Noise spectrum, Noise histogram analysis – Teledyne LeCroy SDA III-CompleteLinQ User Manual

Page 54

Advertising
background image

SDAIII-CompleteLinQ Software

Noise Spectrum

On the Noise Measurement dialog, touch the Noise Spectrum button to display the Noise Spectrum
dialog at the far right of the screen.

Show Rn+BUnSpectrum - Checking this box displays the Rn BUn Spectrum. This plot displays the
frequency spectrum of the signal with the data-dependent effects removed.

Rn+BUn Peak Threshold - Checking this box displays the threshold for the RnBUnSpectrum. Peaks higher
than this threshold are considered Pj and the remaining signal under the threshold is integrated to form
the Rj of the spectral method of jitter composition.

Show Pn Inv. FFT plot - Checking this box displays the Pn Inverse FFT. This plot is the inverse FFT of only
the points of the RnBUnSpectrum above the threshold. This allows you to view the peaks in the time
domain and is very useful for viewing time domain Pn effects.

Show Peaks - Checking this box displays the peaks of the RnBUnSpectrum. This annotates the points of
the RnBUnSpectrum above the threshold.

Noise Histogram Analysis

The RnBUn histogram can be viewed along with the Q-fit plot. RnBunHist is a histogram of the data in
the RnBUn track, and shows the distribution of the noise that is super-imposed on the data pattern.

Zoom the plot by touching the trace descriptor, and then selecting the zoom tab on the right side dialog.

Show Rn+BUn Histogram - Displays Rn+BUn Histogram. Vertical axis is in number of edges in a particular
jitter bin. Horizontal axis is the RnBUn jitter value. Scale is linear. Sometimes it is useful to use the log10
math function to get a log vertical scale to make it easier to view the tails.

Show Q-Fit for RnBUn- Plots the histogram in the Q-Scale representation. In the Q-Scale representation,
Gaussian tails result in straight lines whose slope is equal to 1/Rn. Teledyne LeCroy uses a special Q-scale
which we call nQ-Scale or normalized Q-Scale. This allows for Gaussian distributions with variable

52

921143 Rev A

Advertising