Engineering Group Conference Call

20 August 1997
11:30 Am Central Time
call-in number: 314-232-8169


Participants:
Dai Kai Sze (ANL), Don Steiner (RPI), Jake Blanchard (UW), Hesham Khater (UW), Mark Tillack (UCSD), Fred Dahlgren (PPPL), Les Waganer (Boeing), Ron Miller (UCSD)

AGENDA

  1. Review of ENG goals, preparation for next project meeting
  2. Design Assessments
    1. CP design options, issues and plans
    2. Power core configuration design options, issues and plans
    3. Blanket design options, issues and plans
    4. Divertor design options, issues and plans
  3. Other topics
    1. Neutronics
    2. Safety
    3. Materials
  4. Status of systems design space

Preliminary remarks

The next project meeting is Sept. 25-26 in PPPL. A Project call is scheduled for Sept. 3, and one more ENG call for Sept. 10.

The purpose of this call is to report on the status of activities and plan for the period of work up to that meeting. The assessments should be mostly finished by then, and hopefully we can agree on the ENG strategy before the project meeting. In addition, we need to start considering documentation needs before moving to a design phase.

The thrust of work this year has been on:

  1. exploration of design options and novel design solutions
  2. analysis and assessment of options
  3. description of key issues

This must lead to a process of narrowing the design options to one or two. For example, we could continue to maintain a Li/V and PbLi/SS blanket, but the system code runs require choosing one.

We anticipate assessments in 5 design areas:

  1. CP internal design options (Reiersen)
  2. power core configuration/maintenance options (Najmabadi, Wang)
  3. divertor options (Tillack, Zhou)
  4. blanket options (Sze)
  5. IB shielding options (El-Guebaly)

Beyond that, we have activities in neutronics, safety and materials which cross these component boundaries.

Design Assessments

General Discussion

Steiner asked, how will we choose design concepts?

Waganer felt that discovery of critical flaws may make the decision, for example, the inability to identify a high-temperature HX for the PbLi dual-coolant design. Sze suggested that a design change (to He-cooled, stagnant PbLi breeder) could make this design more feasible. All designs have critrical issues, so it will be difficult to use this approach effectively. Tillack remarked that he hoped we could evolve a group position throughout the next month so as to avoid a contentious debate. the decision usually ends up with significant qualitative aspects, and requires good judgement.

Discussions of the documentation of work ensued. Tillack commented that the FED special issue is in production at Elsevier (Oxford), and that it will include a substantial number of color figures. The group agreed that a detailed outline should be developed by October. Realistically, this Fall is a busy time, and we would not be likely to have a report ready until early (calendar) 1998.

Centerpost

Degradation of Cu(Al-15) fracture toughness at higher temperature due to irradiation has led to considering Cu-Cr-Zr as a better choice. Warm operation (up to 300 C) may keep the material reasonably resistant to fracture.

Power conversion issues were discussed. The expectation of providing useful heat for the power cycle from a 300 C centerpost was questioned. A similar issue arises for the divertor. Sze, Tillack and Dahlgren agreed to continue to pursue this matter. (A subsequent e-mail from Sze concluded that any heat below 250-300 C, depending on the cycle type, would be waste heat and not useable in the cycle). Dai Kai also pointed out that we need to be careful in quoting coolant temperatures, because there is usually up to a 50 C drop across heat exchangers.

Don Steiner made a case for using Li everywhere. The total mass flow in the CP probably doesn't go through the blanket and divertor, and so careful examination of the various coolant streams and their coupling to the power cycle is needed.

One of the critical unknowns for the centerpost design is PF coil locations and currents, needed to perform stress analysis. This is important for the centerpost as well as the outboard TF legs. The wedge configuration is still being examined. Tillack agreed to discuss with Miller whether we can obtain some initial numbers from the JSOLVER part of ASC.

Tillack explained that definition of the outboard TF geometry may depend on the choice of maintenance scheme. The nominal goal is a constant-tension "D-shape" with top and bottom rings to resolve radial loads, but vertical downward power core maintenance would require reconsidering the optimum shape.

Since it appears the wartm CP option is receiving greater attention and is still considered a primary design option, it is important to address the remaining issues before the project meeting: internal design details, need for insulator to eliminate massive MHD forces on the liquid, power conversion, etc.

Sze pointed out that an attempt has been made to eliminate coatings in the blanket, taking advantage of the lower field strength. Adoption of insulators in the CP brings back this critical issue for LM coolants.

