This grid is VERY similar to Grid #1 (case3-struct-1-p2dfmt), 
with the exception that the top wall's shape has been adjusted to 
APPROXIMATELY account for blockage caused by the side plates in 
the experiment.

This grid is in nondimensional units (nondimensionalized by hump chord
of 16.536 inches).

This is a 4-zone grid, with 1-to-1 interface connectivity,
created using Gridgen V14.

The zone sizes are as follows:
 793x217
 161x121
  65x121
  49x217


The 1-to-1 blocking data is as follows:

   1-1 BLOCKING DATA:
      NBLI
         3
 NUMBER   GRID     :    ISTA   JSTA   KSTA   IEND   JEND   KEND  ISVA1  ISVA2
      1      1             1    233      1      2    353      1      1      2
      2      3             1      1      1      2      1    121      1      3
      3      1             1      1      1      2      1    217      1      3
 NUMBER   GRID     :    ISTA   JSTA   KSTA   IEND   JEND   KEND  ISVA1  ISVA2
      1      3             1     65    121      2     65      1      1      3
      2      2             1    161      1      2    161    121      1      3
      3      4             1     49      1      2     49    217      1      3

The way to read the above table is to match the number 
from the 1st set with the number from the 2nd set.  For
example, for interface number 1, grid 1 connects with grid 3.
Because this is a 2-D grid, the i-indices ISTA and IEND
are irrelevant.  The j-index from zone 1 from 233-353 
(at k=1) matches up with the k-index from zone 3 from
121-65 (reverse ordering) (at j=65).  The parameters isva1 
and iva2 indicate which index is varying (but because
this is a 2-D grid, the fact that the i-index is varying
(ISVA1=1) is irrelevant)... for interface
number 1 the j-index (ISVA2=2) is varying in grid 1, 
and the k-index (ISVA=3) is varying in grid 3.


The B.C.s are as follows:

Zone 1 J0  : 1-to-1 interface with zone 4
       JDIM: downstream tunnel outflow (set back pressure)
       K0  : wall from 1-233 and 353-793; 1-to-1 interface with zone 3 from 233-353
       KDIM: inviscid wall
Zone 2 J0  : bottom of bell-shaped chamber
               this BC may be inviscid wall when there is no flow control
               or else it may be the BC that defines suction/blowing
       JDIM: 1-to-1 interface with zone 3
       K0  : wall
       KDIM: wall
Zone 3 J0  : 1-to-1 interface with zone 2
       JDIM: 1-to-1 interface with part of zone 1
       K0  : wall
       KDIM: wall
Zone 4 J0  : upstream tunnel inflow
       JDIM: 1-to-1 interface with zone 1
       K0  : wall
       KDIM: inviscid wall

----------

This grid system extends to x/c=-6.39 upstream of the
bump's leading edge, and to x/c=4.0 downstream.  The grid
height is modified from the original height of
y/c=0.90905 (which corresponds with the actual
tunnel height of 15.032 inches from the splitter plate to
the top tunnel wall).  The tunnel top wall is adjusted
to APPROXIMATELY account for tunnel blockage due to the
side plates.  The top wall contour shape was obtained by computing
the approximate local cross-sectional area of the side
plates and dividing it by the local cross-sectional
area of the tunnel (above the splitter plate and hump), then 
scaling the top wall shape in the 2-D grid by this factor.

Also, note that if you desire to input boundary conditions
directly at the "inflow" location of x/c=-2.14 (the location
where upstream data was taken in the experiment), then zone 4
can be omitted (i.e., zone 1 has a forward extent to 
exactly x/c=-2.14).

The min spacing at viscous walls was set to be approximately
8.e-6.  This minimum spacing yields a y+ value less than 1, 
even for the grid when every other point is removed in each 
coordinate direction.  The top tunnel wall has inviscid-type 
wall grid spacing.

Format for the structured grid is PLOT3D-type, formatted, 
MG, 2D:

      read(2,*) nbl
      read(2,*) (idim(n),jdim(n),n=1,nbl)
      do n=1,nbl
        read(2,*) ((x(i,j,n),i=1,idim(n)),j=1,jdim(n)),
     +            ((y(i,j,n),i=1,idim(n)),j=1,jdim(n))
      enddo
