Saturday, April 26, 2014

18th c. Virginian folk houses, Part 1

Henry Glassie in his book,  Folk Housing in Middle Virginia, U. of Tennessee, 1975, discussed the 18th c. folk (now usually called vernacular) houses of Louisa and Goochland Counties, Virginia. He recognized that their plans were based on variations of a square and analyzed the ratios he saw.
His illustrations include photographs and measured floor plans.

Because I have only examined buildings in one part of the country, the Northeast, I wanted to look carefully at the vernacular houses he recorded.
From Henry Glassie's drawings could I determine whether or not the house-wrights use geometry? If so what kind? Did settlers in Virginia bring or develop different framing geometries from those in the north?

 Many of these houses had back-to-back fireplaces, one in the main house, one in the adjoining shed. A chimney and fireplace in the 1700's was essential; its cooking fire never went out; it was the source of heat and light. So I include it in my analysis. Here are 3 similar geometries. I have redrawn the plans on graph paper.

Please click the diagrams to enlarge them.

This is The Parrish House, Figure 35,Type 3, in Dr. Glassie's book, mid- 18th century, built of sawn logs dovetailed at the corners.

Its geometry, shown below, begins with  a square, drawn with two solid lines and its diagonals in red. The location of the left window. and the doors are determined by the center. The upper window is located by the half square and its diagonal - a dash-dot red line. The window and door locations are  drawn with a dash-dot green line. 
The rest of the room, including the chimney, comes from half  of the square extended on the right side, dashed red line. That rectangle, divided in half , red dash-dot lines,  and its diagonals, green dash-dot lines, position the right wall and the chimney mass.  I have added green arrows to show those intersections. 
It feels crowded on the drawing, but on site there would have been plenty of space and it is the same geometry each time, just on a smaller scale.

I think the partition wall was a later addition, blocking the stair as it does.  

When I first drew this I thought the size was derived from the Golden Section. I kept looking for a solution based on just a square because the house is static, solid, not fluid. The Golden Section grows; it is dynamic.

This house Dr. Glassie called Unique House. It is House G, Fig. 29 in the book. He notes that it originally had a chimney on both ends.

The red square on the left begins with the chimney. Its right side is the partition wall between the rooms. the right room is a square of the same size. The left outside wall is determined by the square again, just as in the Parrish House above.
If the square starts at the left wall, the right side is at the edge of the stair. See the green square. If the missing chimney is added to the right, the square that would begin on its outer side would also have its left side at the stair. Each side would be the mirror image of the other.
The dot and dash green verticals through the centers of the squares determine the door frame and the center lines of the windows. The right green center-line is canted about a degree, not truly vertical. But the Glassie drawing indicates that the fireplace is also not centered, so I drew it in.  

This is The Moore House, Figure 31, Type 5, House H in Dr. Glassie's book. There is also a front elevation of this house in the book.
The plan shows two main rooms with a shed on each end. A turning stair is located in the back left corner of the larger right hand room.

The main house is composed of 2 squares, drawn in red.  The square to the left includes the chimney. Its green center line locates the center of the windows. The diagonal of half the square crossing one of the 4 small squares within the main one determines the placement of the left wall for the room. See the green dot and dash lines on the left square.
The chimneys on both sides are centered on the squares. But the right chimney is outside the square. Here the smaller square's diagonal and the diagonal of its half rectangle position the chimney base. I've used a green dot and dash line here to mark the intersection as well as the window and door locations for this room.

The sheds on both ends were probably added after the house was built. They are both  laid out as 3-4-5 triangles insuring that they would sit parallel to the existing house. See the dashed and broken black lines on the rooms on either end.

Unlike the buildings in New England and New York, the Virginia houses have only approximate dates and many of them no longer exist.

5/14/2014: I have just returned from driving back roads in Louisa and Goochland Counties, VA, looking at houses that are similar to these recorded by Dr. Glassie. The form is still visible, sometimes as a wing of a larger house, sometimes an out-building on a farm, sometimes in ruins. However, not once did I see a firebox that could have been used outside. This makes me think the houses Dr. Glassie recorded were half of an original house.

Friday, April 18, 2014

Basic geometry - Sandown Meeting House revisited

I have just revised the post I wrote on the Sandown, NH, Meeting House.

I have learned a lot more about early New England geometry since I wrote about Sandown in February. I have been in the Sandown Meeting House, sat in the pews, climbed among the trusses, stood at the lectern in the pulpit. I have visited and explored the geometry of the Parson Capen House, 1683; and the Rockingham, VT, Meeting House, 1785-1800. I am reading Palladio's 4 Books.
I revisited the Sandown measured drawings, found simpler, cleaner geometries which could have been employed.
A new post would have let the old diagrams remain. They are better discarded. I really didn't want anyone to come across the old post through an internet search.  So I redrew and rewrote it.

People had looked at that post 84 times before I revised.
If you are one, please go back and read the new version.

Some of what I learned:

 - How to divide a square into halves; more importantly, why.
Follow the progression on the graph paper here: #1-#2-#3-#4 - counter-clockwise.
I saw how that geometry informed the Parson Capen House.

