Scribd
Upload
43 views

Chapter 1 Notes

1) A measurement system consists of four main parts: a sensor-transducer stage to sense the physical variable, a signal conditioning stage to modify the transducer signal, an output stage to display the measured value, and sometimes a feedback-control stage for process control applications. 2) Developing a measurement test plan involves defining the test objective, identifying process variables and parameters to control, selecting measurement techniques and equipment based on tolerance limits, and determining how data will be analyzed. 3) Calibration establishes the relationship between a measurement system's inputs and outputs by applying known standard values and observing the system's response. This allows the measurement value to be interpreted.

Uploaded by

Souvik Das
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
43 views

Chapter 1 Notes

1) A measurement system consists of four main parts: a sensor-transducer stage to sense the physical variable, a signal conditioning stage to modify the transducer signal, an output stage to display the measured value, and sometimes a feedback-control stage for process control applications. 2) Developing a measurement test plan involves defining the test objective, identifying process variables and parameters to control, selecting measurement techniques and equipment based on tolerance limits, and determining how data will be analyzed. 3) Calibration establishes the relationship between a measurement system's inputs and outputs by applying known standard values and observing the system's response. This allows the measurement value to be interpreted.

Uploaded by

Souvik Das
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 11

Chapter 1

Basic Concepts of Measurement


Methods

RS Figliola and DE Beasley; Theory and Design for Mechanical


Measurements; 2000

Basic Concepts of Measurement


Systems
Definitions:
A measurement assigns a specific value to
a physical variable.
The physical variable becomes the
measured variable.
A measurement system is a tool used to
quantify the physical variable.

Four Parts of Measurement


System
Sensor-Transducer Stage
Signal Conditioning Stage
Output Stage
Feed-back Control Stage

Definitions
The sensor is a physical element that uses
some natural phenomenon to sense the
variable being measured.
The transducer changes this sensed
information into a detectable signal form
(electrical, mechanical, optical, etc.)
The goal is to convert the sensed
information into a form that can be easily
quantified.

Intermediate Stage
Signal conditioning equipment modifies the
transducer signal to a desired form.
This optional intermediate stage can be used
to increase the magnitude of the signal
through amplification, remove portions of
the signal through some filtering technique,
provide mechanical or optical linkage
between the transducer and the output
range, or other similar activities.

Feedback-Control Stage
In those measurement systems involved in
process control, a fourth stage, called the
feedback-control stage, contains a controller
that interprets the measured signal and
makes a decision regarding the control of
the process.
This decision creates a change in a process
parameter that affects the magnitude of the
sensed variable.

Experimental design involves developing a


measurement test plan.

Q: What makes up a measurement


test plan?
1.

Parameter Design Plan: Defines test objective and


identifies process variables, parameters, and a means
of their control.

2.

System and Tolerance Design Plan: Selection of


measurement technique, equipment, and test
procedure based on some preconceived tolerance limit
of error.

3.

Data Reduction Design Plan: How the data will be


analyzed, presented, and used.

Variable Definitions:
Independent Variable can be changed independent of
others
Dependent Variable is affected by change in one or more
other variables
Control of Variables hold at constant value or prescribed
condition over duration of measurement
Discrete Variable possible values can be enumerated (dice)
Continuous Variable values can take on any level within
range of operation (pressure, temperature)
Extraneous Variables cannot be controlled during
measurement but affect the measured variable (operator or
machine particular functioning

More Definitions:
Parameters - functional relationships between variables
Control Parameter - has an effect on the behavior of

the measured variable and should be controlled during


test period
Noise - a random variation of value of the measured
signal due to variation in extraneous variables
A completely controlled variable contains no noise
Most signals have some noise; the question is how much!
A complete control
Environmental conditions
Thermal noise
Radio frequency

More Definitions Cont:


Interference: produces undesirable

deterministic trends on measured values,


such as a sinusoidal wave superimposed on
the signal path
Hum

and acoustic feedback in PA system

Well-defined

environmental conditions

AC Power 60 H3
Fluorescent Lighting 120 H3
Electromagnetic Interference EMI
Radio Frequency RF

It is a major objective to break up interference trends so


that they appear as random variation in data set. These can
be handles by statistics.

