D7 LAMP Manual

After starting up LAMP you need to define the instrument and where to find the data files. Click on the “DATA Access” button in the top left-hand corner of the window and new buttons will appear. The second button from the left (next to the “Self ...” button) defines the instrument name. From the pull-down menu here, select “TOF” and then “D7”. The button to the right of this defines where to find the data files. From the pull-down menu, select “d7” (not “On-Line”). This reads the data directly from the DEC Alpha workstation and means that you can read data from a run which is still going.

You should also define your printer by clicking in the “Options” -> “Titles..” menu (just above the plot region). This will open a new window and at the bottom here, you will see a field for “Default Postscript DEVICE”. Here you enter "ep1_d7" for the Epson printer in the cabin, or "lc1_07ver" for the colour LaserJet outside the IN6 cabin,  and you can then return to the main window by clicking “Done”.

These pages are intended as an introduction to LAMP and specifically, D7 data analysis using LAMP. For a more complete description (but without the D7-specifics) see the LAMP manual, which can be obtained by selecting the “Help” ->“manual” menu in LAMP. 

I strongly recommend that you select the option “Extend to classical LAMP” in the “Lamp/layout” menu.  This gives all the functionality of LAMP in a more visible manner than the menus, and in these pages I assume that you have done this.

A workspace is a buffer for storing data. LAMP predefines 20 workspaces which are known as w1, w2, ..., w20. These workspaces can contain whole runs or series of runs – not just single spectra as in Genie. When you read data into a workspace, it automatically adapts to the dimensions of the data. In the simplest case you read data using the “Read” button. Workspaces can then be displayed by selecting them and pressing the “Plot W ” button. Workspaces can be manipulated using intuitive commands such as: 

w3 = w1+w2 or w5 = w4/3.7 or w6 = total(w5) etc.

Standard instrument procedures are provided either at the press of a button or via a function such as: 

w2=normalise(w1)

in which data in w1 are intensity normalised using the counting time.

To read a run or a series of runs into a workspace, enter the run number(s) into the field to the right of the “Read” button and click the button, e.g. 3713 (reads run number 3713), 3713>3716 (reads run numbers 3713 up to 3716 and adds them together), 3713:3716 (reads run numbers 3713 up to 3716 and concatenates them).

There are 7 different numerical arrays associated with each workspace read in by LAMP.  In addition there 6 text arrays which deal with captions.  These are

1. x
2. y
3. z
4. e
5. n
6. p
7. pv
8. captions: w_tit, x_tit, y_tit, z_tit, other_tit, par_txt

For each workspace, a series of parameters are defined. These are (for w3):

x3 
an array containing the x-axis values for the data points in w3. e.g. if w3 has dimensions (128,32), x3 should have dimensions (128)

y3 
the y-axis (only applicable if w3 has 2 dimensions or more). e.g. if w3 has dimensions (128,32), y3 should have dimensions (32)

z3 
the z-axis (only applicable if w3 has 3 dimensions or more)

e3 
the error bars (usually the standard deviation) associated with w3. Must have same dimensions as w3

n3 
the monitor/counting time. On D7, n3 always has dimensions (4,7) and contains the following:

It is seen that index numbers start at 0 in this programming language (IDL). The first line, i.e.  n3(*,0) contains zeros for consistency with the format of other instruments.

p3 
has dimensions (50). Contains instrument parameters relevant to the data in w3, such as flipper currents, time-of-flight parameters, flipping ratios, etc. They can be seen and edited by pressing the “Data Params” button.

pv3 
has dimensions (50,nruns). Contains the same information as p3 if only one run is stored in w3.  If the workspace consists of a concatenation of runs, the parameters for each run are stored  separately in pv3.

captions 
These are: x_tit(3), y_tit(3),w_tit(3) and other_tit(3) with the obvious meanings.

Workspace manipulation can take place in one of three ways:

1. The manipulation can be entered in the “MANIPULATIONS” field followed by return
2. It can be entered in one of the function fields (the five fields to the right of the “User Macros?” button) followed by return or by clicking “Do”
3. It can be written into a command file (.prox file) and executed by entering @filename.prox in the “MANIPULATIONS” or function fields.

Each time workspaces are manipulated, most of the parameters are passed on, with parameters of the first workspace after the “=” taking priority, e.g 

w1 = w2 + w3

automatically invokes: 

p1 = p2
x1 = x2
y1 = y2
x_tit(1) = x_tit(2)  &  y_tit(1) = y_tit(2)  &  w_tit(1) = w_tit(2)

which is what you would expect. However, it also invokes 

n1 = n2 
e1 = e2

which is often not at all what you expect.

