semi合集-English.pdf - 第4028页

SEMI F50-0200 © SEMI 2000 6 Impulse Insulation Levels ( BIL) are just som e of the ways that events can be eliminated. 6.4.3 Ser vice Configurations 6.4.3.1 Determ i ne the factors th at are f i xed for the purposes of i…

SEMI F50-0200 © SEMI 20005

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

110%

1 10 100 1000

Duration (cycles)

Percent of Nominal Voltage

Figure 3

Example of Chart Summarizing

Monitoring Data

6.3.2 Correlate data to impact on factory.

6.3.2.1 For new sites, categorize voltage sag events

within magnitude/duration bins.

6.3.2.2 When evaluating new sites that are located in

different electrical utility service areas it is beneficial to

normalize the voltage sag data prior to comparison.

The use of magnitude/duration bins to place historical

or predicted event data creates discrete blocks of like

kind events. The impact on factories, causes of events,

and potential improvements may be evaluated on each

individual bin or groups of bins. Increasing the number

of magnitude/duration bins used in the data comparison

refines the accuracy, but also increases the effort

needed to translate events into discrete data bins. (See

IEEE 1346 and Related Information 2.)

6.3.2.3 For existing sites, define categories of event

impact on manufacturing process. For example:

• In Spec Event/No known impact

• In Spec Event/Minor impact

• In Spec Event/Major impact

• Out of Spec Event/No known impact

• Out of Spec Event/Minor impact

• Out of Spec Event/Major impact

6.3.2.3.1 The boundary between major and minor

impact is often cost or number of wafer moves lost

converted to an equivalent cost.

6.4 Recommend Improvements

6.4.1 Analyze costs, benefits, and risks.

6.4.1.1 The improvement recommendation process

should include the equivalent of identifying the costs

related to the disturbances, the costs related to

improvements, and the effectiveness of improvements.

The risks should be identified for taking no action, the

possibility that events will occur during the

implementation of improvements, and that events will

occur as a result of unknown factors resulting from the

installation of improvements. Improvements can

include corrective action to eliminate system faults,

changes to service configurations, and power

enhancements.

6.4.2 Corrective action to eliminate system faults.

6.4.2.1 The key to influencing an electric utility’s

voltage sag performance is mutual understanding of

measurement and improvement processes. Voltage

sags on utility electric systems are created because of

faults (short circuits) caused by a variety of events,

including lightning, trees contacting power lines,

equipment failure, and vehicles striking power poles.

In order to reduce the number of voltage sag events, it

is important to understand the specific cause of each

fault. Semiconductor manufacturers should request that

electric utilities share disturbance investigation reports

and statistics. If data is not available, or tracking fault

causes is not a focus (many utilities track only outage

causes) then a fault tracking system should be

established.

6.4.2.2 Many times, the initially identified fault cause

(for example lightning) has a more specific cause (for

example a contaminated insulator), with an even more

specific root cause (for example salt contamination on

coastal power lines in dry weather seasons).

Identifying this root cause helps to establish the

appropriate corrective action (for example, improved

insulator cleaning practices to include weather

considerations on coastal lines). Semiconductor

manufacturers and their electric utilities should work

together to ensure voltage sag event root cause

identification processes exists.

6.4.2.3 Analysis steps for electric utilities to identify

the root cause of system faults include the following:

Step 1 Locate the fault and identify what initiated the

fault.

Step 2 Investigate the underlying causes of the fault

to discover the root cause.

Step 3 Track faults and root causes in a database.

Step 4 Identify corrective actions.

6.4.2.4 Some of the more obvious corrective actions

include additional animal guards on exposed electrical

devices to reduce the effects from inadvertent touch.

Additional patrols and early removal of birds nests,

sources of nesting material, reduction in potential

roosting and nesting sites, sealing any possible entry to

electrical equipment against wildlife intrusion, and

designs using larger phase spacing and higher Basic

SEMI F50-0200 © SEMI 2000 6

Impulse Insulation Levels

(BIL) are just some of the

ways that events can be eliminated.

6.4.3 Service Configurations

6.4.3.1 Determine the factors that are fixed for the

purposes of improvement development and evaluation.

Some examples include, but are not limited to, physical

location of site, utility system configuration beyond the

immediate vicinity of the site, and electric rate

structures.

6.4.3.2 Once the electrical reliability and power

quality needs of a semiconductor factory are identified

and the reliability and power quality of the electrical

network in the area has been characterized, electric

utility service configuration can be considered. The

electric utility and the semiconductor manufacturer

should jointly develop a plan that balances reliability

and power quality. This plan should consider the

following service configuration options.

6.4.3.3 High voltage service configura tions.

6.4.3.3.1 Utility electrical service configurations have

a significant effect on the levels of power quality and

reliability. Semiconductor manufacturing facilities can

typically derive the highest service quality and

reliability from electrical service provided at the highest

voltage level. By bringing service to the semiconductor

factory from the highest available voltage system,

semiconductor facilities can eliminate their exposure to

electrical disturbances on lower voltage systems.

Seldom are disturbances that result from events

originating in lower voltage systems transferred into the

higher voltage systems to any significant degree.

Utility industry studies have indicated approximately

60−75% fewer voltage sag events (below 70% of

nominal) on the high voltage systems.

6.4.3.3.2 For plants with loads greater than 10 MW,

the highest voltage available is usually service at a

voltage between 69 kV and 345 kV. Voltages above

this range, while widely used by utilities, are not

usually economical to adapt to loads less than 60 MW

and may require lengthy regulatory approvals. If new

overhead power lines are required, environmental and

public issues associated with locating the lines may

reduce access to higher voltage lines.

