Artificial neural network application for magnetic core width prediction and modeling for magnetic disk drive manufacture

• US 8,521,670 B2
• Filed: 05/25/2011
• Issued: 08/27/2013
• Est. Priority Date: 05/25/2011
• Status: Expired due to Fees

• Current Assignee: Western Digital Technologies Incorporated (Western Digital Corporation)
• Original Assignee: HGST Netherlands B.V. (Western Digital Corporation)
• Inventors: Paguio, Hernan J. S.
• Primary Examiner(s): Starks, Wilbert L
• Application Number:
  US13/115,884
• Publication Number:
  US 20120303562A1
• Time in Patent Office: 825 Days
• Field of Search: 706/19, 706/45
• US Class Current: 706/19
• CPC Class Codes: G11B 5/3116 Shaping of layers, poles or gaps for improving the form of the electrical signal transduced, e.g. for shielding, contour effect, equalizing, side flux fringing, cross talk reduction between heads or between heads and information tracks G11B5/3113, G11B5/245 take precedence
• CPC Class Codes: G11B 5/3163 Fabrication methods or processes specially adapted for a particular head structure, e.g. using base layers for electroplating, using functional layers for masking, using energy or particle beams for shaping the structure or modifying the properties of the basic layers

Abstract

A method for predicting and optimizing magnetic core width of a write head using neural networks to analyze manufacturing parameters, and determining new manufacturing parameters that will provide more optimal magnetic core width results. The manufacturing parameters can include: write pole flare point; wrap around shield dimension; and side gap dimension.

Claims

1. A method for manufacturing magnetic heads for magnetic data recording, comprising:

collecting data regarding manufacturing parameters;

constructing a neural network;

analyzing the data using the neural network to produce an output from the neural network; and

using the output from the neural network to optimize a magnetic core width of a magnetic write head.

2. The method as in claim 1, wherein the data includes:

a write pole flare point;

a wrap around shield dimension; and

a side gap.

3. The method as in claim 2 wherein the data includes a magnetic sensor stripe height and a reader/writer offset.

4. The method as in claim 1 further comprising adjusting the manufacturing parameters based on the output from the neural network to optimize the magnetic core width.

5. The method as in claim 1 wherein the neural network comprises a multi-layer network of a plurality of perceptrons.

6. The method as in claim 1 wherein the neural network comprises a multi-layer model having an input layer at least one hidden layer and an output layer.

7. The method as in claim 1 wherein the neural network comprises a feed forward neural network.

8. The method as, in claim 1 wherein the neural network comprises a plurality of feed forward neural networks.

9. The method as in claim 1 wherein the neural network includes a perceptron.

10. The method as in claim 1 wherein the neural network includes a multilayer perceptron.

11. The method as in claim 1 wherein the neural network includes a feed forward multi-layer perceptron.

12. A non-transitory computer readable medium comprising instructions for:

collecting data regarding manufacturing parameters;

constructing a neural network;

analyzing the data using the neural network to 1oroduce an output from the neural network; and

using the output from the neural network to provide new manufacturing parameters for optimizing a magnetic core width of a write head.

13. The method as in claim 12, wherein the data includes:

a write pole flare point;

a wrap around shield dimension; and

a side gap.

14. The method as in claim 13 wherein the data includes a magnetic sensor stripe height and a reader/writer offset.

15. The method as in claim 12 wherein the neural network comprises a multi-layer network of a plurality of perceptrons.

16. The method as in claim 12 wherein the neural network comprises a multi-layer model having an input layer at least one hidden layer and an output layer.

17. The method as in claim 12 wherein the neural network comprises a feed forward neural network.

18. The method as in claim 12 wherein the neural network comprises a plurality of feed forward neural networks.

19. The method as in claim 12 wherein the neural network includes a perceptron.

20. The method as in claim 12 wherein the neural network includes a multilayer perceptron.