There are two very important points here, which are often a source of error:

1) If you are adding together raw (un-normalised) data, you should always explicitly add the monitors, i.e. 

n1 = n2 + n3

in the above example

2) You should always explicitly perform the error bar calculation, i.e. 

e1 = sqrt(e2^2 + e3^2)

in the above example. Note that you do not need to do this when the manipulation is done by an existing IDL program, e.g. 

w1 = normalise​​​​​​​(w2)

automatically gets the error bars right.

Many different modes of measurement are possible on D7; there are two types of data format:

Without time-of-flight analysis: Data is one- or two-dimensional: x-axis is detector angle, y-axis is spin measurement (known as “phase”).   e.g. for w1, if the measurement is without polarization analysis (No-PA) w1 has dimensions (64) since there are 64 detectors. For measurement with and without spin-flip (Z-PA) w1 has dimensions (64,2) and for measurement with full XYZ-PA w1 has dimensions (64,6) arranged as follows

w1(*,0) - Z non-flip 
w1(*,1) - Z flip 
w1(*,2) - X non-flip 
w1(*,3) - X flip 
w1(*,4) - Y non-flip 
w1(*,5) - Y flip

With time-of-flight analysis: Data is two-dimensional: x-axis is channel number, y-axis is scattering angle (which is repeated for PA measurements). E.g. for w1, if the measurement is No-PA w1 has dimensions (128,64) (i.e. 128 time channels, 64 detectors). For measurement with Z-PA, w1 has dimensions (128,128) and for measurements with XYZ-PA w1 has dimensions (128,384) arranged as follows:

w1(*,0:63)  - Z non-flip 
w1(*,64:127) - Z flip 
w1(*,128:191) - X non-flip 
w1(*,192:255) - X flip 
w1(*,256:319) - Y non-flip 
w1(*,320:383) - Y flip

1. To see the dimensions and contents of workspaces and parameters, the following two commands can be used:

help, w1

which tells you the type of variable w1 is and its dimensions, and 

print, w1

which prints out its elements. The results are displayed in the winterm window from which LAMP was launched.

2. Some useful functions are given in examples below. For the example we assume that w1 contains XYZ-PA time-of-flight data and therefore has dimensions (128,384):

w2 = w1(*, 0)

This takes all time channels of the first spectrum of w1, i.e. w2 will contain spectrum 1 of w1 (Z non-flip). Remember that index numbers in IDL start with 0, so if w1 has dimensions (128,384), the highest channel number (spectrum number) is 127 (383).

w3 = w2(70:80)

takes out the region from channels 70 to 80 (the elastic peak, for instance). w3 will have dimensions (11)

w2 = total(w1, 1)

integrates along the first dimension (i.e. channel number). If w1 has dimensions (128,384), w2 will have dimensions (384) and contain the integrated intensity as a function of scattering angle. Similarly,

w2 = total(w1,2)

will integrate along the second dimension (i.e. scattering angle). w2 will now have dimensions (128) and contain a summed time-of-flight spectrum for all the detectors and all phases. Combinations are also possible, such as

w2 = total(w1(*,0:63), 1)

which integrates the counts in each spectrum but only does it for spectra 0 to 63, i.e. the Z non-flip phase. w2 will have dimensions (64).

It is usually useful to create command files for reading and treating data. To do this, press the “User Macros?” button. The resulting window is a crude editor for creating macros and command files. The programming language is IDL. Enter the name of the command file you wish to create (must end with .prox) in the file name field, e.g. test.prox. The first line of your command file will usually consist of reading in run numbers. There are three read-in functions:  

w1 = rdrun(3276)

reads a single run (same as entering 3276 in the “Read” field)

w1 = rdsum(3276,3280)

reads and adds together all runs from 3276 to 3280 (same as entering 3276>3280 in the “Read” field)

w1 = rdand(3276,3280)

reads and concatenates all runs from 3276 to 3280 (same as entering 3276:3280 in the “Read” field)

normalise.pro

The next step is then usually to run the macro normalise, which does two things: it normalises to counting time or incident beam monitor (up to six different counting times or monitor counts for an XYZ-PA measurement). It also removes (if desired) every other spectrum, keeping the even detector numbers, since only these have polarizers in front. It is called by entering  

w2 =  normalise(w1)

A simple command file test.prox could then consist of two lines  

w1 = rdsum(3276,3280)  
w2 =  normalise(w1)

To save this file (and simultaneously check the syntax) press the “Compile new file” button and then “Exit”. You can now execute the command file by entering @test (the .prox extension is taken by default) and hitting return. Any error messages or run-time comments are displayed in the winterm window from which LAMP was launched. The workspaces can be examined with the “Plot W ”, “Scroll Spectra”, “GK_Fit” or “SuperPlot” buttons (see LAMP manual for descriptions).

