From:	Lawrence E. Dudek
Sent:	Thursday, May 26, 2011 4:49 PM
To:	NSTX Center Stack Upgrade Team
Cc:	Tiana Dodson
Subject:	Notes from 5/25/2011 NSTX CSU Meeting

Notes from 5/25/2011 Meeting:
Check of Disruption Calcs for NSTXU Passive Plates (Menard)
 (see VGs)
Menard looked at a representative 700kA shot # 140587 from NSTX which as similar to the P1-
P4 GRD VDE approximation.  His calculations closely match the loads and currents calculated 
by Titus.  However it was noted the GRD VDE plasma equilibria are not self-consistent which 
(assuming worst case PF) might be overly conservative.  This analysis also does not take into 
account the skin depth effects (approx 2-3 mm for Cu) which may reduce the net currents and 
loads.

A discussion with the Columbia University group revealed a change to SS for the PP would 
increase RWM growth rate by about 1 order of magnitude (or more) which is unacceptable. This 
was based on calcs done for NSTX 5 year plan and KSTAR.

PP 3D Opera Modeling  (Zhai) (see VGs)
Yuhu has modeled the Passive Plates, and brackets thru to the VV connections.  His PP model 
includes elements through the thickness so current skin effects can be detected.  His analysis 
 confirms the presence of the skin effects and show a significantly different eddy current pattern 
when compated to Titus and Menard.  The model shows current running through the thickness of 
the plates with significant canceling effects.  The next step is to extract the loads from the model 
to compare to the Titus fast disruption model.

TF Temps, LLD Disruption Load Comparison, Lower Secondary Passive Plate Disruption 
with Support Struts (Titus) (see VGs)
A peer review chit suggested we could shorten the TF current pulse slightly to prevent the copper 
from exceeding 100C.  Hans analysis shows the lower and upper TF reaches 113C and 110C 
respectively at EOFT 
ACTION: Jim C to have the CTD 425 resin tested at 115C to determine if there is a problem.

Pete modeled the LLD plates in his ANSYS model to compare the relatively light calculated 
LLD loads to the much larger PP loads.  The reactions at the LLD plate were calculated to be on 
the order of a couple hundred pounds as predicted by Brooks.  Using the NSTXU plasma 
disruption currents the loads increase to approximately 3000 lbs, similar to what is seen on the 
passive plates.

Also ran the PP with Gap elements to simulate a column stop connection to the VV which would 
support the PP in the outward direction.   Analysis indicates they are ineffective. 
FDR Prep (Strykowsky) (see VGs)
Ron provided an update of the agenda and the charge to review what will be needed for the 
Final Design Review.

 
OH Coil lead Area and Embedded G10 (Mardenfeld) (see VGs)
ANSYS model indicates stresses in the copper exceed the fatigue allowable 
(120 Mpa) where the leads exit the OH coil body.  Shear stresses between the 
CTD101K and Cu exceed the strength allowable of 7MPa.  Suggest adding 
some compliance in the area to reduce stresses such as rubber "grommet"  
ACTION: Jim C. to brainstorm with Titus to develop solution.

OTF Support Ring (Rogoff) (see VGs)
ANSYS analysis indicates the outer TF ring looks robust, the extended 
supports on the outer leg are ineffective as demonstrated in Han Zhang 
analysis.  Inside clevis pin on the outer leg bracket is over stressed and will 
need reinforcement to reduce stresses to allowable levels.  Simplest solution 
may be to change pin material and to add reinforcement to the clevis plate to 
increase capacity.   

http://nstx-upgrade.pppl.gov/Project_Meetings/CenterStackUpgrade/index_Meetings.htm