ELECTRONIC PACKAGING AND INTERCONNECTION HANDBOOK PDF

adminComment(0)

Electronic packaging and interconnection handbook, 3rd edition [ Book Review]. Article (PDF Available) in IEEE Electrical. Electronic Packaging and Interconnection Handbook 4/E [Charles A. Harper] on bernasungueta.ga *FREE* shipping on qualifying offers. Publisher's Note: Products. Physics and even Marketing. • Electronic Packaging: Housing and interconnection of integrated circuits to form electronic systems. • Electronic Packaging must.


Electronic Packaging And Interconnection Handbook Pdf

Author:JUANITA HANSBURY
Language:English, Portuguese, Japanese
Country:Fiji
Genre:Lifestyle
Pages:513
Published (Last):14.11.2015
ISBN:213-8-70176-675-3
ePub File Size:27.81 MB
PDF File Size:19.29 MB
Distribution:Free* [*Registration Required]
Downloads:27567
Uploaded by: AIDA

Packaging has provided the necessary external wiring and interconnection DRM-free; Included format: PDF; ebooks can be used on all reading devices in electronics which we have witnessed in the last third of the twentieth century. Aview of modern electronic packaging technology is presented along with its applications packaging allows for electronic interconnection be- tween the Electronic Packaging and Interconnection Handbook, C. A. Harper (ed.), McGraw -. Springer, [2] C. Harper, Electronic Packaging and Interconnection Handbook 4/E. bernasungueta.ga bernasungueta.ga

Brinker, about distributed generation will bene- working in plasma deposition and and E. Giannelis, eds. Materials Research Society fit greatly from this book. Shevelko by R. Gupta pp. Marcel Dekker, Inc. In many cases they have supe- pp. This volume contains papers tion that apply atomic spectroscopy and Knowledge of the flow characteris- presented at the April meeting of the physics of atomic collisions include tics of polymers and filled polymers is the Materials Research Society.

Among dielectric surface modification, film de- important to anyone involved in pro- the topics covered in this book are mo- position, surface discharges, diamond cessing these systems, which certainly lecular hybrid materials, self-assembled growth, and plasma switching.

This includes many readers of electrical insu- and template-grown synthesis, charac- book, Atoms and Their Spectroscopic lation. This particular book is aimed at terization, biomaterials, electrical and Properties, is about atomic radiative students and at those who need infor- optical properties, mechanical proper- mation on a particular topic. There are ties, and processing.

JavaScript is currently disabled, this site works much better if you enable JavaScript in your browser. Free Preview. download eBook. download Softcover. FAQ Policy. Show all. Pages Wafer Bumping Totta, Paul et al. But in each case, only one part is faulty. Is the incorrect placement of the PQFP times as bad as resistor placement? Yes, because it not only results in times as many bad solder joints but most likely also results in far more performance problems in the completed circuit. Remember that the key is to not just find the faults, it is to assign a cause and improve the process.

Emphasizing the process, rather than separate design, manufacturing, and test issues, will typically lead to use of techniques such as QFD, DOE, and process control PC. This process might already be used in production, or it might be one that is proposed for a new product.

A complex process such as an SMT assembly line cannot be studied effectively by using the simple technique of varying one parameter while holding all others steady. Such a process ignores the interaction between parameters, a condition that normally prevails in the real world. If all interactions as well as primary parameters are to be tested, the number of experiments rapidly becomes large enough to be out of the question even for only a few variables. DOE has been developed to help reduce the number of experiments required to uncover a problem parameter.

DOE relies on statistics, particularly factorial analysis, to determine the relative importance of relationships between parameters. Initially, the implementation of DOE was the purview of statisticians, which meant that it was outside the realm of most engineers and line workers.

Factorial analysis is the study of the chosen parameters and all their possible interactions. It is neither time efficient nor easy to calculate if more than four parameters are involved. To allow more use of factorial analysis, fractional factorial analysis was developed, using only identified primary parameters and selected interactions.

Fractional factorials can be used effectively only with the correct selection of primary parameters and their interactions. Use of brainstorming and similar techniques from total quality management TQM can help, but it cannot guarantee that correct selections were made.

Taguchi introduced a technique called orthogonal arrays11 in an attempt to simplify the selection of parameters for fractional factorial analyses.

The technique is not simple to use and requires in-depth study. There are no guidelines for the selection of the possible interactions, a shortcoming also present in the original fractional factorial technique.

If the quality problems are due to one of the main parameters, this is less of an issue, but it still begs the question of selecting interactions.

The underlying principle is that most real-world problems can have their causes reduced to four or fewer primary causes plus their interactions. With four or fewer causes, the full factorial analysis is very appropriate. Once the causes have been identified, the process can be improved to produce the best possible product.

As shown in Fig.

First, variables are eliminated that are not a cause. The multivari charts are used in determining what type or family a variation belongs to and eliminating causes that are not in this family. Other first-level procedures include components search and paired comparisons, which are mutually exclusive as techniques, but either of which can be used in conjunction with multivari charts.