A more sophisticated command file could be

w1 = rdsum(3276,3280) ; sample runs
w2 =  normalise(w1)
w3 =  rdrun(3281)      ; empty can run
w4 =  normalise(w3)
w5 = background(sample = w2, empty = w4)

here it is seen that the semicolon is the comment separator in IDL. The other important point is that the error calculation in the last line has been performed explicitly.  In this batch file we have used the macro background.pro to subtract the empty can run.  This macro will be explained more fully below.

A macro called concatenate.pro exists that allows you to easily concatenate runs that have already been read in. Say you have measured in two different positions of the detector banks, but you measured position 1 before position 2 for the sample run, but position 2 first for the empty run, you would then need to do something like:

w1 = rdand(4910,4911) ; sample runs
w2 =  normalise(w1)
w3 = rdrun(4913)     ; empty can position 1
w4 = rdrun(4912)      ; empty can position 2
w5 = concatenate(w3,w4)
w6 = normalise(w5)
w7 = background(sample = w2, empty = w6)

After subtraction of the empty can signal, the macros used depend on the type of measurement. There are, however, two command files which are common to most types of measurement on D7. These are the quartz and vanadium correction command files. After describing the procedure for these two corrections a typical command file for data analysis of each type of measurement will be explained

The D7 data analysis macros are divided into four sections - 

1. The BASE macros (found in the D7/BASEDIR section of the macros list), 
2. The GENERAL macros (not restricted to D7 data), 
3. The TOF macros (including normalise.pro - kept in the TOF section of the macros list)

Another category of "special" D7 macros visible from the LAMP macros window, are not used by the user under normal circumstances and will not be described here.

1.1. add_xyz.pro

Syntax: 

w3 = add_xyz(w2[, /zpo][, /xyz][, /nopo])

(optional keywords shown in square brackets)

Adds the X, Y and Z measurements together. Must be performed on raw data, i.e. before calling normalise.pro

KEYWORDS: 
 /zpo: (default) 
        three non-flip measurements are added together to give one three flip measurements are also added together. 
        DIMENSIONS: 
                w_in(nspectra,6,nruns) -> w_out(nspectra,2,nruns)

 /xyz: 
        after separately adding together the non-flip and flip measurements, the sums are then copied into the X and Y phases. 
        DIMENSIONS: 
                w_in(nspectra,6,nruns) -> w_out(nspectra,6,nruns)

 /nopo: 
         add all flip phases together to obtain a single measurement. Then copy sum into the Z1, X and Y phases. 
         DIMENSIONS: 
                w_in(nspectra,6,nruns) -> w_out(nspectra,6,nruns)

DIMENSIONS: (see above)

Syntax:   

w3 = background(sample = w1, empty = w2[, cadmium = w3][, T = <T>][, muR = <muR>][,MS_file = '<filename>'])

(optional keywords/arguments shown in square brackets)

Performs background subtraction of normalised data.

ARGUMENTS: 
         sample : workspace containing sample runs (required) 
         empty : workspace containing empty runs (required) 
         cadmium : workspace containing of cadmium runs (optional) 
         T : sample transmission (default: T=1) 
         muR : muR parameter required for Hewat-type angular transmission correction
         MS_file: name of multiple scattering correction file output by MSCATT (see the D7 MSCATT guide for details).

 N.B. Specify either T for a normal background subtraction or muR for an angular dependent background subtraction - NOT BOTH
         If using a MSCATT file, then don't include a transmission factor.

DIMENSIONS: 
         TOF:         w_in=w_out(nchannels,nspectra*nphases) - only deals with 1 run 
        NOTOF:    w_in=w_out(nspectra,nphases,nruns) - unless nruns=1

Syntax:  

w3 = binq(w1, dQ = <dQ>)

Interpolates data along the x-axis to regular steps with point spacing dQ. Only runs on 1d data.