Sze also suggested that electrolysis can be an issue in the CP. This depends on the ability to maintain low conductivity de-ionized water, which shoul dnot pose a problem.

Blanket

Sze summarized the activities in the blanket area. He has looked at MHD issues (trying to eliminate the coatings). Tillack commented that the new strawman looks to have VERY high power density, and so a determination of the limit of a non-insulated blanket would be useful.

A He-cooled blanket with stagnant PbLi appears to be a good variant that avoids the HX problem. The use of SiC is still advocated to enhance the outlet temperature beyond the steel temperature limit.

Tritium issues may become more serious at higher temperature (with the possible exception of Li), and this needs to be looked at.

Tillack asked if the safety attributes of design options are still considered distinguising characteristics, and what is being done. Contact has been made with the UW group doing LM chemical reaction studies (Corradini), and it is hoped they will be able to run their codes to explore various options. This work probably will not be accomplished by the time the report is expected.

Divertor

Tillack mentioned that several engineering options are possible, and characterized them as conventional (ITER-like), advanced (e.g., ARIES-RS) and exotic (e.g., the LM free surface concepts). Many of the choices revolve around edge physics, and so pressure needs to be applied to obtain some physics input to the process.

Sze noted that sweeping of the divertor plasma may be possible, given the low field strength for the LAR machines. It was determined that ITER canÕt do it, but we may be able. It's an area we should consider, as it is unique to this concept. The main task is a magnetic calculation to determine input power needed. It turns out that Peng already considered this and designed it into NSTX.

Tillack reiterated that we need to start getting physics inputs. It's not clear we NEED to sweep the divertor. We have no idea what will be the conditions there.

At the next project meeting, Tillack said he would talk about the LM free surface options and do an assessment of the implication of using conventional designs.

Configuration

Tillack summarized the status of the configuration options. A conference call was held about a week ago between UCSD and PPPL, and the decision was made to focus on the option of horizontal removal of the TF coils and power core. The other remaining options are vertical withdrawal from below and ARIES-RS type of port maintenance.

Other Issues

Neutronics

Laila is expected back from an extended vacation in a day or two. No progress was reported.

Materials

Tillack asked Blanchard to summarize the team's position on fluence lifetime, as a recent paper by Karditsas has raised this issue again. The limitation is based on creep, but a number of uncertainties remain in understanding his result. It appears very conservative compared with other estimates, and depends on his source of data. Blanchard will contact him and write something for the team documentation.

Tillack posted the ITER materials handbook on the ARIES web site, and it has been agreed that we should use that data from now on. The relevant data for us is for V and Cu alloys.

Safety

Khater mentioned that he had spoken with a student of Corradini to determine if they could help us with safety assessment of blanket options. They may have a code working by December, when we hope to get some results. The code apparently is set up to look at a variety of chemical reactions, including Be/water, Li, PbLi, etc.

Mogahed needs some additional dimensions to continue his LOCA analysis. Tillack said Mogahed should contact him if he needs something.

Systems Studies

Miller described the current status. An aspect ratio 1.6 case is being gone through Š numbers have started to leak out, so we hope to have those distributed to the team soon. The 1.6 case looks quite a bit better. It's a smaller device with higher power density and uses the extra space inboard to expand the centerpost. This case is approaching a point where breeding is a concern. The point was made that taking space away for breeding might eliminate much of the benefit because the centerpost would go back to a small cross section. It would be very helpful if we could identify the point at which we can start putting breeder/multipier in there, and understand what would be the impact in a self-consistent way.

The hole in the CP has not yet been included in the systems runs, but Dahlgren pointed out it should be a minor (5%) effect.

The higher wall loading also raises the old issue of how to properly penalize designs for lower reliability/availability. A shortened calendar lifetime will result from a fixed fluence lifetime. All of the tradeoffs of these effects remain to be fully explored.

The PF system has gotten more attractive with aspect ratio 1.6. The relative distance to the plasma is smaller, and smaller PF coil currents are needed because the1.6 case is much smaller.

Steiner asked how far the code is from looking at a LM CP. After resolving some confusion over the LM design vs. IBC, it seems that a LM cooled CP is not a perturbation on the code itself, but only requires proper input parameters. Miller will try to do a system-level case by next project meeting for the Li CP.

Tilack asked if there is any intent to go to still higher A. Miller said he would like to do it in principle. We need the physics for this case.