6.4.3.4 Redundancy.

6.4.3.4.1 All on-site facilities and internal factory

distribution should have at least N+1 component

redundancy. Where N is the number of components

required to operate for maximum loading conditions

and +1 indicates a single additional component that will

operate to maintain the system capability in the event

that one of the original components is out-of-service. If

the plant is to be operated with no annual shutdowns for

maintenance, then the system should be designed with

enough redundancy to maintain every component in the

plant without dropping service to any load. This

requires at least a dual feed system that originates with

two or more utility sources and continues throughout

the semiconductor factory with appropriate transfer

schemes to keep the loads energized at all times and to

transfer loads without interruptions.

6.4.3.4.2 The most reliable service is one where there

are multiple sources connected in a network to the

semiconductor factory. This allows for adequate power

supply, even if one of the sources fails. If a network is

not available, a dual feed system can be configured to

provide an immediate transfer to the backup system in

the case of primary source failure, reducing outage time

to near zero. Additionally, if the two sources are

independent, a static transfer switch may increase

quality to a level higher than that of a network. If only

one source is available to a semiconductor factory with

a load of less than five megawatts, an alternative is on

site generation combined with a voltage sag ride-

through system.

6.4.3.5 Minimize exposure.

6.4.3.5.1 When choosing source configurations, it is

important to consider exposure at the semiconductor

factory site. The more line length the factory is

connected to, the more exposure there is.

a) It might not be desirable to have three lines serving

the plant if one is a long line that is prone to

failures. As a rule of thumb, more than three lines

connected in a network may reduce quality without

adding significantly to reliability.

b) If service is taken from a distribution class circuit,

then consider purchasing a so-called express or

dedicated feeder from the utility to isolate the plant

from neighboring facilities.

c) Review the line routes with the utility and consider

changes to reduce exposures, such as: where the

poles are vulnerable to being struck by vehicles, or

where trees are growing close to the transmission

lines.

6.4.4 Power Enhancements Technologies

6.4.4.1 All disturbances will not be eliminated from

the utility grid. In order to achieve the next step in

plant protection, it may be necessary to implement

some type of custom power option. Custom power is

so called because it is thought to be a custom solution

tailored to the needs of the process and the unique

situation of the site. Custom power options usually

involve some type of power enhancement and

conditioning system. These power electronics systems

are most often connected between the semiconductor

factory and the electric utility at the point of common

coupling (also known as the electrical service point).

Many custom power options include some energy

storage to ride through the disturbance, but they are

usually designed to carry the factory through only

momentary interruptions. In fact, for large factories

(>10 MW) some available ride-through systems only

operate to boost the voltage during sags and will not

carry the factory site through even short outages. The

trade-off is cost versus protection. The systems should

be economically evaluated as well as matched

technically to the needs of the site.

6.4.4.2 Electric utility provided custom power options

should be balanced against factory system voltage sag

immunity covered in industry guide for factory systems

voltage sag immunity.

6.5 Select and Implement Improvements

6.5.1 Both the electric utility and the semiconductor

manufacturer should agree criteria methodology for

prioritizing improvements. The following are examples

of prioritization criteria.

• Expected frequency of disturbances.

• Impact of fault on electric utility and

semiconductor processing.

• Relative cost of system improvement.

• Ability of action to reduce effects.

6.5.2 Select improvement to be implemented, identify

the schedules for installation, and define the new

system performance expectations.

6.5.3 Implement selected improvements.

6.5.4 With identified fault tracking and root cause

analysis processes electric utilities will be in a position

to communicate the cause of events, their corrective

actions, and the impact of improvements. Results will

be both immediate and long-term, but to ensure that the

continuous improvement process remains successful,

the impacts of improvements should be tracked. The

results of this tracking should provide feedback to the

continuous improvement process as a whole.

7 Related Documents

7.1 SEMI Standards

SEMI F42 — Test Method for Semiconductor

Processing Equipment Voltage Sag Immunity

SEMI F47 — Provisional Specification for

Semiconductor Processing Equipment Voltage Sag

Immunity

SEMI F49 — Guide for Semiconductor Factory

Systems Voltage Sag Immunity

SEMI E51 — Guide for Typical Facilities Services and

Termination Matrix

7.2 IEEE Standards

IEEE 141  IEEE Recommended Practice for Electric

Power Distribution for Industrial Plants

IEEE 446 — IEEE Recommended Practice for

Emergency and Standby Power Systems for Industrial

and Commercial Applications

IEEE 493 — IEEE Recommended Practice for the

Design of Reliable Industrial and Commercial Power

Systems

IEEE 1100 — IEEE Recommended Practice for

Powering and Grounding Sensitive Electronic

Equipment

IEEE 1250 — IEEE Guide for Service to Equipment

Sensitive to Momentary Voltage Disturbances

NOTE 3: As listed or revised, all documents cited shall be the

latest publications of adopted standards.

NOTICE: SEMI makes no warranties or

representations as to the suitability of the guides set

forth herein for any particular application. The

determination of the suitability of the guide is solely the

responsibility of the user. Users are cautioned to refer

to manufacturer’s instructions, product labels, product

data sheets, and other relevant literature respecting any

materials mentioned herein. These guides are subject to

change without notice.

The user’s attention is called to the possibility that

compliance with this guide may require use of

copyrighted material or of an invention covered by

patent rights. By publication of this guide, SEMI takes

no position respecting the validity of any patent rights

or copyrights asserted in connection with any item

mentioned in this guide. Users of this guide are

expressly advised that determination of any such patent

rights or copyrights, and the risk of infringement of

such rights, are entirely their own responsibility.