ARGUMENTS: 
         dQ :desired constant Q-bin width

DIMENSIONS: 
         w_in(nspectra) -> w_out(nQ)

Syntax:   

w3 = components(w2[, /Schaerpf])

(optional keywords shown in square brackets)

Separates data taken with Z or XYZ polarization analysis into the constituent components:

index 
     0: total 
     1: incoherent 
     2: coherent + isotope-incoherent 
     3: paramagnetic (average of spin-flip and non-spin-flip paramagnetic) 
     4: non-spin-flip paramagnetic 
     5: spin-flip paramagnetic

KEYWORD: 
         /Schaerpf:  use original Schaerpf expressions for component separation

DIMENSIONS: 
         No-TOF: w_in(32,nphases,nruns) -> w_out(32,ncomponents,nruns) 
         TOF: w_in(128,nspectra*nphases) -> w_out(128,nspectra*ncomponents)

Syntax:  

w3 = concatenate(w1, w2)

Concatenates two workspaces. Must be performed on raw data, i.e. before calling normalise.pro

DIMENSIONS 
         any - but first n-1 out of n dimensions must be equal in w1 and w2 
         e.g. w1(nspectra,nphases,nrun1) 
         w2(nspectra,nphases,nrun2) -> w3(nspectra,nphases,nrun1+nrun2)

Syntax:  

w2 = corr_xyz(w1, filenum = <filenum>[, temp = <temp>])

(optional arguments shown in square brackets)

Corrects for flipping ratios and optionally for multiple scattering

ARGUMENTS: 
         filenum : numor(s) of flipping ratio file(s) to use:  e.g. filenum=27134 to use the file "quartz_27134.dat" 
                      filenum=[27134,27135,27136] to use multiple files 
         temp : specified temperature in case of temperature dependent flipping ratio files as created by quartz.pro 
                    e.g. filenum=27134, temp=100 - reads file "quartz_27134_100.dat"

DIMENSIONS: 
         TOF -> w_in=w_out(nchannels,nspectra*nphases) 
         NOTOF -> w_in=w_out(nspectra,nphases,nruns)

Syntax:  

w10 = fold(w9)

Folds the four quadrants of the Qx-Qy plane, output from QxQy.pro, into one.

DIMENSIONS: 
         w_in(nQ.nQ) -> w_out(nQ/2,nQ/2)

Syntax:  

w1 = input_quartz(numor = <numor>[, temp = <temp>])

(optional arguments shown in square brackets)

Read a quartz flipping ratio file 'quartz_<numor>_<temp>.dat' and loads it into the desired workspace

ARGUMENTS: 
         numor : numor corresponding to quartz file of interest 
         temp : temperature parameter of corrected quartz file (see quartz.pro)

Syntax:  

w1 = input_vanadium(<numor>[, nspectra = <nspectra>][, /number][, /angle])

(optional arguments/keywords shown in square brackets)

Reads the vanadium file 'vanadium_<numor>.dat', and puts it into a 1-D workspace. Also reads in multiple numors (e.g numor=[47324,47336]) and puts them into a 2-d workspace.

ARGUMENTS: 
         numor :numor(s) corresponding to vanadium file(s) to be read in 
         nspectra :number of spectra to read in from file

KEYWORDS: 
         /number :x-axis is spectrum number (default) 
         /angle :x-axis is scattering angle

Syntax:  

w6 = isotropic_ms(w5, roh = <roh>, muR = <muR>, abs = <abs>, nuc = <nuc>, inc = <inc>[, mag = <mag>])

(optional arguments shown in square brackets)

Takes background subtracted D7 data of 1, 2 or 6 phases and corrects for multiple scattering using the isotropic approximation of Wells and Cywinski, and the secondary scattering coefficients of Blech and Averbach. At present, deals only with cylindrical samples. Correction of TOF data not yet implemented.

 References 
 ---------- 
 P. Wells and R. Cywinski, Aust. J. Phys. 34 (1981) 193 
 T. M. Harders, T. J. Hicks and P. Wells, J. Appl. Cryst. 18 (1985) 131 
 J. Mayers and R. Cywinski, Nucl. Inst. Meth. Phys. Res. A241 (1985) 519 
 I. A. Blech and B. L. Averbach, Phys. Rev. 137 (1965) A1113

ARGUMENTS: 
         roh : R/h - radius of cylinder over height 
         muR : Nsigmat*R calculated from transmission measurements 
         abs: Absorption Cross-section (at experimental wavelength) in barns 
         nuc: Nuclear Cross-section in barns 
         inc: Spin-Incoherent cross-section in barns 
         mag: Magnetic cross-section in barns (optional)

DIMENSIONS 
         w_in = w_out(nspectra,nphases,nruns) - unless nruns=1

Syntax:  

w2 = norm_inc(w1, inc_xs = <inc_xs>[, debye = <debye>])

(optional arguments shown in square brackets)

Normalise intensities to the incoherent scattering of the sample.  Must be called after components.pro and theta_scan.pro and phi2q.pro.

