MAX785 데이터 시트보기 (PDF) - Maxim Integrated

부품명

상세내역

제조사

MAX785

Product Reliability Report

Maxim Integrated 

Product Reliability Report

processes with extremely tight control. Each wafer

must pass numerous in-process checkpoints (such

as oxide thickness, alignment, critical dimensions,

and defect densities), and must comply with Maxim’s

demanding electrical and physical specifications.

Finished wafers are inspected optically to detect any

physical defects. They are then parametrically tested

to ensure full conformity to Maxim’s specifications.

Our parametric measurement system is designed to

make the precision measurements that will ensure

reliability and reproducibility in analog circuits.

We believe our quality-control technology is the best

in the industry, capable of resolving current levels

below 1pA, and of producing less than 1pF capaci-

tance. Maxim’s proprietary software allows automatic

measurement of subthreshold characteristics, fast

surface-state density, noise, and other parameters

crucial to predicting long-term stability and reliability.

Every Maxim wafer is subject to this rigorous screen-

ing at no premium to our customers.

Failure-Rate History

The graph below (Figure 8) illustrates Maxim’s

Failures-in-Time (FIT) rate performance. It also

highlights the progressive improvements made in

this FIT rate, a trend that we expect to see contin-

ue, thanks to our established continuous-improve-

ment methodology.

Infant Mortality Evaluation

Maxim evaluates each process and product family’s

Infant Mortality rate immediately after achieving

qualified status. Through Infant Mortality analysis, we

can identify the common defects for each process or

product family. For an illustration of Maxim’s low

Infant Mortality rate, refer to Table 2. Figure 9 is an

Infant Mortality pareto chart showing each category

of failures; categories are prioritized based on rela-

tive frequency.

______________________________Reliability Data

Merits of Burn-In

Figure 10 plots Failure Rate versus Time for the

metal-gate CMOS process. The plot is based on

Table 3’s Life Test data and Table 2’s Infant Mortality

evaluation data, both applied to a General Reliability

model. From this data, the benefit of production

burn-in can be derived.

8

7.52

7

6

5

4

7.49

6.80

5.63

4.79

3

2.88

2.68

2

1

Sept '85

May '86

July '87

April '88

Mar '90

Mar '91

Jan '92

Figure 8. Maxim FIT Rates Over Time

2.47

Jan '93

2.51

Jan '94

1.31

Jan '95

1.29

Jan '96

1.25

Jan '97

8 ______________________________________________________________________________________

Share Link: