general design guidelines for manual assembly

7/18/2019 GENERAL DESIGN GUIDELINES FOR MANUAL ASSEMBLY

http://slidepdf.com/reader/full/general-design-guidelines-for-manual-assembly-56d615a0c2fe1 1/74

GENERAL DESIGN GUIDELINES FOR

MANUAL ASSEMBLY

1



Objective of design for assembly1-Provide a tool for designer or design team for the

of product complexity and assembly

2-Guiding the designer to simplify the product for savingthe both manufacturing and assembly cost

3-Establishing a database that consists of assembly timeand cost factor for various design situations andproduction conditions

2



GENERAL DESIGN GUIDELINES FOR

MANUAL ASSEMBLY The process of manual assembly can be divided into

two separate areas:

1. handling (acquiring, orienting, and moving the parts)

2. insertion and fastening (mating a part to another partor group of parts)

3



Design Guidelines for Part Handling

1-Design parts that have end-to-end symmetry and

rotational symmetry about the axis of insertion.

4

pin washer Dowel pin

Parts with end-to-end symmetry

nail bulb Screw disk

Parts with no end-to-end symmetry



Parts with rotational symmetry

washer pin

bulbscrew

diskkey

Parts with no rotational symmetry

If this cannot be achieved, try to design parts having themaximum possible symmetry.




2-In those instances where the part cannot be made

symmetric, Design parts that are obviously asymmetric .

6




3-Provide features that will prevent jamming of partsthat tend to nest or stack when stored in bulk.

7




4- Avoid features that will allow tangling of parts whenstored in bulk.

8




5- Avoid parts that:

Stick together or are slipperydelicate, flexible, very small

Are very large

Are hazardous to thehandler (parts that aresharp, splinter easily, etc.).

9



Design Guidelines for Insertion and Fastening

Design so that there is little or no resistance toinsertion and provide chamfers to guide insertion oftwo mating parts.

Generous clearance should be provided, but care mustbe taken to avoid clearances that will result in atendency for parts to jam or hang-up during insertion(see Figs. 3.3 to 3.6).

10




11




12




13




14




Standardize by using commonparts, processes, and methodsacross all models and evenacross product lines to permitthe use of higher volumeprocesses that normally resultin lower product cost.

15




Use pyramid assembly -provide for progressiveassembly about one axis of

reference.

In general, it is best toassemble from above.

16




Avoid the necessity for holding parts down to maintaintheir orientation during manipulation of thesubassembly or during the placement of another part.

If holding down is required, then try to design so thatpart is secured as soon as possible after it has beeninserted.

17




Design so that a part is located before it is released.

18




When common mechanicalfasteners are used, the followingsequence indicates the relativecost of different fastening

processes, listed in order ofincreasing manual assembly cost:1. Snap fitting2. Plastic bending3. Riveting

4. Screw fastening

19



ASSEMBLY EFFICIENCY

20



Assembly Efficiency

Two main factors that influence the assembly or

subassembly cost of a product :1. No. of parts in a product.

2. Ease of handling, insertion, and fastening of parts.

21



Assembly Efficiency

DFA index: a figure obtained by dividing the theoretical min

assembly time by actual assembly time.

N min: Theoretical min No. of parts

ta: Basic assembly time for one part. It is the average time fora part that presents no handling, insertion, or fasteningdifficulties (about 3 s).

tma: Estimated time to complete assembly of the product.

22



EFFECTS OF DIFFERENT FACTORS INFLUENCING

MANUAL ASSEMBLEY

23



EFFECT OF PART SYMMETRY ON

HANDLING TIME

Assembly operations involve at least two component parts:

1. Part to be inserted

2. Part or assembly (receptacle) into which part is inserted

Orientation involves the proper alignment of the part to beinserted relative to the corresponding receptacle.

Orientation can be divided into two distinct operations:

1. Alignment of axis of part that corresponds to axis ofinsertion

2. Rotation of part about this axis.

24




HANDLING TIME

Two kinds of symmetry for a part:

1

a symmetry: depends on angle through which apart must be rotated about an axis perpendicular tothe axis of insertion to repeat its orientation.

2 b

symmetry: depends on angle through which apart must be rotated about the axis of insertion torepeat its orientation.

25




HANDLING TIME

a-symmetry and b-symmetry for various parts

26




HANDLING TIME

Time standard systems are used to establish assemblytimes in industry.

Several different approaches have been employed todetermine relationships between amount of rotationrequired to orient a part and the time required toperform that rotation.

