Engineering Group Conference Call

10 September 1997
11:30 Am Central Time
call-in number: 314-232-8169


Participants:
Ron Miller (UCSD), Xueren Wang (UCSD), Mark Tillack (UCSD), Laila El-Guebaly (UW), Don Steiner (RPI), Les Waganer (Boeing), Wayne Reiersen (PPPL)

AGENDA

  1. Status of systems design space (Miller)
  2. Design Assessments
    1. CP design options, issues and plans (Reiersen)
    2. Power core configuration design options, issues and plans (Wang)
    3. Blanket design options, issues and plans
    4. Shield design options, issues and plans (El-Guebaly)
    5. Divertor design options, issues and plans (Tillack)
  3. Other topics
    1. Neutronics
    2. Safety
    3. Materials

Preliminary discussions

Miller indicated that the EPRI utility requirements document is available for $2500. He is negotiating a lower price, since it would be nice for the team to have at least one copy.

The last UAC meeting was Sept. 1996. The question arose, should we schedule another one? No concrete plan emerged. The report from the last meeting still has not been released.

Centerpost and TF coils

Lengthy discussion of the CP and coil issues took place. There is a need to verify the IB VV thickness requirements for acceptable stresses in order to determine the IB radial build.

Miller noted criticism at the FPA meeting in Snowmass ragarding the lack of serious stress analysis of the GA LAR design and other "cartoonish" features.

Tillack reviewed a discussion with Bromberg, who is performing some coil stress analysis. Reiersen explained that Dahlgren is also examining TF coil stresses. Initial results for a constant-tension outboard coil leg suggests that the elongated LAR plasma fits nicely with a constant-tension shape. The coil follows the blanket shape well.

Reiersen also remarked that the material stress allowables for low-T and high-T operation are being used to determine the attractiveness of the two major options. He does not plan on considering cyclic behavior in the analysis.

Tillack asked if the busbar and joint problems have been adequately addressed, or if further work is needed. Reiersen answered that we can do a normal bus design with only 10's of MW of dissipated power, so this is not a big issue.

Some discussion ensued on the IB radial build, and specifically the CP, FW and VV configuration and location of the vacuum break.

Miller described the previous use of CP flaring to minimize resistive losses. Reiersen noted that flaring is not really possible until the divertor location due to high elongation and triangularity. Miller will compare flared and non-flared options for the reference design to understand the penalty of a cylindrical CP.

Reiersen pointed out a strong incentive to decouple the power core and magnet. the temperatures will be different, and differential deflections will cause all sorts of problems. Miller added that a gap will be needed when these elements are decoupled.

Tillack asked if we can feed back engineering requirements on the CP shape and resistive losses and ask the physicists to change the equilibrium. Reiersen replied that we already pushed them to reduce the TF coil current, and that probably this would be a losing battle.

The configuration questions all depend on the maintenance scheme. Reiersen mentioned that a vertical maintenance scheme would imply a different coil shape than the one currently being examined.

Systems Studies

Miller noted that the A= 1.25, 1.4, and 1.6 physics cases are in hand. the A=1.8 and 2.0 cases are expected, but it's not clear they will be ready by the next project meeting. The case A=1.6 is the best so far, due in part to extra room inboard which significantly lowers the coil current density without reducing beta too much.

Reiersen asked is it possible to generate a 1.6 variant with a cylindrical CP? Yes, but it will take some time.

Steiner asked if Ron's cases assume a water-cooled CP. The answer is yes, but other options are still possible. Steiner indicated that we have stopped work on the IBC, but a warm CP is still an option.

Reiersen suggested that in the near term we should understand what the candidates are. For example, for a water-cooled CP, is hot a Li blanket OK? Tillack answered: NO. There is no one remaining on the team who appears willing to advocate that position. Miller said that we need to document this.

Reiersen said that the system runs should be self-consistent. We need to define the power cycle. Miller added that the code allows a separate divertor cycle, since this issue of different coolant conditions has come up before.

Tillack explained that the two blanket cycles considered are the supercritical Rankine cycle and He Brayton cycle. Either of these is considered compatible in principle with any set of blanket materials, although the optimum choice depends on the details. These high-performance cycles require high coolant temperatures, and do not benefit much from "low-grade" heat.

Reiersen mentioned that CuCrZr will operate at or below 300 C max. Al-15 could in principle get to 350 or even higher, depending on irradiation effects.