ARGUMENTS: 
         inc_xs :incoherent cross-section of the sample (required) 
         debye :Debye-Waller factor (optional)

DIMENSIONS 
        x-axis must be in Q -> w_in=w_out(nQ)

Syntax:  

w2 = norm_van(w1, filenum = <numor(s)>[, S_mass = <S_mass>, fwt = <fwt>, V_mass = <V_mass>])

(optional arguments shown in square brackets)

Correct for "apparent" detector efficiencies using vanadium data. If desired, normalise intensity to an absolute scale.

ARGUMENTS: 
         filenum :number of vanadium integral file - "vanadium_<filenum>.dat" created by vanadium.pro. 
                Is possible also to input multiple vanadium file numbers for use with theta_scan data, 
                e.g. filenum=[13487,13491] - use 5 vanadium files numbered, 13487,13488,13489,13490,13491 
                i.e. sequential order 
                filenum=[13487,13490,13499] - use the three specified files,  i.e. non-sequential order

         S_mass :sample mass (g) -optional 
         V_mass :V mass (g) -optional 
         Fwt :formula weight (amu) per f.u. -optional

DIMENSIONS: 
         w_in=w_out(nspectra,nphases,nruns) unless nruns=1

Syntax:  

w5 = omega_scan(w4[, /all_angles][, /pos_angles][, /neg_angles])

(optional keywords shown in square brackets)

Input: 2-D or 3-D workspace containing a non-tof omega-scan read in with rdand, normalised with normalise.pro and optionally separated into components with components.pro.  Rearranges to a 1-D or 2-D workspace with detector angle theta as x-axis and omega angle as y-axis.

KEYWORDS: 
         /neg_angles : only use data with negative angles 
         /pos_angles : only use data with positive angles 
         /all_angles : use all banks (default)

DIMENSIONS: 
         input format: w(32_or_64,nphases,nruns) - unless nphases or nruns is 1 
         output: w(nspectra,nruns,nphases) - unless nphases is 1

Syntax:  

w10 = phi2q(w9)

Transforms scattering angle on x-axis to Q, assuming elastic scattering.

DIMENSIONS: 
         w_in=w_out(nspectra)

Syntax: 

w10 = qrebin(w9[, omegashift = <omegashift>][, /fold])

(optional arguments shown in square brackets)

Takes output workspace of omega_scan.pro containing omega-scan, converts to Q-space and rebins to a constant Qx-Qy grid, giving sensible error bars! Deals with only one spin phase at a time.

ARGUMENTS: 
         omegashift : offset in degrees of omega=0 position

KEYWORDS:
         /fold:  optionally fold data on to + + quadrant

DIMENSIONS: 
         w_in(nspectra,nruns) ->w_out(nQ,nQ) (folded),  w_out(2*nQ,2*nQ) (unfolded)

Syntax:  

w5 = quartz(w4[, /nosave][, FR_depol = <FR_depol>, temp = <temp>])

(optional arguments/keywords shown in square brackets)

Takes 2-D or 3-D workspace containing a normalised non-tof z-PA or xyz-PA background-subtracted quartz run (or runs). Puts the flipping ratios into w_out and also into the file 'quartz_"numor".dat'. Will also correct for depolarization due to the sample (assuming cylindrical geometry) if desired.

ARGUMENTS: 
         FR_depol :if depolarization is present in the sample -  FR_depol is flipping ratio (in M2) 
                in the presence of depolarization 
         temp :if depolarization is present in the sample - temp is the temperature (used to name the output file 
                "quartz_<numor>_<temp>.dat") to identify the depolarisation correction at that temperature

KEYWORDS: 
         /nosave :don't save flipping ratios to file

DIMENSIONS: 
         w_in(nspectra, nphases) -> w_out(nspectra)

Syntax:

w6 = QxQy(w5[, omegashift = <omegashift>])

(optional arguments/keywords shown in square brackets)

Takes output workspace of omega_scan.pro containing omega-scan, converts to Q-space and interpolates to a constant Qx-Qy grid (rough and ready version of qrebin.pro) Deals with only one spin phase at a time.

ARGUMENTS: 
         omegashift : angle (in degrees) of omega=0 position

DIMENSIONS 
         w_in(nspectra,nruns) -> w_out(nQx,nQy)

Syntax:

w2 = remove_zeros(w1)

Removes points with y=0 and e=-1 from a 1-d workspace.

DIMENSIONS: 
        1d workspace only

Syntax:  

w5 = slab(w5, angle = <angle>, muR = <muR>, Tmin = <Tmin>)

For D7 data only - (useful for slab geometry single crystal omega scans).  Corrects for sample attenuation of the scattered neutrons assuming  infinite slab geometry.