Two of the most commonly used systems:1. Methods Time Measurement (MTM)

2. Work Factor (WF).

27




HANDLING TIME

MTM system:

Max possible orientation is employed, which is one-

half the beta rotational symmetry of a part. Effect of alpha symmetry is not considered.

MTM system classifies max possible orientation intothree groups:

1. Symmetric2. semi-symmetric

3. non-symmetric

28




HANDLING TIME

WF system: Symmetry of a part is classified by the ratio of the number

of ways the part can be inserted to the number of ways thepart can be grasped preparatory to insertion. For a square prism to be inserted into a square hole:

One particular end first can be inserted in four ways out ofthe eight ways it could be suitably grasped.

Hence, on average, one-half of the parts grasped wouldrequire orientation, and this is defined in the WF system as asituation requiring 50% orientation.

Account is taken of alpha symmetry, and some account istaken of beta symmetry.

29




HANDLING TIME

A single parameter: give a satisfactory relation between

the symmetry of a part and the time required fororientation.

Total angle of symmetry = a + b

30




HANDLING TIME

Fig. 3.19: Effect of total

angle of symmetry ontime required to handle(grasp, move, orient,and place) a part

Shaded areas: values oftotal angle of symmetrythat cannot exist.

31




HANDLING TIME Symmetry of a part can be conveniently classified into

5 groups.

The first group, which represents a sphere, is notgenerally of practical interest.

Therefore, four groups are suggested that are employedin the coding system for part handling (Fig. 3.15).

Comparison of these experimental results with MTMand WF orientation parameters showed that theseparameters do not account properly for the symmetryof a part.

32



EFFECT OF PART THICKNESS AND SIZE

ON HANDLING TIME

Thickness of a "cylindrical" part: radius

Thickness for non-cylindrical parts: max height of part

with its smallest dimension extending from a flatsurface.

33




ON HANDLING TIME Cylindrical parts: parts having cylindrical or other regular

cross sections with five or more sides. Non-cylindrical parts: Diameter is greater than or equal

to its length. Parts with a "thickness" greater than 2mm present no

grasping or handling problems. For long cylindrical parts this critical value would have

occurred at a value of 4mm if the diameter had been usedfor the "thickness".

Grasping a long cylinder 4mm in diameter is equivalent tograsping a rectangular part 2 mm thick if each is placed ona flat surface.

34




ON HANDLING TIME Size of a part:

largest non-

diagonaldimension ofpart's outline when projectedon a flat surface.

It is normally thelength of thepart.

35




ON HANDLING TIME Parts can be divided into four size categories as illustrated.

Large parts involve little or no variation in handling time

with changes in their size Handling time for medium and small parts displays

progressively greater sensitivity with respect to part size.

Since the time penalty involved in handling very smallparts is large and very sensitive to decreasing part size,

tweezers will usually be required to manipulate such parts. Tweezers can be assumed to be necessary when size is less

than 2 mm.

36



EFFECT OF WEIGHT ON HANDLING

TIME

The effect of increasing weight on grasping andcontrolling is found to be an additive time penalty and

the effect on moving is found to be a proportionalincrease of the basic time.

For the effect of weight on a part handled using onehand, the total adjustment t pw to handling time can be

represented by the following equation:

37




TIME W (Ib): weight of the part

th (s): basic time for handling a "light" part when no

orientation is needed and when it is to be moved ashort distance.

An average value for th is 1.13, and therefore the totaltime penalty due to weight would be approximately

0.025 W.

38




TIME If we assume that the max weight of a part to be

handled using one hand is around 10-20 Ib, the max

penalty for weight is 0.25-0.5 s and is a fairly smallcorrection.

Eq. (3.3) does not take into account the fact that largerparts will usually be moved greater distances, resulting

in more significant time penalties.

39



EFECTS OF COMBINATIONS OF FACTORS

Penalties associated with each individual factor are notnecessarily additive.

For example, if a part requires additional time to moveit from A to B, it can probably be oriented during themove. Therefore, it may be wrong to add the extra timefor part size and an extra time for orientation to the

basic handling time.