Steiner reiterated that we need to consider both lower and upper temperatures. Reiersen asked if we may want to use a separate cycle for the CP. Steiner felt probably not. Reiersen added that we can't say if Cu can go to 350 or not. It looks tough. We may need to limit the peak temperature to 300 C.

Tillack said that we should get some feedback from the materials people on how high we can operate different copper alloys. Reiersen suggested we wait a bit before contacting them, so that we obtain info in the context of a design point.

Steiner asked if we can use PbLi in the CP? Reiersen answered that the temperature drops in the Cu, SS clad, insulator and film is at least ~30 C from the peak Cu temperature to the coolant. If the limit is 300, then there's no margin to the melting point. We really would need a material we can run significantly higher than 300 C.

Steiner asked if we have a SS sleeve on the coolant channels anyways, could the SS take the stresses and fracture toughness? Reiersen answered that cladding may eliminate the coolant leakage concern, but cracks would propagate quickly and degrade the integrity of the CP. A SS sleeve has to be thin to allow heat to pass. Structurally it has no strength.

Steiner asked what mechanical load does the Cu really carry. Reiersen answered if we need to add structure, then the Cu packing fraction goes down. With a hexagonal pack, the packing fraction is ~80%.

Steiner asked if we need to evaluate this tradeoff.

Back to discussion of the system studies, Miller mentioned that he sent the poloidal distribution of wall load to El-Guebaly. He may need to recalculate it for the A=1.6 case. The issue of economy of scale (>1000 MW net) was mentioned. Since fission people are pushing higher, we should be able to do it. But it implies higher power densities, which may be difficult to accommodate.

Waganer noted that in Prometheus we did the design at 1000 MWe but used the system code to look at parametrics. That approach could be used here.

Configuration and Maintenance

Wang reviewed the 3 options being looked at.

  1. Vertical CP removal from bottom. Radial Motion of the TF coils requires a large space around machine. We would need to move the PF coils also.

  2. Vertical maintenance from below. A primary concern is that we would need an outboard demountable joint.

  3. ARIES-RS type. This configuration is understood.

Tillack asked about shielding requirements to allow rewelding of the VV. El- Guebaly indicated that ~1 m of shield is needed to allow rewelding through lifetime.

Reiersen said we need to consider the radiating mantle and heat flux limits on the FW. Tillack asked for the FW surface area so that these simple calculations could be done. Reiersen said that a simple steel FW looks appealing.

Shield

El-Guebaly will design a shield for both Li and PbLi blankets. Without SiC we could breed with the PbLi blanket. With the 1-cm SiC inserts proposed in the original dual-blanket concept definition, we would need to breed on the inboard. He or water-cooled inboard first walls are considered. For the Li/V option, we assume the IB shield and VV would be Li cooled.

Tillack asked what is the difference in inboard shielding thickness between He and water coolants. El-Guebaly said ~5 cm.

NEUTRONICS

At A=1.6, the peak wall load is ~10 MW/m2. El-Guebaly wondered if this is a problem. Steiner answered: Yes.

Reiersen asked how does ASC treat the blanket changeout frequency? Miller explained that replacement cost is included, but currently the plant capacity factor is not adjusted. It could be. [improving the available model in ASC is an action item]. Reiersen mentioned this may be a large effect.

Miller continued, at 75% capacity factor, there are 90 days per year of downtime. This comes from 60 days unscheduled and 30 days scheduled.

Tillack asked where does 60 days come from? Miller answered: we might be able to lower this.

SAFETY

Mogahed is including the coil leads (5x the CP area) in the LOCA thermal analysis. Reiersen suggested to use a concentric connection to the CP at the bottom, with branching as many leads as needed. It's safe to say that the cross sectional area would be much larger.

Miller noted that the issue of top or bottom placement of leads affects the maintenance scheme (and vice-versa).

DIVERTOR

Tillack indicated that our initial thermal analysis of the liquid film and droplet concepts was relatively pessimistic. Peak heat flux of only 520 MW/m2 are obtained using Liao's maximum permissible edge density. More detailed analyses will be presented at the project meeting.

Reiersen asked if it's conceivable that with a radiating mantle the divertor heat loads would be low enough? Tillack indicated that some balance of power flow to both the FW and divertor probably will be needed, as the first wall most likely would not be able to absorb all of the transport and radiation power.