GEOMETRIC TOLERANCING

Geometric dimensioning and tolerancing (GD&T) is a symbolic language used on

engineering drawings and computer generated threedimensional solid models for explicitly
describing nominal geometry and its allowable variation.

TYPES OF CONTROL NEEDED FOR PARTS

POSITION OF FEATURES WITHIN PARTS FORM or SHAPE of FEATURES

(linear dimensions and diameters) Such as straightness, angularity
Such as holes or parallelism of spesific portions
of the part

A datum is a feature of a part that acts as a master

reference used to locate other features of the part.
A datum can be a point, a line, or a plane
Theoretically exact dimensions

The dimension determining the theoretically exact form, orientation or position

respectively must not be toleranced. The corresponding actual dimensions may only
vary by the tolerances of form, orientation or position specified within the tolerance
frame. This is illustrated in the figure below
 Geometric
Type of tolerance Symbol
characteristics

Form Straightness
Form Flatness Tolerances are given
Form Circularity along with modifiers
Form Cylindricity

Profile Profile of a line

Profile Profile of a surface

Orientation Perpendicularity
Symbol Modifier
Orientation Angularity
Free state
Orientation Parallelism
Least material condition (LMC)
Location Symmetry
Maximum material condition (MMC)
Location Positional tolerance

Location Concentricity Projected tolerance zone

Runout Circular runout

Runout Total runout

LMC: Least material condition of a component at that limit of size where the material of the feature is at
its minimum everywhere, e.g. maximum hole diameter and minimum shaft diameter.

LMS: The dimension defining the least material condition of a feature.

LMVS :The size generated by the collective effect of the least material size, LMS of a feature and the
geometrical tolerance (form, orientation or location).

LMVC: Least material virtual condition. The condition of a component of least material virtual size.

LMR: Least material requirement, defining a geometrical feature of the same type and of perfect form,
with a given dimension equal to LMVS, which limits a feature on the inside of the material e.g wall
thickness.

MMS :Maximum material size; The size of the maximum material condition.

MMVC: Maximum material virtual condition, is a perfect form condition of the feature.

MMVS: Maximum material virtual size, is the size generated by the collective effect of the maximum
material size, MMS, of a feature of size and the geometrical tolerance (form, orientation or location)
given for the derived feature of the same feature of size. Maximum material virtual size, MMVS, is a
parameter for size used as a numerical value connected to MMVC.

PJ: Projected tolerance zone. A geometric tolerance zone which projects from an actual design features
such as a hole
Maximum Material Requirement
The minimum assembly clearance occurs when each of the mating components is at its
maximum material size (e.g. the largest pin size and the smallest hole size) it additionally occurs
when the geometrical deviations e.g. the form, orientation and location deviations of the
components size and their derived features centre line surface form are also at their maximum.
Least Material Requirement
Assembly clearance are maximized when the sizes of the assembled features of size are at least
material size e.g. the smallest shaft size together with the largest hole size and when the
geometrical variations e.g. the form, orientation and location deviation and their derived
features are at zero.

30.1

30.0
Projected tolerance zones

In some cases, the tolerance of orientation and position shall apply not to the feature itself but to
the external projection of it. Such projected tolerance zones shall be indicated by the symbol P (in
circle)
Free State

This condition applies to parts made of elastic or plastic materials or of thin flexible materials. The free

state condition of a part: it is not restrained and subjected only to the force of gravity.

The distortion of such a part should be such that it will be brought within the specified tolerances for
verification at assembly or by assembly using forces expected under the normal assembly conditions.

An example of how a free state geometrical tolerance is applied is provided in the figure below. The
interpretation of this figure is that the tolerances followed by shall apply in the free state. The other
tolerances apply in the assembled condition
 Measurement Error and Uncertainty

Measurement
Uncertainty

Error

Measured Value True Value

The difference between measurement error and measurement uncertainty

should be explained clearly to evaluate the measurement results correctly. The
term measurement error means the difference between the "true value" and the
value found by a measurement.

According to the Guidance to Expression of Uncertainty in Measurement

(GUM), measurement errors are distinguished as random and systematic errors.
 Random Error
Random error in measurement typically arises from unpredictable variations of influence
quantities. These random effects under apparently equal condition at a given position give
rise to variations in repeated observations of the measurand. Although it is not possible to
compensate for the random error of a measurement result, it can usually be reduced by
increasing the number of observations and only be expressed statistically. Random errors
include: positioning and allignment errors, non-determinable fluctiations in ambient
conditions, transient fluctiations in the friction in the instrument and operator errors such as
reading errors.

Systematic Error
Systematic error, like random error, can not be eliminated but it can be reduced. If a
systematic error arises from a recognized effect of an influence quantity on a measurement
result, hereafter termed a systematic effect, the effect can be quantified and a correction or
correction factor can be applied to compensate for the effect. Recognized systematic errors
can generally be correlated with position along an axis and can be corrected for if the
relative accompanying random error is small enough. Systematic errors can often be
compensated to a certain degree using calibration techniques. Systematic errors include
calibration errors and changes in ambient conditions.

Random errors cannot be compensated for without real time measurement and
feedback into correction servo loop. Thus when evaluating the error budget for a machine
two distinct sub-budget based on systematic and random errors should be kept.

See ME410 lecture notes for more information about random and systematic errors:
http://www.me.metu.edu.tr/courses/me410/notes/Week3/Week3.PPT
 Measurement Uncertainty
The lack of exact knowledge of the value of the measurand explains the
uncertainty of the result of a measurement. The result of a measurement after
correction for recognized systematic effects is still only an estimate of the
measurands value due to the uncertainty arising from random effects and from
imperfect correction of the result for systematic effects.
It is used to describe an interval in which the true value can be expected to lie with a specified level of confidence
(p). For example, if uc is assumed to represent a normal (Gaussian) uncertainty distribution, then a 99.7% confidence
interval corresponds to using an expanded uncertainty obtained with kp = 3
0,30

1 x
e z 2 , z
2
0,25
f (x)
0,20
2
n
Standart
P 0,15

1

n 1 i1
(xi )
Deviation
0,10
z z
0,05 Uncertainty u(x)

n
x
0,00
5,9 6,1 6,3 6,5 6,7 6,9 7,1
Arithmetic Mean of
U p k p uc i

y U p Y y U p (8)
n Repeated Measurements
Type A Uncertainty uc Combined Standart Uncertainty n Number of Repeated Measurements
U95% Expanded Uncertainty (k95% 2) Standart Deviation
U99% Expanded Uncertainty (k99% 3)
 Measurement Uncertainty (Continued)

3 2 + + 2 + 3

68.3%
95.5%
99.7%
A generally accepted rule:

Measuring accuracy should be within the 10% of the workpiece tolerance.

Suppose that standart deviation in the dimensions of products is p due to
random causes only. Then in order to have at least 99.7% of the products
acceptable, the manufacturing tolerance should be T 6p.
 Least Square Principle
Most probable value of the observed quantities is the one which renders sum of the
squares of residual errors a minimum.

- Single variable arithmetic mean: X = xi/n

- Two dependent variables ex: straight line Y = aX + b

yn

x , y
yn1
:
: x i
y i

: n n
:
:
y4
a
x x y y
i i
, b y ax
x x
y3 2

y2 i

y1

x1 x2 x3 x4 ................. xn1 xn
Uncertainty Source Chart