Alumni Page
Reunions Register here

Virtual 1221

The red owl logo of The Hawthorne School
The Hawthorne School
Washington, D.C.

Virtual 1221 -- how it's constructed

Measuring the Building

1963 yearbook coverThe first problem was (and still is) to determine the building plan and dimensions.  I began with the picture on the cover of the 1963 yearbook, which shows the the front of the school viewed almost perpendicularly from across Massachusetts Avenue.  This photograph by Howie Mitchell (or Grayson Mattingly?) is so clear that it is possible to count the courses of bricks -- though probably not on the smaller version shown at left.  So I counted the bricks to determine that the parlor bay window openings are 40 courses tall, the sills are five courses, and so on.

The picture was taken with the camera tilted very slightly upward, but the cameras appears to have been in approximately same plane as the tops of the basement windows.  At the intersection of that plane with the picture, the horizontal and vertical scales should be the same, so I deduced that the width of the parlor bay window openings is equal to the height of 15 courses of bricks. 

Sandy and Eleanor still have a few bricks from 1221, but these are special decorative red bricks which are not representative of the yellow pressed brick used in the front part of the building.  Fortunately, the dimensions of pressed brick (unlike common brick) were highly standardized: 8 3/8 x 4 1/8 x 2 3/8 inches.  Because pressed brick was very smooth and uniform, mortar joint dimensions of 1/8 to 3/16 inches were typically used, and I think the mortar at 1221 was close to 1/8.  This means that the height of one brick plus one course of mortar would have been 2.5 inches, so for example the window openings would have been 40 * 2.5 = 100 inches, or 8 feet 4 inches high, and 15 * 2.5 = 37.5 inches wide.

The total width of the front facade (not including OSH) in the "Virtual 1221" model is about 43 feet; this compares favorably with the width read out from Sanborn insurance maps, which ranges from 40 to 46 feet in various editions.

Building a Wall

box representing wall
box with window openings
wall with exploded lintels and band course
complete wall, assembled

Armed with a few dimensions and some guesswork, I've begun to model the building.  The model is built up in POV-Ray from simple geometric shapes such as  rectangular boxes, cylinders, and spheres, with limited use of more complex shapes defined by quadratic splines (polynomials which can form more or less arbitrary curves).

The first step is simply to define a box which represents a section of wall (in this case, the south wall of the parlor bay), ignoring details such as window openings, lintels, and sills.

Next, the window openings are subtracted from the wall volume.  The openings have some extra space removed to accomodate the sill -- this wouldn't be necessary except that some of the sills have a bevel, and the wall would show through in that area.

The sills are just blocks with an inclined box subtracted to make the bevel.  The lintels are more complex -- the model doesn't yet really do them justice -- and are built up out of a number of overlapping boxes, some of which are tilted, as shown in the exploded view above.  Until recently I hadn't found a good picture of the lintel brackets, so I'm still using a rough approximation.  These are made from a quadratic spline curve fitted by hand.

Finally, all the decorations -- lintels, sills, and raised band course, are assembled and added to the wall.  In some cases, such as the OSH corners, the wall then needs to be rotated or translated (moved) into its final position.


Photogrammetry is a method used to find the relative dimensions and orientation of objects from photographic views.  It relies primarily on the inversion of geometrical perspective to obtain the size and shape of the original object.   Ideally, the photographs are taken with calibrated cameras at a known distance and direction from the building, but of course that is not possible with 1221.  Recently, new techniques have been developed which allow the use of uncalibrated photographs, but it is still necessary to have some known dimension (in this case, the pressed bricks) and to have photographs taken from many different angles.  To further correct and complete the Virtual 1221 model, I am in the process of learning more about photogrammetry.

-- Dana Sawyer

Copyright © 2004-2006 by Dana Sawyer

This page last updated May 21, 2006

Valid HTML 4.01!