MSD - Getting started
 

Deciding what to plot [AL(i)]
You need to have available the relevant equations for the experiment you will carry out or which has already been carried out. You need to know which variables have been altered and which variables are, for this particular data set, constant. The constants might be universal constants (e.g. G, e, h, c), local constants (density of fluid being used, focal length of lens, resistivity of wire) or pure numbers.

Here is a diagram of the standard straight-line equation Y = M X + C

A function containing variables or pseudo-variables and numbers only

=

A combination of constants and numbers only

´

A function containing variables or pseudo-variables and numbers only

+

A combination of constants and numbers only
Your primary task is to manipulate the given equations into such a form that you can fit it into these boxes. Suppose, for example, that you start with a modified version of Van der Waals' equation

where a depends on what gas you use, T is the temperature and R the universal gas constant. For the purposes of this exercise, a, T and R are constants, while p and V are the raw variables. Multiplying out and re-arranging gives

so our box equation becomes

pV

=

- a

´

1/V

+

RT
and if we plot pV on the vertical axis against 1/V on the horizontal axis, we will obtain a straight line whose gradient is -a and whose vertical intercept is RT.
There is no limit to the number of ways in which you can generate functions of the raw variables to produce new 'plotting' variables to go into the purple boxes. What is crucial is that neither of the raw variables variable can be allowed to spill over into the green 'constant' boxes.
 

Rough calculations, etc. [BL(ii)]

Method 1

Open up a spreadsheet and populate a column with possible values of one of your raw variables

Use your basic equations to fill a second column with the corresponding values of the other raw variable. This may involve guessing or looking up values for some of the constants.

Once you've generated the columns containing your two raw variables, transfer both columns to Sheet2 of the Excel file. [Keep Sheet1 as evidence for BL(ii) later].

Fill two new columns with the values of the 'plotting' variables specified in the two purple boxes. Chart these two columns, using the Scatter option.

Inspect the resulting chart and table to check for

  • physical scale - will your experiment fit into a lab conveniently?
  • range - do you have a reasonable number of plots, evenly spread along the graph?
  • sign - are you going to be able to make/interpret measurements of either sign?

If there are any shortcomings, make a note of additional values of the independent variable that will have to be used.

  

Method 2

(This method is only of use if your raw equation is not linear and your pseudo-variable equation is.)

Obtain a two-column table containing the raw variables.

Chart them un-processed.

Observe that you have not got a straight line, and that you intend to

  • consider the theory

  • develop a pseudo-variable equation

  • plot a straight line graph

 

 
 

Effects [BR(ii)]
Think about precision and uncertainty in relation to your raw readings. You'll want to do something quantitative on this later in C/D, but for now just think about the root causes of inaccuracy. These can constitute 'effects which might have an effect'. Think about how to minimise them or take them into account. You might, for example, decide to do repeats so that error bars can give an indication of how damaging the effects might be. Or you might decide to modify techniques.