TF coil geometry not well documented and not very precise #603

Open jonmaddock opened 6 years ago

In GitLab by @mkovari on Jan 24, 2018, 09:08

The html output contains some very useful images.

[ ] 1. The image TFWindingPack.png has some discrepancies, which make it impossible to understand the TF coil geometry precisely:

Variables radwp, bcylth, tcan1, tcan2 and tcan3 are shown but do not exist.
Features and variables not shown:

thkcas: (inboard TF coil case outer (non-plasma side) thickness) 
casthi: inboard TF coil case plasma side thickness
casths: inboard TF coil sidewall case thickness 
thicndut: conduit insulation thickness (m) 
tfinsgap: TF coil WP insertion gap
thwcndut: TF coil conduit case thickness (m) (iteration variable 58)
tinstf: ground insulation thickness surrounding winding pack (m) 
rbmax : radius of maximum TF B-field (m)

[x] 2. The radial build for the centre-line of the TF coil should be included.
[ ] 3. It would be very nice to plot the TF winding pack diagram to scale. This would certainly help to identify errors and apparent discrepancies with results from other codes, and to liaise with the CAD designers.
[x] 4. This description is confusing:

bmaxtfrp : peak field at conductor with ripple (T)

as most people use the term ripple only for the outboard side. Perhaps we should write

bmaxtfrp : peak field at discrete cable (radius rbmax) (T)

[ ] 5. The peak field is calculated at radius rbmax, given by
```
rbmax = rcoil - casthi
```
This is not very accurate. The actual winding array doesn't include the insertion gap tfinsgap and the ground insulation thickness surrounding winding pack tinstf.
[ ] There is a discrepancy in the radial build of the TF coil, equal to twice the ground insulation thickness.

In GitLab by @mkovari on Feb 5, 2018, 14:46

marked the task 4. This description is confusing: as completed

In GitLab by @mkovari on Feb 5, 2018, 14:46

marked the task 2. The radial build for the centre-line of the TF coil be included as completed

In GitLab by @jmorris-uk on Feb 14, 2018, 14:21

mentioned in commit 62742b19fdb5f9cc97f816d75395fd1eaa69366f

In GitLab by @skahn on Nov 28, 2019, 15:57

mentioned in commit dca25082c3f1ece6f096e5a32e461dfa7d2bf592

In GitLab by @skahn on Nov 28, 2019, 15:58

Point 6 is related to the fact that r_wp_inner calculation does not take into account the insulation and insertion gap thickness

r_wp_inner = r_tf_inleg_in + thkcas

while the WP thickness does :

thkwp = tfcth - casthi - thkcas - 2.0D0*tinstf - 2.0d0*tfinsgap

The outer WP radius being calculated with thkwp, the discrepancy appears

r_wp_outer = r_wp_inner + thkwp

Point 5 and 6 are being solved on branch https://git.ccfe.ac.uk/process/process/tree/issue-'Reference to deleted milestone 23602'

Best regards
Seb

In GitLab by @jmorris-uk on Dec 13, 2019, 12:24

mentioned in commit f52441588e0b9653b40f1e3e738f1058dc877bae

As I am currently re-writing the docs I will put this down as a point of reference when I get to the TF section

ukaea / PROCESS

TF coil geometry not well documented and not very precise #603