40



EFFECT OF SYMMETRY FOR PARTS THAT SEVERELY NEST OR TANGLE AND MAYREQUIRE TWEEZERS FOR GRASPING AND MANIPULATION

A part may require tweezers when:

1. Its thickness is so small thatfinger-grasp is difficult.

2. Vision is obscured andprepositioning is difficultbecause of its small size.

3. Touching it is undesirable

4. Fingers cannot access thedesired location.

41



EFFECT OF SYMMETRY FOR PARTS THAT SEVERELY NEST OR TANGLE AND MAY

REQUIRE TWEEZERS FOR GRASPING AND MANIPULATION

A part is considered to nest or tangle severely when anadditional handling time of 1.5s or greater is required

due to these factors. In general, two hands will be required to separate

severely nested or tangled parts.

Helical springs with open ends and widely spaced coils

are examples of parts that severely nest or tangle.

42





Figure 3.23: how thetime required fororientation is affected

by a and b angles ofsymmetry for partsthat nest or tangleseverely and may

require tweezers forhandling.

43





In general, orientation using hands results in a smallertime penalty than orientation using tweezers.

Therefore factors necessitating the use of tweezersshould be avoided if possible.

44



EFFECT OF CHAMFER DESIGN ON

INSERTION OPERATIONS

Two common assembly operations:

1.Insertion of a peg into a hole.

2. Placement of a part with a holeonto a peg.

The dimensionless diametral

clearance c between the peg andthe hole is defined by (D - d)/D

45




INSERTION OPERATIONS

46

FIG. 3.25: Effect of clearance on insertion time




INSERTION OPERATIONS From Fig. 3.25, the following conclusions have been drawn:

1. For a given clearance, the difference in the insertion time for twodifferent chamfer designs is always a constant.

2. A chamfer on the peg is more effective in reducing insertion timethan the same chamfer on the hole.

3. Max width of chamfer that is effective in reducing the insertiontime for both the peg and the hole is approximately 0.1D.

4. For conical chamfers, the most effective design provides chamferson both the peg & the hole, with w1 = w2 = 0.1D and q 1 = q 2 < 45.

5. Manual insertion time is not sensitive to variations in the angle ofthe chamfer for the range 10 < q < 50.

6. A curved chamfer can have advantages over a conical chamfer forsmall clearances.

47




INSERTION OPERATIONS The long manual insertion time for the peg and hole

with a small clearance is probably due to the type ofengagement occurring between the peg and the holeduring the initial stages of insertion.

48




INSERTION OPERATIONS Figure 3.26 shows twopossible situations that willcause difficulties:

1. In Fig. 3.26a, the two pointsof contact arising on thesame circular cross sectionof the peg give rise to forcesresisting the insertion.

2. In Fig. 3.26b, the peg hasbecome jammed at theentrance of the hole.

49




INSERTION OPERATIONS For a chamfer conforming to a

body of constant width,insertion time is independentof dimensionless clearance cin the range c > 0.001.

Curved chamfer is the

optimum design for peg-in-hole insertion operations(Fig. 3.25).

50



ESTIMATION OF INSERTION TIME

For conical chamfers (Fig. 3.24), where the width of 45° chamfers is 0. 1 d, the manual insertion time for a plaincylindrical peg ti; is given by

Whichever is largerf(chamfers) = -100 (no chamfer)=- 220 (chamfer on hole)=- 250 (chamfer on peg)=- 370 (chamfer on peg and hole)

51



EFFECTS OF OBSTRUCTED

ACCESS AND RESTRICTEDVISION ON INSERTION OF

THREADED FASTENERS OF

VARIOUS DESIGNS

52

Fig. 3.30a shows theeffects of the shape ofthe screw point and holeentrance, when the

assembly worker cannotsee the operation and

when various levels ofobstruction are present.



EFFECTS OF OBSTRUCTED ACCESS AND RESTRICTED VISION ON

INSERTION OF THREADED FASTENERS OF VARIOUS DESIGNS

When the distance from the obstructing surface to thehole center was greater than 16mm, the surface had noeffect on the manipulations and the restriction of vision

was the only factor.

Under these circumstances, the standard screw insertedinto a recessed hole gave the shortest time.

For a standard screw with a standard hole an additional

2.5s was required.

When the hole was closer to the wall, thereby inhibitingthe manipulations, a further time of 2 or 3s was necessary.

53





Restriction of vision had little effect when access wasobstructed.

This was because the proximity of the obstructingsurface allowed tactile sensing to take the place ofsight.

When the obstruction was removed, restricted vision

could account for up to 1.5s additional time.

54





Figure 3.31: Effect of number of threads on time topick up the tool, engage the screw, tighten the screw,and replace the tool.

There was no restriction on tool operation for any ofthese situations.