Random Test
Suppose y = f(xa, xb,) y=dependent, xa=independent
But y is also influenced by extraneous variables
z1, so that y=F(xa, xb,;z1,z2) the dependence
of y on xi can be found by holding all but one
independent variable constant and varying one
variable.
Extraneous variables can not be controlled, but
their influence can be minimized by randomized
test procedures.

Definition:

A random test is defined by a measurement matrix that sets


a random order on the value of an applied independent
variable.

Trends normally introduced by the coupling of a


relatively slow and uncontrolled variation in the
extraneous variables with a sequential application in
values of the independent variable applied will be
broken up.
This type of plan is effective for the local control of
extraneous variables that change in a continuous
manner.

Discrete extraneous variables such as instruments,


test operator, test conditions, etc can be broken up
using randomized test blocks.

A block is a data set whose extraneous variable is


fixed over a range of independent variable
applications. Subsequent block vary the extraneous
variable.

Definitions
Repetitions repeated measurement of
dependent variable during a single test run
(diameter of ball bearing batch - multiple
measurement on same batch/day)
Replication an independent duplication of
a series of test measurements under similar
operating conditions (ball bearing batch on
separate day under similar conditions)

Definitions
Calibration the act of applying a known
input to a measurement system in order to
observe the system output
The known value used for the calibration is
called the standard.

Definitions
Static Calibration value of variable remains
constant during each response observation. By
incrementing change in input variables and recording
the output at each level, a calibration curve can be
made.
Dynamic Calibration determines the relationship
between an input of known dynamic behavior and the
measurement system output.

Usually such calibrations involve either a sinusoidal signal or


a step change as the known input signal.
when the variables of interest are time dependent and timebased information is sought, dynamic information is needed.

Definition
Calibration Curve describes the static
input-output relationship for a measurement
system and provides a means to interpret the
output.

Static Sensitivity slope of the static calibration curve


K = K(x1) = (

dy
)x = x1
dx

K may not be constant over range if curve is nonlinear.


The static sensitivity is a measure relating the change in the
indicated output associated with a given change in a static input.

Definitions
Range - a measurement system should not be
operated outside its calibration range

Input Span ri = xmax xmin


Output Span ro = ymax ymin

Accuracy can be estimated during calibration


and represents the ability to indicate the true value.
Absolute Error = = true value indicated value

Percent Accuracy

= (1

truevalue

) *100

Precision, or repeatability, measures a systems ability to


indicate a particular value upon repeated but independent
applications of specific input.
Precision Error is the measure of random variation in
repeated measurements.
The average error in a series of repeated calibration
measurements defines the bias error.
The Bias Error is the difference between the average
value and the true value.

Definitions
Sequential Test incrementally varies the input value
over prescribed range (ramp up or ramp down)
Random Test randomly selects input values over
prescribed range (discrete or continuous values)
Random test minimizes interference and break up
hysteresis and observation errors. It insures that each
success measurement is independent of the previous.
Overall Instrument Error based on all known
error!
Ec= [e12 + e22 + e32 + + em2]1/2 (estimate)

Difference in output:
eh=yupscale-ydownscale
Hysteresis error:
%ehmax=(ehmax/ro)x100
Where ro=ymax-ymin
Predicted output:
YL(x)=a0 + a1x
Linearity Error:
eL(x)=y(x)-yL(x)
Maximum expected
Linearity Error:
%eLmax=(eLmax/ro)x100

Sensitivity Error:
eK a statistical measure of
precision error in estimate
of slope of calibration
curve. May be corrected
by adjusting the gain of
the system.
Zero Error:
ez is error in output at zero
input, can be compensated
by adjusting the systems
zero effect.

Repeatability
Error:
%(eR)max=(2(sx)/ro)x100
95% of observation
within 2 standard
deviation
68% within 1
observation

10

Standards
Standards:

When measurement system is calibrated, it is


compared with some standard whose value is
presumably known
There are hierarchy of standards
International Primary Standard
Transfer
Local
Working

11

You might also like