ARGUMENTS: 
         angle :angle in degrees of sample normal rel. to incident beam at omega=0.  Counterclockwise is +ve 
         muR :muR is Nsigmat where sample transmission=EXP(-Nsigmat)  N*sigma=0.037 mm^-1 for vanadium 
         Tmin :Minimum transmission of sample (defined by real thickness and length)

DIMENSIONS 
 w_in(32,nphases,nruns) -> w_out(32,nphases,nruns)

Syntax: 

w10 = theta_scan(w9[, /all_angles][, /pos_angles][, /neg_angles][, detector = <det>][, /coder])

(optional keywords/arguments shown in square brackets)

Puts all detector angles onto a common x-axis. Removes zeroed data points.

ARGUMENTS: 
         detector :optional argument specifying an individual detector to use

KEYWORDS: 
         /neg_angles :only use data with negative angles 
         /pos_angles :only use data with positive angles 
         /all_angles :use all detector banks (default) 
         /coder :only used if an individual detector is specified 
             If set: 
                     use coder angle as x-axis, 
             otherwise: 
                     use calculated detector angle as x-axis

DIMENSIONS: 
        input format: w(nspectra,nphases,nruns) - unless nphases or nruns is 1 
        output: w(nruns*nspectra,nphases) - unless nphases is 1

Syntax: 

w10 = unfold(w9)

Reproduces one quadrant of the Qx-Qy plane (from fold.pro or qrebin.pro) into all four quadrants.

DIMENSIONS: 
         w_in(nQ,nQ) -> w_out(2*nQ,2*nQ)

Syntax:

w5 = vanadium(w4[, /nosave])

(optional keyword shown in square brackets)

Takes 2-D or 3-D workspace containing one or more normalised non-tof z-PA or xyz-PA background-subtracted vanadium runs. For quartz and multiple scattering corrections, data should be run through corr_xyz first. Vanadium.pro adds the flip and non-flip intensities and puts the sums into w_out and also into the file 'vanadium_<numor>.dat'

KEYWORDS: 
         /nosave :Don't save in file format

DIMENSIONS: 
     input format: w_in(32,nphases,nruns) 
     output format: w_out(32,nruns) (saved in <nrun> individual files)

Syntax: 

w2 = zero_detectors(w1, bad1[, bad2][, bad3][, bad4][, bad5][, /angle])

(optional keywords/arguments shown in square brackets)

Zeroes non-contributing spectra or angular regions - (I=0, dI=-1) Data should already have every other detector removed (i.e. must have been put through normalise.pro).

ARGUMENTS: 
         bad1, bad2, ... :Can either be 
                 a) detector number arrays 
                 b) ranges in scattering angle 
                (at least one array - bad1 - is required)

KEYWORD: 
         /angle :specifies that angular ranges are used

DIMENSIONS 
         w_in=w_out(nspectra,nphases,nruns)

2.1. average.pro

Syntax:  

w3 = average(w1, w2)

Takes the weighted average of two workspaces, w1 and w2.

Syntax: 

w3 = bin(w1, ibin)

Bins the points in the x-axis of a 1-d workspace, w1 by an integer multiple, ibin.

Syntax: 

w3 = divide(w1, w2)

Divides workspace w1 by workspace w2.   Ensures proper error propagation on division.

Syntax: 

w3 = fit_poly(w2, degree = <npoly>)

Fits a polynomial of degree npoly in the x-dimension to one- or  two-dimensional data

ARGUMENTS: 
         degree :degree of polynomial to fit the data with

DIMENSIONS: 
         1-d or 2d data

Syntax: 

w1 = input(file = '<in_file>')

Reads a file previously saved by output.pro (with titles), or any XYE formatted file.

ARGUMENTS 
         file :name of file to input 
                 NB - don't forget to include single quotes

Syntax: 

output, w1, file = 'myfile.dat' <,/params>

Takes a workspace w1 and writes it to the file, "myfile.dat". 1-d workspaces are saved in a standard (x,y,e) 3-column format. Larger workspaces are saved differently.

ARGUMENTS
          file: name of file to output

KEYWORDS
          params:  specifies whether to output the run parameters with the data

Syntax:

output_gsas, w1, file = 'outfile'

Takes a 1-d workspace an writes it to a GSAS format data file.  Since GSAS requires constant angular steps, binq.pro should be run on the data first.

Syntax:  

w2 = remove(w1, min = <xmin>, max = <xmax>)

Removes points from 1-d data within x-region specified by xmin, xmax.

