Saturday, August 26, 2006

0 Variables and Parameters : Are they two the same?

Notes for myself and others based on my readings today from Theory and Design for Mechanical Measurements (Figliola,3rd edition).

1. The primary objective of measurement is to establish the value or tendency of some variable, which is that suggested by measurement device.

2. Any measurement problem is an open ended design problem, whose outcome will not have a unique solution. There may be several approaches to solving it.

3. Human senses of measurement are limited, and measurement systems and tools extends the capabilities of humans.

4. The constituents of a measurement system are :
a. sensor-transducer stage
b. signal conditioning stage
c. output stage control stage

5. Calibration is done by comparing the measurement device (an unknown) against an equal or better standard.

6. Sensor = physical element ----> senses variable being measured.
Transducer converts this sensation to a detectable signal form (electrical, mechanical etc)
Eg : Internal capillary design of a bulb in a thermometer acts as a transducer
(thermal information converted to mechanical displacement of mercury in the capillary)

7. Functions of signal conditioning :
a. modifying signal to a desired form
b. amplifying signal
c. removing portions of signal
d.providing mechanical mechanical or optical linkage between transducer and output stage.

8. Ouput stage indicates value of measurement.

9. Feedback control stage interprets the output signal and takes appropriate action regarding control of process. Simple example : Furnace thermostat at home.

10. 3 things a successful engineer would do before an experimental test :
a. parameter design plan - What am I trying to answer? What has to be measured? What variables will affect my results?
b. system and tolerance design plan - Selection of measurement device and test procedure based on some preconceived tolerance limits for error.
c. data reduction design plan - How will I interpret the data? How will I use the data to answer my question.

11. 2 types of variables - discrete (outcome of the roll of a dice), or continuous (pointer dial measuring pressure,strain, temperature etc).

12. Control extraneous variables that can have different outcomes on the output of a measurements. Else, be prepared to solve a puzzle!

Question : What is the difference between a variable and parameter, if there is any? I understand that a parameter can have an effect on the behavior of the measured variable. Is it not then a variable itself? Why the name 'parameter' ? I have posed this question to my friends at Eng-tips forum so lets see what they have to say.

Why can't a parameter also be an extraneous variable?

For example, in testing the boiling point for water, one can get different results if the barometric pressure is unchecked.

So we do another set of tests, this time controlling the pressure of water and we get consistent values.

In the first case, pressure was like an extraneous variable we forgot to control. In the second series of tests, we controlled it (so with the definition that a parameter is controlled or moved by something or someone, pressure, in this case, is a parameter, right?)

Am I making sense here? If what I said is true, then I'll truly understand that these two terms are interchangable.


Post a Comment

Thank you. I read every single comment.