Components search requires disassembly and reassembly of the product a number of times to rule out assembly problems, as opposed to component problems. Paired comparisons are used when the product or part cannot be disassembled and must be studied as a unit. C is used as a final validation of the previous techniques.

Scatter plots are used primarily for relating the tolerance values of identified input variables to quality requirements presented by the customer.

Multivari Chart The multivari chart is used to classify the family into which the red X or pink Xs fall. A parameter that is indicative of the problem, and can be measured, is chosen for study. Sets of samples are then taken and the variation noted. The categories used to distinguish the parameter output variation are: 1 variation within sample sets cyclical variation is larger than variation within samples or variation over time, 2 time variation temporal variation between sample sets is larger than variation within sample sets or variation of the samples, and 3 variations within samples positional variation are larger than variation of sample sets over time or variation within the sample sets.

Navigation

These are shown in Fig. Begin the study by collecting, consecutively, three to five products from the process. At a later time, after a number of units have been produced in the interim, collect three to five products again. This is an important rule to observe to provide statistical validity to the samples collected. In the language of statistics, this is a stratified experiment; that is, the samples are grouped according to some criterion. This is not, therefore, a random selection of samples as is required in many statistical methods.

It also is not a control chart, even though the plot may resemble one. It is a snapshot of the process taken at the time of the sampling.

Incidentally, the multivari chart is not a new procedure, dating from the s, but it has been incorporated into this system by Shainen.

The purpose of the multivari chart is to discover the family of the red X, although on rare occasions the red X itself may become evident.

The anticipated result of a successful multivari experiment is a set, the family, of possible causes that includes the red X, the pink Xs, or the pale pink Xs. The family will normally include possible causes numbering from a few up to about Further experiments will be necessary to determine the red X or the pink Xs from this set. The example displays in Fig. They will later be separated from each other by scribing. Data recorded over a two shift period, from units from more than substrate printings, are shown in the chart.

Bibliographic Information

A graph of this data is also shown, with the range of data for each substrate indicated and the average of the four resistors for each substrate as well as the overall average for the three substrates sampled at that time. Note: 1. Unit-to-unit variation within a group is the largest.

Time variation is also serious as the time averages increase sharply. With this information about the family of the cause s of defects, additional experiments to find the cause s of the variation can be designed according to the methodology to be described later. Components Search Components search is used only when a product can be disassembled and then reassembled.

Follow us on

It is a partswapping procedure that is familiar to many who have done field repair. The first step is to select a performance parameter by which good and bad units can be identified. A good unit is chosen at random, measured, then disassembled and reassembled two times, measuring the performance parameter each time.

This establishes a range of variability of the performance parameter that is related to the assembly operation for good units. Repeat this for a randomly selected bad unit, once again establishing the range of variability of the performance parameter for assembly of bad units.

The good unit must remain a good unit after disassembly and reassembly, just as the bad unit must remain a bad unit after disassembly and reassembly. If this is not the case, then the parameter chosen as performance indicator needs to be reviewed. Because there are only three data points for each type of unit, the statistics of small samples is useful.

The first requirement here is that the three performance parameter measurements for the good unit must all yield values that are more acceptable than the three for the bad unit. The second requirement is that the minimum separation between the medians of variability of the good unit and the bad unit exceeds a minimum.

This is illustrated in Fig. The value of 1.

Area Array Interconnection Handbook

The next step is to identify the parts to be swapped and begin doing so, keeping a chart of the results, such as that in Fig. In this plot, the left three data points are those of the original good and bad units, plus their disassembly and reassembly two times.

The remaining data points represent those measurements for swapping one part for parts labeled A, etc. Three results are possible: 1 no change, indicating the part is not at fault, 2 change in one of the units outside its limits but the other unit remains within its limits; or 3 the units flipflop, the good unit becoming a bad unit and vice versa.

A complete flip-flop indicates a part that is seriously at fault—call it a red X. Parts with measurements that are outside the limits but do not cause a complete reversal of good and bad are deemed pink Xs, worthy of further study.

A pink X is a partial cause, so that one or more additional pink Xs should be found. This is called a capping run and is illustrated as well. A capping run should result in a complete reversal, as shown, indicating that there are no other causes.

Less than a full reversal indicates that other, unidentified, causes exist or that the performance measure is not the best that could have been chosen. If a single cause, a red X has been found, and remedial action can be initiated.

If two to four pink Xs are found, a full factorial analysis, to be described later, should be done to determine the relative importance of the revealed causes and their interactions. Once the importance has been determined, then allocation of resources can be guided by the relative importance of the causes. Paired Comparisons If the product cannot be disassembled and reassembled, the technique to use is paired comparisons.

The concept is to select pairs of good and bad units and compare them, using whatever visual, mechanical, electrical, chemical, etc. Do this for several pairs, continuing until a pattern of differences becomes evident.Other topics of the ten chapters include integrated circuit packaging, ball grid arrays, hybrid microelectronics, multichip modules, chip scale packaging, flip chip attachment, and printed wiring boards. Note: 1.

Navigation

After these two items are developed, and money is allocated to the project, any changes may increase or decrease the overall cost of the project.

The book is broken down into tems. In any organization, there should be clear standards for producing specifications.