PC-Based System Offers Practical Approach to Disk Drive Testing

Personal computer (PC) applications are proliferating at an astonishing rate. To accommodate these applications, many popular PC software programs share a common appetite for large volumes of disk space and fast access speeds.

To meet these demands, the areal density of disk drives is growing at a rate of 60% per year. Access speeds are also improving at a steady clip. Much of this explosive growth in performance is attributed to improvements in technology for rigid disk drive heads.

Several factors affect the performance of head assemblies for high-capacity disk drives, but the primary concerns are:

The flying-height characteristics of disk drive heads.

The role played by the resonance of the structural support (or flexture) in determining the flying characteristics.

Briefly put, the requirement for higher tracking density has made head resonance in the track-to-track (radial) axis a critical consideration. The measurement of the radial movement (resonance) of head-suspension systems poses many challenges for test equipment and methodology.

Disk Drive Performance

Standard computer disk drives store data magnetically. The magnetic read/write head moves over the magnetically coated surface of a disk.

The head must be very close to the surface to “write” a series of tiny magnetic fields (bits) on the thin magnetic coating of the disk. The combination of moving the recording head and rotating the disk enables reliable, cost-efficient storage of large volumes of information and high-speed random access to all data stored.

Today, many disk drives spin at speeds >7,000 rpm, more than 200 times the speed of the bygone 33-rpm record. The recording head flies over the surface of the disk at heights measured in microinches. Head assemblies for high-capacity disk drives fly at less than 2 microinches above the surface, a distance that equals about 1/1,500th the diameter of the typical human hair. Some heads require suspension systems that descend below the microinch level, leading to pseudo-contact recording.

Tracking Errors

Given the high speeds and low tolerances involved, reliable performance depends on precise control over the disk drive surface and the flying characteristics of the recording head. To ensure precise flatness of the surface, manufacturers measure and smooth the disk with special burnish and glide equipment.

To detect and quantify defects in the magnetic surface of the disk, media-certification equipment is used. The flying height of the head is measured with polychromatic interferometric™-based testers. To determine the resonance characteristics of the recording head and suspension assembly, called the head gimbal assembly or HGA, the industry uses HGA resonance testers.

Resonance testing focuses on the suspension assembly that holds the recording element, called a slider, in place. Precise understanding of suspension resonance is critical because the tracking performance of the head is controlled by algorithms that have very limited control in the high-frequency range of mechanical resonances.

The track density of disk drives has increased to the point that microinch variations in the radial (side-to-side) resonance of the suspension may cause the head to mistakenly write data on previously written tracks. Such errors are catastrophic when data is destroyed. At the very least, they degrade read/write performance.

Two small components are of particular interest in resonance testing: the slider and the flexture. The slider is the element containing the electromagnetic transducer or head. The flexture holds the slider in position at the end of the drive arm. Figure 1 shows the components of the HGA.

Less than an inch in length, the flexture has become the target of an extensive testing effort. One example of the equipment necessary to test resonances in flextures and HGAs is the Phase Metrics HRT-1 Resonance Test System. This system was initially developed to meet the production test requirements of Hutchinson Technology, a manufacturer of flextures for HGAs.

HRT-1 Resonance Tester

The HRT-1 includes:

A PC for overall system control.

A laser printer for hard copies of measurements and plots.

A shaker table for vibration input.

A laser Doppler vibrometer that measures light coming off the slider component of the HGA.

A color video camera that provides an on-screen image for laser alignment.

A DSP Technology SigLab 20-22 Signal Analysis System that excites the HGA over a precise frequency range and employs digital signal processing hardware and software to calculate the HGA’s mechanical transfer function.

Theory of Operation

The HRT-1 detects and measures mechanical resonance peaks in HGAs. This is accomplished by calculating the mechanical transfer function of the HGA. To begin the process, the HGA-under-test is attached to the HGA mounting block using a small screw hole through the HGA baseplate. The mounting block is then attached to an adaptor block on the diaphragm of the shaker. The shaker excites the HGA in a forward and backward (radial) direction.

An accelerometer embedded in the mounting block measures the acceleration of the HGA baseplate. The resulting velocity of the slider is measured with a laser Doppler vibrometer. Figure 2 shows a block diagram of the HRT-1.

The next step is to obtain the transfer function, relating the acceleration of the HGA baseplate to the velocity of the slider as a function of frequency. The procedure involves dividing the ratio of the cross-power spectrum between the velocity and acceleration signals by the auto-power spectrum of the acceleration signal.

The cross- and auto-power spectra are determined by real-time digital filtering, windowing and the FFT algorithm. This ultimately produces both a magnitude plot and a plot of phase vs frequency of the mechanical transfer function.

PC-Based Signal Analysis System

In the HRT-1, the SigLab 20-22 Signal Analysis System drives the shaker over a user-specified frequency range and obtains the transfer function relating the baseplate acceleration and the slider velocity. The two-channel system takes measurements directly into MatLab, an environment for numerical and computational analysis.

Test Setup

To acquire test measurements, you usually adjust instrument knobs and switches to proper levels or write specific lines of code. With SigLab, you configure tests in software from the PC using an application called Virtual Network Analyzer. Figure 3 shows a screen capture of a test setup with Virtual Network Analyzer software.

A graphical user interface allows you to adjust test parameters to view the most accurate information in the least amount of time. Once test setups are defined, you have a simple means of running a wider variety of complex resonance tests, which can be easily invoked and modified from the controlling PC.

On the Factory Floor

The system is now performing—in a production environment—the kind of measurements that were previously obtainable only with specialized instruments in a laboratory setting. Progress in PC-based analysis tools allows you to perform higher levels of resonance testing out on the factory floor. The capability to analyze track-to-track mechanical resonance in HGAs is one example of the many new test technologies combining to enable data storage capacities to continue climbing at rates of 60% per year.

About the Authors

Rich Freedland is Director of Marketing for Phase Metrics. Before the merger with Phase Metrics, Mr. Freedland was President of Helios. He holds a B.S.E.E. degree from Syracuse University and a master’s degree in business administration from Golden Gate University. Phase Metrics, 47475 Fremont Blvd., Fremont, CA 94538, (510) 624-5178.

Dick Benson is Product Specialist at DSP Technology. Previously, he was an Engineering Manager at Tektronix. Mr. Benson has an M.S. degree in electrical engineering. DSP Technology, 48500 Kato Rd., Fremont, CA 94538-7385, (510) 657-7555.

August 1996