ARGUMENTS: 
         min : min of x-region 
         max : max of x-region

DIMENSIONS 
         1-d data only

Syntax:

w2 = smoothx(w1, nsmooth)

Smoothes data using a moving filter in the x-dimension of one- or two-dimensional data.  Smoothes over -nsmooth to +nsmooth

Syntax: 

w2 = smoothy(w1, nsmooth)

Smoothes data using a moving filter in the y-dimension of  two-dimensional data.  Smoothes over -nsmooth to +nsmooth

Syntax: 

w3 = splice(w1, w2, min = <min>, max = <max>)

Splices together two data sets together. Takes w_in1 outside the  xmin->xmax region and w_in2 inside.

ARGUMENTS: 
         min :min of w_in2 region 
         max :max of w_in2 region

DIMENSIONS 
         must be 2 1-d arrays input

3.1. corr_tof.pro

Syntax: 

w6 = corr_tof(w5[, /det_eff][, /frameoverlap][, /ana_eff])

(optional keywords shown in square brackets)

KEYWORDS: 
         /det_eff : corrects for energy-variation of detector efficiency 
         /frameoverlap: subtracts a t^-4 tail from the beginning of the time frame 
         /ana_eff : correct for energy dependence of analyser transmission

DIMENSIONS 
         w_in=w_out(nchannels,nphases*nspectra)

Syntax: 

w3 = elastic(w2[, min = <min>, max = <max>][, /save])

(optional arguments and keywords shown in square brackets)

Finds the elastic peak position in each spectrum by fitting a Gaussian and a flat background. Looks for peak in region xmin<x<xmax

ARGUMENTS: <br />          min: minimum x-value for peak search 
         max: maximum x-value for peak search

KEYWORDS: 
         /save : Write out results to file 'elastic.dat'

DIMENSIONS: 
         w_in(nchannels,nspectra) OR w_in(nchannels)

Syntax: 

w10 = estrip(w9, E = <Evalue>, dE = <dE>, dQ = <dQ>[, /all_angles][, /pos_angles][, /neg_angles])

(Optional keywords shown in square brackets)

Rebins output data from t2e.pro and reb.pro to regular-grid S(Q,w) data using the old KHA IN6 rebin algorithm. Selects a single E-value. Input workspace must be in energy transfer versus scattering angle, i.e. only one component or spin phase.

ARGUMENTS: 
         E: E-value at centre of extracted strip 
         dE : Width of E-strip 
         dQ : Q bin width

KEYWORDS: 
         /neg_angles : use only negative angles 
         /pos_angles : use only positive angles 
         /all_angles : use all angles (default)

DIMENSIONS: 
         w_in(nE,nphi) -> w_out(nE,nQ)

Syntax: 

w2 = normalise(w1[, /raw][, /monitor][, /time][, detector = <det>][, /all][, /madangles][, zeroshift = <zeroshift>])

(optional keywords/arguments shown in square brackets)

Normalises raw data to monitor or counting time. Extracts every other spectrum if desired(=> nspectra=32) and normalises intensity

ARGUMENTS 
         detector : detector number to normalise to 
         zeroshift : angle in degrees of 2theta=0     (depends on wavelength and setup)

KEYWORDS 
         /raw : no normlisation (error bars calculated) 
         /monitor : normalise data to 1000 monitor1 counts (DEFAULT) 
         /time : normalise data to counting time 
         /all : supresses the removal of the odd numbered detectors 
         /madangles : supresses the recalculation of the detector angles using the current YIG calibration

DIMENSIONS 
         non-TOF data: w_out(nspectra,nphases,nruns) -unless nphases is 1 
         TOF data: w_out(nchannels,nspectra*nphases,nruns)

Syntax: 

output_allangles, w17, file = 'outfile'

Takes a 2-D workspace from t2e.pro and writes it to a series of files, one for each detector. x-axis must be energy transfer and y-axis detector number. 

Output files have names: 
<file>.det2 for detector #2, 
<file>.det4 for detector #4,  
etc...

ARGUMENTS 
         file : name of output file stem

DIMENSIONS 
         w_in(nspectra,dE) output from t2e.pro [and reb.pro]

Syntax: 

output_allqs, w17, file = 'outfile'

Takes a 2-d workspace from sqw_rebin.pro and writes it to a series of files, one for each Q. x-axis must be Q and y-axis energy transfer.

Output files have names:    
<outfile>.q010  for  Q=0.10 Å-1
<outfile>.q010  for  Q=0.10 Å-1 
etc...

