Browse Prior Art Database

Method and System for reducing Frequency Sweep Test Cycle Time using a Sampling Algorithm

IP.com Disclosure Number: IPCOM000197613D
Publication Date: 2010-Jul-16
Document File: 3 page(s) / 40K

Publishing Venue

The IP.com Prior Art Database

Abstract

A method and system for reducing Frequency Sweep Test Cycle Time using a sampling algorithm is disclosed.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 53% of the total text.

Page 1 of 3

Method and System for reducing Frequency Sweep Test Cycle Time using a Sampling Algorithm

Disclosed is a method and system for reducing Frequency Sweep Test Cycle Time using a sampling algorithm.

In the manufacturing process of a Tape Head Actuator, a swept sine frequency response test is performed on the actuator to calculate the natural frequency and resonant rise of the device. The test involves repeatedly energizing the Device Under Test (DUT) with a sine wave at a constant amplitude over a range of different frequencies, capturing the physical response signal of the DUT for each iteration in order to determine the natural frequency and resonant rise of the DUT. In addition to calculating the natural frequency and resonant rise of the DUT, the test also checks for a specific behavior over a broad range of frequencies (e.g. 10 Hz - 3000 Hz). For a typical swept sine test, a standard logarithmic sweep is used to determine the sampled frequencies. However, using equally spaced intervals between the start and stop frequencies along a log scale with enough frequency samples to prevent under sampling at higher, critical frequencies results in excessive over-sampling at very low frequencies. Increasing the frequency sample size reduces the under sampling problem, but results in a longer test time.

The method disclosed herein involves selecting frequency samples for a frequency sweep response test using a modified logarithmic scale. Instead of using an equally log-spaced delta to determine the frequency samples, there are three distinct frequency windows. First, the range of frequencies between the start frequency to the minimum critical frequency. Second the range of frequencies from the minimum critical frequency to the maximum critical frequency. Third, the range of frequencies between the maximum critical frequency to the stop frequency. Within each range a delta is calculated (end frequency - start frequency / remaining number of samples). In addition, for all three ranges the algorithm checks whether the difference between each adjacent frequency is greater than a defined minimum difference. Until the minimum difference requirement is achieved, the algorithm re-calculates the current frequency by continuously adding the log-based delta to the sample calculation. The algorithm is given below:

User-defined constant parameters

// start frequency
double startFrequency = 10;

//

  end frequency
double stopFrequency = 3000;

//

  number of desired frequency samples int numberOfSamples = 75;

//

//

      minimum difference required between samples, else a new sample is calculated

1

Page 2 of 3

double minimumDifference = 1.5;

//

      defining a critical frequency range allows more refinement and control of frequency samples

//

  start of critical frequency range double criticalFrequencyMin=50;

//

  end of critical frequency range double criticalFrequencyMax=150;

//

  final array of frequencies to sample double testFrequencies[75

];

//

  Initialize sampling paramet...