TJ-II:Magnetic coordinates

Revision as of 09:27, 23 July 2009 by Admin (talk | contribs)

The diagnostics perform measurements in real space. The location of points is given in one of these co-ordinate systems (units according to the S.I. system, m and rad):

  1. Cartesian (X, Y, Z) with its origin at the point of gravity of the TJ-II device, the X-axis pointing due East, the Y-axis due North, and the Z-axis up.
  2. Cylindrical (R, φ, Z), where R2 = X2 + Y2 and tan φ = Y/X.

In order to make comparisons between diagnostics, it is useful to convert these real-space co-ordinates to flux co-ordinates. This co-ordinate transform depends on the particular magnetic configuration used in a given experiment. Two tools are available to do so (See the on-line documentation - only internal laboratory access):

First, vacuum equilibrium calculations from VMEC. These are then used to obtain magnetic flux co-ordinates (ψ, θ, φ). A set of routines is available to perform the corresponding co-ordinate transforms.[1] The drawback of the VMEC calculations is (a) that magnetic islands are ignored, and (b) that only a limited number of configurations is available.

Second, magnetic field line calculations using the Biot-Savart Law. The approximate magnetic flux is recovered from an interpolation procedure. A set of routines is available to perform the corresponding co-ordinate transforms. More information can be found in files g3d_readme.doc and g3d_gridfile.doc. Since the latter procedure is more flexible and generally applicable than the VMEC-based calculations, the latter is preferred.

It should be noted that these co-ordinate transforms are approximate and not error-free. The errors in the vacuum field calculation are due to three sources:

  1. The error in the placement of the coils.
  2. The error in the value of the currents flowing through the coils.
  3. The error due to the fact that the model uses a finite amount of filaments to model the current in the coils (which has a continuous distribution).

Additional errors are due to finite-pressure effects (estimated to be quite small in TJ-II) and net plasma currents.

References