ISG CSV format

Description

The simplest and most flexible file format is a tabular ISG CSV format.

In terms of its structure it is very similar to an Excel spreadsheet.

The columns are named in the header. The header can contain both argument and parameter names.
The actual records are specified in the rows below.
Row comments start by a double slash "//".
A "file break" can be created with a hash "#". After this break, a new header and the associated records can be specified.

Programming Example

Machine position errors

// The header only contains position errors.

Y0A Z0A X0B …

// Position errors are typically (but not necessarily)

// constant. Therefore, one data row is usually sufficient here.

0.1 0.2 0.11 …

Programming Example

Component errors of the X axis

// The header contains the argument "X" and component errors.

X EXX EYX EZX EAX EBX ECX

0.0 0.02 0.01 0.07 0.00 0.0042 0.003

10.0 0.00 -0.1 0.01 0.00 0.0011 0.002

…

// Further records depending on X

Extended functional dependencies

Any functional dependencies can be implemented in the ISG CSV format. For example, it is conceivable that two data blocks for component errors in the X axis, one each for Z = 0 and one for Z = 100, must be specified due to sag effects on the machine. Parameters are interpolated as described in Section “File formats supported”.

Any higher-dimensional dependencies can be implemented with this method.

Programming Example

Component errors of the X axis for Z = 0 and Z = 100

// The argument "Z" now crops up in the header.

Z X EXX EYX EZX EAX EBX ECX

// In the first block Z is always = 0.

0.0 0.0 0.02 0.01 0.07 0.00 0.0042 0.003

0.0 10.0 0.00 -0.1 0.01 0.00 0.0011 0.002

…

// In the second block Z is always = 100.

100.0 0.0 0.01 0.02 0.03 0.07 0.0040 0.003

100.0 10.0 0.03 -0.09 0.03 0.06 0.0015 0.000

…