NCSX Engineering meeting topics

December 6 2000

1. Cost estimate

The current cost estimate for the modular coil configuration is $66.7M. This includes a core cost of $34.6M. The modular coil set cost is $21.7M, or $1M per coil. The cost needs to be updated, but for the core there needs to be more design work in the areas of machine structure, cryostat, PF and TF coils. Costs for the modular coils themselves are in the process of being refined. Dave Williamson and Brad Nelson will meet with the representative of New England Wire today, and Brad talked to Mike Tupper of CTD about possible revisions to the potting scheme.

A kickoff meeting will be held this Friday to begin a series of cost estimate reviews. The review schedule is:

2. Modular Coil design

Work on the “clocked” coil set has been stopped, and all efforts have returned to the 1017 coil set. A few troublesome regions of coil twist and curvature are being tweaked to avoid later problems in analysis and to allow “good” stereolithography models to be constructed. A better shell surface is also being constructed that should be more robust to later manipulation by Pro-E and to possible improve the ANSYS modeling.

HM Fan and Dave have been discussing the coil system analysis, and are passing models back and forth.

Since the coils are cooled to LN2 temperatures and have a relatively small temperature rise during a pulse, we may now have the option of potting the cable conductor. There are heat transfer and stress implications of this. In discussions with Mike Tupper, it looks like we may be able to obtain some potted cable samples for compression testing near term for a few $k.

3. Machine structure and facility interfaces

This area needs attention and Tom Brown will begin working to define the machine gravity supports and their interfaces with the stellarator core and foundation. The elevation and plan locations of the machine in the enclosure will be defined so the length of bus runs, coolant piping, etc. can be estimated.

4. Cryostat

The cryostat envelope must be defined in more detail, including its interface with the machine structure. The basic concept is a frame and panel design similar to what is proposed for FIRE and what was proposed for BPX.

5. PF coils

Wayne has modified the PF coil number, size, and location. The new coils have currents of up to 20kA/turn for a 1 x 1 inch turn. The current density is around 3000 A/cm^2, and the coil cross section has increased somewhat from the original design. With the new information it should be possible to make some cost estimating packages to send to Everson for ROM costs.

6. TF coils

The TF coil concept consists of a pair of multi-turn windings attached to both sides of the radial structure plates. These plates help register the modular coil shell segments and provide a mounting surface for the TF and PF coils. It appears that a 3 turn winding of hollow copper conductor on each side of the plate would be adequate, requiring only about 14kA/turn. As with the PF coils, a cost estimating package for the TF coils will be prepared and sent out to Everson to obtain some ROM cost.

7. Design Integration

The primary effort in design integration remains accommodating all the access requirements. In addition, work has begun to investigate the possibility of using TFTR or JT-60 beamlines in place of the PBX beamlines.

Another large effort has been to enter all the baseline Pro-E models into the Intralink database at PPPL.