema functions (roof/maximum). The step/impulse response method, requiring only the "fast edge" signal source on the scope, can be utilised without the need of an external signal source. The other techniques require an external generator. The table summarises the characteristics of each method.

Table: Summary of input signals used to measure frequency response.

These techniques might save you a lot of time the next time you need to make a quick frequency response measurement.

About the author
Arthur Pini is a technical support specialist and electrical engineer with over 50 years experience in the electronics test and measurement industry. He has supported oscilloscopes, real-time spectrum analysers, frequency synthesisers, digitisers and arbitrary waveform generators for leading manufacturers.