- What patterns happen when the square in question is divided both horizontally and vertically - see A, B, C, D, E.
- And all the diagonals added.
- And then, imagine 2 squares overlapping! Since this happens at all 3 of the meeting houses I've studied, I needed to think about it more carefully - see F, G.

- And if overlapping is reasonable, how about turning the square 45*, on its point?  See H. The circle surrounding the square is really unnecessary, and, I think, probably not used. Imagine G or H with all its diagonals as in F!
I presented some of that in my diagram for the doors of the Rockingham Meeting House.

Finally I have been trying to understand the influence of Palladio on American vernacular architecture before the 1820's in terms of geometry.
We seem to use squares as a base in New England, and daisy wheel circles in New York. We use 3-4-5 triangles, especially when building an addition. About 1800 the more urban builders seem to begin to explore the Golden Section and less traditional ways to use geometry.

Currently, I do not see builders north of Boston using circles before 1790.
And I will now be skeptical when I see what appears to be a Golden Section. The turned square - H above - uses diagonals just as does the Golden Section. But within the confines of the square, not as an extension.

Wednesday, April 9, 2014

Rockingham Meetiinghouse, Rockingham, VT

The Rockingham Meetinghouse  was begun in 1787, dedicated in 1800.

I had done some preliminary analysis of the design and frame and realized I needed to see it. I wondered if it would feel as spare as the Rocky Hill and Sandown meetinghouses.
It does.
The site, on top of a hill with a view all around, makes up for the simplicity of the structure. One can only come to it from below, and like many 18th c. New England buildings it sits upright and confident. It is very impressive.

I will go back in the summer, when it is open and I can go in. I will also try for a better photograph - the slope of the hill and the snow made it hard to find a good vantage point.

The Town Fathers specified a building 44 ft. by 56 ft. The HABS drawings read 44'-4" x 56'-6". The difference could easily be the addition of the sheathing and siding to the frame. The porches (the end staircases) are square: 12'-2" x 12'-2".

As an architect I like think of General Fuller, the master builder  - who in 1785 was also the architect - considering what would visually, spatially make the statement. I think of him mulling over the open space in the middle with the pulpit centered on one wall, a window behind, the balcony on the other 3 sides.

He laid out the square which defined the open space in the center and then divided that square into thirds to set the columns for the balcony and the posts for the frame. See the black square and columns.

He extended the column spacing - the dotted black lines - to place the posts on the front and rear walls.

Then the diagonals of the square were extended, creating 2 overlapping squares. This made the balcony the same depth all the around. See red square 'a-a-a-a' and green square 'b-b-b-b' both with diagonals.
The porches are squares set in the middle of the west and east walls. See the small green square on left.

I may be romanticizing: he was building in the tradition of Sandown and Rocky Hill. Like those meeting houses, the plan is made up of 2 crossed squares. The width, spec'ed by the Town, was divided in thirds, the squares overlapped.
However when I tried to draw it from the large squares to the central one nothing quite lined up and there were so many lines it was confusing.
It could not have been any easier on a framing floor. I think once he laid out the thirds, he concentrated on the central square.

The front elevation follows the same proportions - the square divided into thirds. The arrows point to where the diagonals cross, the center lines of the windows.
The porches' widths are half the square. The only deviation is the center bay which is a 3-4-5 triangle.

Some of the HABS prints for the Rockingham Meeting house are very faint. The main floor plan, the interior section which shows the roof trusses, the end elevation which I reproduce here, are very hard to read.

I have outlined the shape. I think the roof is 3-4-5 triangles back to back, shown in green. The drawing says  a 9/12 pitch - the little triangle to the right of the roof. A 9/12 pitch is 37*. A 3-4-5 triangle is 36.5*. I would like to read the original dimensions taken on site. I think it is probable that the delineator was not aware of the use of geometry.

The side view shows squares begun from each corner, crossing, both divided into thirds. At the third of each is the wall for the porch.

To make this clearer I had added diagrams:
A) a square with its diagonals, its half rectangle, and the division into thirds, vertically and arrows showing
B) the square made up of 3 rectangles.
C) the same square, divided horizontally, and arrows showing
D) the square made up of 3 rectangles.
Both together produce
E) with only the intersections noted.
F) is the square with all the lines generated from A and B.
G) is the 2 squares crossed.  I will leave it up to you to imagine the confusion of lines resulting.
H) is the diagram for the doors.
Since I have not seen any use of a circle in the plan or elevations I am hesitant to think that the door dimensions were governed by a circle. However, the joiner may not have been the master builder, and may have used a different system.
Or the square was simply turned 45*.

Here is the side door. Its height and center line are the determining dimensions.
From that comes the square and its diagonals, its thirds. The crossing points determine the center lines of the pilasters, the height and width of the door, the width of the architrave.

Now add the square turned 45*.
For this I have used the main door.
I have drawn in only the green 'star'.

If you look back at the front elevation the peak of the pediment is in the center - that dimension determined the height of the entrance, threshold to pediment.
I have added dots at some of the important intersections.

If I add the red star other dimensions like the width of the moldings, the door panels are found.
If a circle is drawn around the squares, the arc defines the bases for the pilasters.

Last picture: Just as at the Rocky Hill Meeting House the eaves on the porches bump into the windows. Neither master builder had solved that problem.