PF coils

[jonmaddock]

In GitLab by @mkovari on Nov 11, 2014, 12:27

Tidy up inconsistencies.
Change default to NbTi.
Remove "steel thickness" column.
Remove 2/3 factor from stress calc.
Max permissible tensile stress same default as for TF and CS.
No need for temp margin.
Current density safety factor (input).
Calculate actual J.
Derive sizes incl. steel.

[jonmaddock]

How much of this is already done, or no longer relevant? Perhaps it can be finished, or split up into the remaining tasks?

[ajpearcey]

Tidy up inconsistencies.
Change default to NbTi. - dont change
Remove "steel thickness" column. - remove as there should only be a conduit
Remove 2/3 factor from stress calc. - has been replaced by sigpfcf which is set to 2/3. will be removed.
Max permissible tensile stress same default as for TF and CS. - now TF and CS stress are separate - will be same as CS
No need for temp margin. YES!
Current density safety factor (input). - rjconpf - check for safety factor and how it's calculated
Calculate actual J.
Derive sizes incl. steel. - add conduit

[ajpearcey]

Going through an making sense of what is in the PF coils models

I will leave the default PF coil conductor material as isumatpf=1 which is the ITER Nb3Sn critical surface model.

The model assumes that all the steel in the PF coil is in a case that encloses the PF coil winding pack and its thickness is determined by a calculation of the thickness required to keep the stress below sigpfcalw (defaultly 500MPa). The model used to assume the 2/3 of the force was supported by the steel. The steel thickness is used is other areas of the code to calculate things like weights and costs. Something similar is done in the CS coil, but with the comment

Thickness of hypothetical steel cylinders assumed to encase the PF along
its inside and outside edges; in reality, the steel is distributed
throughout the conductor.

The factor 2/3 will be replaced with a factor 1. The steel requirement calculation based on the stress limits will be left but a comment will be added to reflect that this steel is spread across the PF coil. This steel thickness will be removed from the output, while PF coil steel mass will be left.

There is no safety factor for the current density in the winding pack. This could be added, thoughts? @mkovari

[mkovari]

1. We definitely need a user-specified margin below the critical current density. Since the PF coils are exposed to large field variations resulting in AC losses (as is the CS), this needs to be fairly substantial - maybe a factor of 2 by default? We need to describe these parameters fairly carefully -
   "Critical conductor current density based on short-sample DC measurements"
   "Ratio of permissible conductor current density to critical conductor current density based on short-sample DC measurements"

2. Does the code include a copper stabiliser?

@ajpearcey