55





Fig. 3.32 shows the time toturn down a nut using a

variety of hand-operated

tools and where theoperation of the tools wasobstructed to variousdegrees.

The penalties for a box-end

wrench are as high as 4s perrevolution whenobstructions are present.

56



MANUAL ASSEMBLY DATABASE ANDDESIGN DATA SHEETS

57



MANUAL ASSEMBLY DATABASE AND

DESIGN DATA SHEETS For the development of the classification schemes and timestandards presented earlier it was necessary to obtain an estimateof the average time, in seconds, to complete the operation for all

the parts falling within each classification or category. Forexample, the uppermost left-hand box in Fig. 3.15 (code 00) gives1.13 for the average time to grasp, orient, and move a part

1. That can be grasped and manipulated with one hand

2. Has a total symmetry angle of less than 360° (a plain cylinder,

for example)3. Is larger than 15mm

4. Has a thickness greater than 2 mm

5. Have no handling difficulties, such as flexibility, tendency totangle or nest, etc.

58



MANUAL ASSEMBLY DATABASE AND

DESIGN DATA SHEETS To illustrate the type of problem that can arise through the use of the group

technology coding or classification scheme employed in the DFA method, wecan consider the assembly of a part having a thickness of 1.9 mm.

We shall assume that, except for its thickness of less than 2 mm, the part

would be classified as code 00 (Fig. 3.15). However, because of the part's thickness, the appropriate code would be 02

and the estimated handling time would be 1.69 instead of 1.13s, representing atime penalty of 0.56s.

Turning now to the results of experiments for the effect of thickness (Fig.3.20), it can be seen that for a cylindrical part the actual time penalty is on theorder of only 0.01 to 0.02 s.

We would therefore expect an error in our results of about 50%. Under normal circumstances, experience has shown that these errors tend to

cancel— with some parts the error results in an overestimate of time and withsome an underestimate.

59



FURTHER DESIGN GUIDELINES

60




Avoid connections

If the only purpose of a part or assembly is to connect A to B, then try to locate A and B at the same point

61



FURTHER DESIGN GUIDELINESDesign so that access for assembly operations is not restricted

62




Avoid adjustments

63




Use kinematic design principles

64




Use kinematic design principles

65



TYPES OF MANUAL ASSEMBLY METHODS

66



TYPES OF MANUAL ASSEMBLY

METHODS Part acquisition time is highly

dependent on:

1. Nature of layout of assembly area2. Method of assembly

For small parts placed within easyreach of assembly worker, handling

times given in Fig. 3.15 areadequate if employing:

bench assembly

multi-station assembly

67




METHODS It will not be possible to place an adequate

supply of parts within easy arm's reach ofassembly worker for volumes that: Do not justify transfer systems

Assembly contains several parts that weighmore than 5 lb or are over 12” in size. Largest part is less than 35” in size No part weighs more than 30 lb

In this case, modular assembly center mightbe used.

This is an arrangement of workbench andstorage shelves where parts are situated asconveniently for assembly worker as possible.

68




METHODS Because turning, bending, or walking may be

necessary for acquisition of some of the parts,

handling times will be increased. Three modular work centers to accommodate

assemblies falling within three size categories wherethe largest part in the assembly is:

1. less than 15 in2. from 15 to 25 in

3. from 25 to 35 in

69




METHODS For products with even larger parts,

custom assembly layout can be used. Product is assembled on a worktable or

on floor and the various storage shelves

and auxiliary equipment are arrangedaround the periphery of the assemblyarea .

Total working area is larger than thatfor modular assembly center anddepends on size category of largestparts in the assembly.

Three subcategories of customassembly layout are employed forassemblies whose largest parts are :1. from 35 to 50 in2. from 50 to 65 in3. larger than 65 in.

70




METHODS For large products a more flexible

arrangement can be used calledflexible assembly layout.

The layout would be similar insize to custom assembly layoutand the same three subcategories

would be employed according to

the size of the largest part. Use of mobile storage carts and

tool carts can make assemblymore efficient.

71




METHODS For high-volume assembly

of products containing

large parts transfer linesmoving past manualassembly stations would beemployed.

72




METHODS Two other manual assembly situations exist:

1. Assembly of small products with very low volumes.

This would include the assembly of intricate andsensitive devices such as the fuel control valves for anaircraft.

2. Assembly of large products is mainly carried out on

site.

73



THANK YOU

general design guidelines for manual assembly

Documents