need just the error processor? see T-Series Embedded

Since its introduction in 2003, the installed base of the T-Series has become the industry's workhorse of magnetic certification. Its speed, sensitivity, and accuracy outperform other systems (legacy or new) by a large margin. For a high sensitivity test (85% threshold) the T-Series will detect previously unseen errors (often surpasses a drive-level scan) giving the user critical process control information. The T Series Certifier is built upon the footprint of your legacy tester. Upon the trade-in of your old tester, you will get the world's FASTEST and most SENSITIVE Magnetic Certifier in return. This is an ideal upgrade for your factory to extend or revive the life of your obsolete certifiers. The package is an essential dump of the older electronics and control system; about the only components that survive the legacy to T-Series rebuilt/upgrade are the motion control (spindle and stages). If your magnetic certification budget is 1 minute: 5000 track magnetic certification (0.2 mil step) 200 kfci 65 MHz (75, 100 versions available) 85% threshold (yes, not a typo), changing threshold would NOT affect test time. With 30 more seconds, you can increase the density to 300 kfci. Or would you rather keep the same density and have a 10,000 track test in 2 minutes? Just imagine the wealth of process control possibilities with a 0.1 mil track pitch at 85% threshold? And still get 5000 disks/day/workcell ! * If you are ready to move your old tester to the 21st century, this upgrade is for you. And the upgrade will take about 30 minutes ** (reversible). * cert only ** Some pre-requisite work may be needed depending on robotics protocol.

Key Features

Industry's Best Sensitivity with Real Time Error Processing

Testing is done during spiral. No more stop-on-error and its associated problems, most notably random-error-blindness. This will allow testing speed to reach the theoretical limit of spiral. The fast Error Processing Unit (EPU) can process the signal with the threshold set at any level - this is about 10,000,000 times faster the EPU it replaces. Whether the test yielded only 1 error or 1 Billion errors, the test time is the same.

This will allow setting the tester at its most sensitive detection ability without the lost of throughput. With the base upgrade, a MG250A can comfortably run at 65MHz, 300 kfci, 85% threshold, at the unprecedented 1 rev per test track throughput rate. The T-Series7 and T-Series10 further improve the test frequency.

The proven T-Series has dramatically changed the magnetic test culture for its installed base. Before the T-Series, typical production is 3K tracks at 65% threshold; and the "audit" cells were running at 5K tracks and 70% threshold. With the T-Series, production is running at 10K tracks, 85% threshold; and the audit is done at 30K tracks. Since test time is always constant, the improvement in precision and throughput are also predictable, further improve product flow.

Windows XP Operating System for Control PC

The advantages of a modern operating system are numerous. Besides a modern code base (C++), the system can now employ: networking, web-enablement, centralized data and test parameter control, real-time data reporting. Also with a potential larger error dataset, other grading schemes can be implemented; mainly pattern grading (scratch, patch), and lot based grading (grade for a group of disks). The Control PC is actually an AMD Athlon64 based system (laptop).

Accuracy: Superior Repeatability and Reproducibility (R&R)

Comparator  threshold accuracy of better than 0.1 %, coupled with a Defect Avalanche test used to compensate for hot/cold heads as well as head rejection; this gives the T-Series superior accuracy over other magnetic certifiers on the market.

Green: 300 Watt Reduction in Power Consumption

Immediately after the upgrade (T7 and T10), the system will consume at least 300W less. Coupled with air conditioning (COP:3), this is approximately a 1.2KW saving per tester. A 50 workcell test floor will have a continuous saving of ~ 0.5 MegaWatt. Depending on the local energy cost, this is approaching a $1.0 Million in electricity annually (or about 2,500 metric tons of CO2, or about 350 Camrys).

Continuous Improvement

At this time, the code base is very dynamic. We are responding to the customers' wish-lists quickly to add new software and hardware features. Previously, adding new (hardware of software) features were severely limited by the DOS based environment. Some of the new abilities include:

• Faster, more intelligent alignment; such as a multi-track alignment with zero-skew check.
• Centralized (via network) real-time monitor of head performance/degradation (maintenance)
• User definable spiral scan direction with disk erase routine.
• Bi-Directional spiral scan. Test on both directions (ID->OD, OD->ID) to further optimize throughput
• Improved electronics such as higher frequency certification and digital glide.
• Customized Pattern Grading schemes.