ARGUMENTS 
         file : name of output file stem

DIMSENSIONS 
         w_in(dQ,dE) output from sqw_rebin.pro

Syntax: 

w10 = qstrip(w9, Q = <Qvalue>, dQ = <dQ>[, /all_angles][, /pos_angles][, /neg_angles])

(optional keywords shown in square brackets)

Rebins output data from t2e.pro and reb.pro to regular-grid S(Q,w) data using the old KHA IN6 rebin algorithm. Selects a single Q-value. Input workspace must be in energy transfer versus scattering angle, i.e. only one component or spin phase.

ARGUMENS: 
         Qvalue: Q-value at centre of extracted strip 
         dQ : Width of Q-strip

KEYWORDS 
         /neg_angles : use only negative angles 
         /pos_angles : use only positive angles 
         /all_angles : use all angles (default)

DIMENSIONS: 
         w_in(nE,nphi) -> w_out(nE)

Syntax: 

w10 = reb(w8, dE = <dE>[, /force])

(optional keywords shown in square brackets)

Rebins output workspace from t2e.pro to constant energy bin width.

ARGUMENTS: 
         dE : required energy bin width

KEYWORDS: 
         /force : rebin over entire energy range otherwise: rebin only where dE is > point spacing (default)

DIMENSIONS 
         w_in=(nE,nspectra) -> w_out(dE, nspectra)

Syntax: 

w7 = slab_tof(w6, angle = <angle>, inc = <inc>, abs = <abs>, thick = <thick>, N = <N>)

Takes a 2-D tof workspace and corrects for sample attenuation of the scattered neutrons assuming infinite slab geometry.

ARGUMENTS 
         angle : the angle in degrees of sample normal relative to incident beam 
         inc: incoherent cross-section of sample (barns) 
         abs: absorption cross-section for thermal neutrons (lambda=1.8Å) (barns) 
         thick : thickness of sample (mm) 
         N : number density of sample (*10^22 per cm^3)

DIMENSIONS 
         w_in(nchannels,nspectra) -> w_out(nchannels,nspectra)

Syntax: 

w10 = sqw_interp(w9, dQ = <dQ>, dE = <dE>, Emin = <emin>[, /all_angles][, /pos_angles][, /neg_angles])

(optional keywords shown in square brackets)

Converts output data from t2e.pro to regular-grid interpolated S(Q,w) data. Input workspace must be in energy transfer versus scattering angle.

ARGUMENTS: 
         dQ : Q bin width 
         dE : energy bin width 
         Emin : minimum energy (neutron energy gain is defined as negative)

KEYWORDS 
         /neg_angles : use only negative angles 
         /pos_angles : use only positive angles 
         /all_angles : use all angles (default)

DIMENSIONS: 
 w_in(nE,nphi) -> w_out(nQs,nEs)

Syntax: 

w10 = sqw_rebin(w9, dQ = <dQ>, Emin = <emin>[, /all_angles][, /pos_angles][, /neg_angles])

(optional keywords shown in square brackets)

Rebins output data from t2e.pro and reb.pro to regular-grid S(Q,w) data using the old KHA IN6 rebin algorithm. Proper rebinning routine with error analysis (unlike sqw_interp.pro).

ARGUMENTS: 
         dQ : Q bin width 
         Emin0: Minimum energy value (meV) - neutron energy gain is negative

KEYWORDS: 
         /neg_angles : use only negative angles 
         /pos_angles : use only positive angles 
         /all_angles : use all angles (default)

DIMENSIONS: 
         w_in(nE,nphi) -> w_out(nQs,nEs)

Syntax: 

w10 = t2e(w9[, /ken])

Converts tof data in w_in to energy transfer in w_out. Includes the factor of ki/kf.

KEYWORDS: 
         /ken : use Ken's expression for dEdt in t2e_tof.pro (not recommended)

DIMENSIONS: 
         w_in(nchannels,nspectra) -> w_out(nE,spectra)

Syntax: 

w8 = vnorm(w7, w20, min = <min>, max = <max>[, tv = <tv>, Ns = <Ns>, ts = <ts>])

(optional arguments shown in square brackets)

Normalises data in w7 vanadium data in w20 integrated between given limits.  Absolute normalisation is possible if desired - although it this assumes flat-plate geometry.

ARGUMENTS 
         min : lower time channel limit of integration over vanadium elastic peak 
         max : upper time channel limit 
         tv : thickness of vanadium sample 
         Ns : number density of sample 
         ts : effective sample thickness

DIMENSIONS 
         w_in=w_out(nchannels, nspectra)