50 IEEE TRANSACTIONS ON MAGNETICS, VOL. 38, NO. 1, JANUARY 2002 Design and Performance of an Inductive Current Probe for Integration Into the Trace Suspension Assembly Anthony B. Kos, Thomas J. Silva, Pavel Kabos, Matthew R. Pufall, Donald C. DeGroot, Larry Webb, and Marc Even Abstract—A low-cost, inductive current probe designed for in- tegration into a disk drive trace suspension assembly is described. The main consideration for this design was to use the same mate- rials currently found in trace suspension assemblies, and thus re- ducing costs, while at the same time providing a drive characteri- zation tool capable of measuring 100 ps head field rise times. The inductive current probe consists of a pair of differential copper conductors fabricated adjacent to the write driver interconnects and magnetically coupled via a Ni–Fe thin film placed on top of these conductors. The differential conductor pair is connected to a high-speed sampling oscilloscope to measure the speed of the write current pulse and thus infer the write head field rise time. Data are shown for high-speed pulses generated with rise times of less than 100 ps. Index Terms—Current probe, head field rise time, inductive voltage, trace suspension assembly, write head current. I. INTRODUCTION I N THE design of high-data-rate disk drives, it is instructive to know the rise time of the write head current pulse. Ex- ternally inserted current probes have been used under labora- tory conditions to measure write head current waveforms and infer field rise times [1]. However, it would be useful to be able to measure actual current waveforms in a commercial drive or prototype device under normal operating conditions using the existing interconnects. We present an inductive current probe as a tool for accomplishing this goal while perturbing the drive sig- nals as little as possible. The costs associated with this method are low, since conven- tional trace suspension materials and manufacturing techniques are used to fabricate the current probe. This is of some advan- tage, since the current probe could even be present on every as- sembly manufactured for use in quality assurance. In addition, since the probe pickup is incorporated between the highly over- driven write drivers (required to provide short head field rise times [2]) and the thin film write head, this is a true in-situ cur- rent rise time measurement. Manuscript received June 25, 2001. A. B. Kos, T. J. Silva, P. Kabos, and M. R. Pufall are with the National Insti- tute of Standards and Technology, Magnetic Technology Division, Boulder, CO 80305 USA (e-mail: kos@ boulder.nist.gov). D. C. DeGroot is with the National Institute of Standards and Technology, Radio Frequency Technology Division, Boulder, CO 80305 USA (e-mail: deg- root@boulder.nist.gov). L. Webb and M. Even are with Hutchinson Technology, Hutchinson, MN 55350 USA (e-mail: larry.webb@hti.htch.com). Publisher Item Identifier S 0018-9464(02)01288-8. Fig. 1. System diagram for inductive current probe. The arrows indicate the direction that the longitudinal bias and transverse saturating fields are applied. To perform an impulse response measurement, the passive impulse-forming network shown is inserted into the output path of the pulse generator where indicated. The inductive current probe (Fig. 1) consists of two sets of differential copper conductors over a dielectric-covered, stain- less steel ground plane, which are magnetically coupled via a Ni–Fe thin film placed upside-down onto the conductors. Since the Ni–Fe film is not actually part of the trace suspension fab- rication process, the process does not need to be modified to include Ni–Fe thin film production. The copper conductors are coated with a layer of polyimide, approximately 1 m thick, to prevent the Ni–Fe film from shorting to the conductors. In an actual drive, the first set of conductors would be the write driver interconnects integrated into the trace suspension assembly. The other set, used for signal pickup, is connected via microwave probes to a high-bandwidth digital sampling oscilloscope. To test the design and performance of the current probe, a com- mercially available, broadband step generator is connected with a single-ended connection via the ground plane to the inner con- ductor on the drive side. The other end is terminated in 50 ohms. This pulse generator is capable of delivering 10 V steps with 50 ps rise times and 10 ns duration, to simulate very fast write current pulses. Each end of the inner conductors on the pickup side is connected, in a single-ended fashion, to the two channels of a dig- ital sampling oscilloscope. This sampling oscilloscope, which has the 14-bit dynamic range necessary to extract millivolt-level signals, averages the waveforms more than 500 times to achieve the required signal-to-noise ratio (SNR). The signals present on the two oscilloscope channels are differentially detected, thus re- moving capacitively coupled signals that are common to both channels. Two measurements, one with the Ni–Fe film saturated and one with the film left free to rotate, are taken and subtracted to yield a result independent of background signals. This sub- traction process is very important to the measurement; it removes 0018-9464/02$17.00 © 2002 IEEE