The improvements are not limited to the local tester. Advanced server based modules can be added for real-time process control needs. Since the test time now is constant, robotics motion can be optimized since the robot knows ahead of time when the test will end. New possibilities are only emerging.

Specifications

Control PC with TestXP Control Software

• Athlon64 or Turion64 Laptop with mounting hardware (mount in place of existing keyboard/LCD screen)
• Windows XP Professional operating system
• USB 2.0 interface for system control
• Fast Ethernet (100 Mb/s) Networking
• TestXP Control software
  • Count based grading and Pattern grading modules.
  • Fast Dual track alignment (zero skew verification)
  • Single direction spiral scan with optional erase scan, or Bi-directional spiral scans.
• TestXPMap Error Mapping software
  • Fast 2D and 3D mapping of errors
  • OpenGL accelerated (typical: 10,000 track with 1 million error data set is plotted in less than 1 second).

Real-time Error Channel

• Filter Equalizer with 50, 65, 75, 100 MHz modules (other frequencies available)
• Programmable Frequency Synthesizer (50 to 100 MHz at 0.1 MHz step)
• Threshold Accuracy of 0.1% (internal resolution of 0.01%)
• USB 2.0 Interface to Control PC
• 50 pin Interface to MG Bus (motion control)
• RS-232 Interface to Robotics
• Top and bottom Error Channels, for each head:
  • Two simultaneous Missing Pulse Thresholds (50 to 99% at 0.1% step) and (simultaneously) one dual-comparator  (super + missing) Thermal Asperity Threshold.
  • Three programmable error categories for MP: Correctable, Uncorrectable, Huge. The user can set the bit length and boundary length of each category.
  • Two programmable error categories for TA: Correctable, Uncorrectable. The user can set the bit length and boundary length of each category.
  • 200 MFlux/s data processing rate (FIFOless design).
  • 256,000 maximum categorized errors per track for each error type (i.e: correctable,  uncorrectable, huge) per threshold. This equals to a maximum 2.0 M errors maximum per track for all categorized error types.
  • Maximum errors per surface depends of total tracks tested. (i.e a 10,000 track test has a maximum of 2.56 Billion errors for each error type).

Optional Packages

• Digital Glide Channel
  • AE/PZT Pre-Amplifier
  • Amplifier with programmable Filter-Equalizer
  • Full Bandwidth Digitizer (5 MS/s)
  • Programmable Digital filter with multi-band hit/event detection.
• TestNet Centralized Web-based Control (from any web-browser)
  • Networking Hardware
  • HTML and SQL servers
  • Software Modules for Parameters control
  • Software Modules for Process control

Features Comparison

Feature	MG250	TSeries	TSeries7	TSeries10,12,15
Frequency (MHz)	50 (max usable ~35)	65	75	100,125,150
PMR Capable	no	yes
Defect Avalanche Test	no	yes
Test Head Compensation/Rejection	no	yes
kfci	200	300	300+	Contact us for more information
Threshold Accuracy/Resolution (%)	1.5/1	0.1/0.01	0.1/0.01
1 minute throughput, low threshold (tracks)	~1K to 4K *	5K	5K
1 minute throughput, high threshold (tracks)	~100 **	5K	5K
Simultaneous thresholds (MP/TA)	1	2/1	2/1
Maximum Usable Threshold (%)	~70	> 90	> 90
Maximum Errors per Track	512	256,000 (per type) 2 Million for all types	256,000 (per type) 2 Million for all types
Throughput depends on threshold	yes	no	no
Operating System	DOS 6	WinXP	WinXP
Networking	limited to none	yes	yes
Web-based Parameter Control	no	yes***	yes***
Dual Comparator Thermal Asperity Channel	no	yes***	yes***
Digital Glide with Multi-Band detection	no	yes***	yes***
Typical Daily throughput (disks @ 2K tracks, 65%, 200 kfci)	500	1000♪	1000♪
Typical Daily throughput (disks @ 10K tracks, 85%, 200 kfci)	<50	600	700

* 50MHz, 200 kfci, 65% threshold; actual throughput depends on errors on disk
** 50MHz, 200 kfci, 85%; actual throughput depends on errors on disk
*** Optional
♪ speed limited by robotics