Wright Chat

pharding

I am building a fledgling archive and database of Frank Lloyd Wright's Prairie Period. This has grown out of our research on the Davenport House and my life long interest in history. I am interested in acquiring intellectual materials and information as opposed to historic artifacts. Thus far I have accumulated a substantial collection of first edition classic books on FLW and Prairie School Architects. I have created a searchable database of the early FLW writings so that I can glean information from what Mr. Wright wrote in an efficient manner. I am also interested in acquiring technical information about the construction of the Prairie Houses. I am especially interested in testing and consultant reports. I and my staff have field measured, to the shop drawing level, and photographed 7 FLW furniture pieces. I have purchased an original specification on a Usonian House. If one would like to contribute information, including photocopies please contact me.
_________________
Paul Harding FAIA Owner and Restoration Architect for FLW's 1901 E. Arthur Davenport House, the First Prairie School House in Chicago

SDR · Joined: 17 Jun 2006 Posts: 4444 Location: San Francisco

Here are some illustrations from an article (28 pp) published in Modulus 22, the Architectural Review at the University of Virginia (1993) by Kenneth Martin Kao. The piece is titled "Frank Lloyd Wright: Experiments in the Art of Building"; it discusses technical innovations in Wright's practice during the Prairie period.

"Comparative wall sections of the E H Pitkin Summer Lodge, Sapper Island, Canada (1900); F B Henderson Residence, Elmhurst, Illinois (1901); Ward Willits House; Arthur Heurtley Residence, Oak Park Illinois (1902); and Frederick B Jones Residence, Lake Delavan, Wisconsin (1901)."

American System Built Houses window detail

Mr Kao, a practicing architect, is a graduate of U Penn, Harvard, and the Lehrstuhl fur Architektur und Konstruktion at the Swiss Federal Institute of Technology (ETH); his dissertation "focussed on the role of technology and experimentation in the work of Frank Lloyd Wright."

pharding · Posted: Sun Apr 08, 2007 5:12 pm Post subject:

Thank you for the information. The batten strip details are somewhat suspect. The board and batten system must stop water travelling horizontally between members, therefore the the board and batten must interlock. Also, the system must account for wood shrinkage and building settlement. The back side of any any board and batten system must be flush because it is applied on top of 1 x 6 horizontal sheathing. The 1901 Davenport board and batten system had these features and they worked exceedingly well for 106 years. I will post a drawing from Davenport on Monday when I get to the office.

Also, the gutters are not shown in the wall sections which are generally concealed.
_________________
Paul Harding FAIA Owner and Restoration Architect for FLW's 1901 E. Arthur Davenport House, the First Prairie School House in Chicago

SDR · Joined: 17 Jun 2006 Posts: 4444 Location: San Francisco

Indeed, that is curious about the batten wall sections.

What is it that is "generally concealed" ? I think Wright might have preferred chain rain leads over conventional downspouts -- just as auto designers of the 30's drew their fenders as if no bumpers would be defacing them. . .?

SDR

pharding · Posted: Sun Apr 08, 2007 5:46 pm Post subject:

SDR: You are correct to question that statement. I checked in my trusty 12 volume FLW Monograph. Most generally had applied gutters. Some had no gutters. A few had concealed gutters.
_________________
Paul Harding FAIA Owner and Restoration Architect for FLW's 1901 E. Arthur Davenport House, the First Prairie School House in Chicago

SDR · Joined: 17 Jun 2006 Posts: 4444 Location: San Francisco

Thank you. I wonder if he ever concealed downspouts ? I know that some architects have run a roof drain from the middle of a flat roof through a central stack. Seems risky to me. . .

But what could be riskier than the roof-edge detail and the "zero-protrusion chimney" of some of the Usonians. . .!

SDR

SDR · Joined: 17 Jun 2006 Posts: 4444 Location: San Francisco

It's a little hard to decipher the caption of figure 8., above, as there are six illustrations and apparently only five buildings cited -- but if I were building I would be tempted to try a version of the lower left example -- perhaps with a less aggressive batten ? ( You gotta love the sheet-metal closure/support piece indicated.) All pieces could be re-ripped/milled to the same angle, and the drainage/water penetration issue seems well addressed.

How about the metal coping added to the top window trim in the ASBH detail !? The man was mad. . .but you have to admire the relentless drive to conceive and to try everything -- sort of like Mother Nature ?

SDR

outside in · Joined: 29 Jul 2006 Posts: 403

In the systems built houses, Wright drained the roof to the center and used a 8 inch piece of cast-iron to take the roof water down through the building (adjacent to the chimney) where it joined up with the sanitary piping before it headed out to the street. I believe it may the only time he did it!

SDR · Joined: 17 Jun 2006 Posts: 4444 Location: San Francisco

You mean. . .these houses ?

Model A231

Model D101

pharding · Posted: Tue Apr 10, 2007 6:00 am Post subject:

Look at the headers above the windows in the wall sections. Basically vertical double 2 x headers as one would expect today, are missing. That is what we found on Davenport. In The Details of Modern Architecture, Vol. 1 by Edward Ford, a superb book about the technology of modern architecture, has exterior wall details of the 1902 Ward Willits House. If those details in that book are accurate, a critical difference exists between the exterior wall detailing on the 1902 Willits and the 1901 Davenport House. On Davenport the ribbon windows are just big slices out of a balloon frame with no stud continuity at the windows and no header other than a horizontal 2 x 4. On Ward Willits, the studs actually run continuous through the window mullions. Thus at the window mullions the stud would run continuous from the foundation wall through the window mullions up to the roof framing. This was an enormous improvement over Davenport. On Davenport at three ribbon window locations we had to cut in a single 2 x 12 header to address this. If Ward Willits was designed on a grid, it probably grew out of this technical requirement.

Another key difference between Ward Willits and Davenport is that the top chord of the truss is a 2 x 6. On Davenport this is a 2 x 4. This was critical improvement for the overhangs because the top chord of the truss extended out for the broad overhang framing. FLW's house technology evolved with the development of new design ideas, Unfortunately along the way there were some technical failures with many successes. FLW was a technical optimist.
_________________
Paul Harding FAIA Owner and Restoration Architect for FLW's 1901 E. Arthur Davenport House, the First Prairie School House in Chicago

outside in · Joined: 29 Jul 2006 Posts: 403

I think the spelling is Willits - You're describing a balloon-frame house, and the Willits House was built similar to Davenport, Ed's drawings were accomplished mostly by students and they didn't get everything quite right.

I believe the only time the wall framing was designed to be integrated with the window and door system was in the System-Built houses, where studs were spaced at two foot and the windows fit precisely between the wall framing.

outside in · Joined: 29 Jul 2006 Posts: 403

I forgot - many times what we consider to be conventional headers (2x10.s etc.) over window openings were not used in early Prairie Homes (and many homes of this period) because some of the load was transferred over the opening by the exterior sheathing, which usually consisted of 1x8 boards nailed to each stud, which acted like a beam. Also the casement windows were set into thick jambs, some of which were 1-1/2 inches wide and firmly shimmed into place. Structurally, the jambs acted almost like studs, transferring the weight of the wall down to the sill plate below.

pharding · Posted: Tue Apr 10, 2007 7:05 am Post subject:

Each individual window can deflect in the center with this approach because the load transferred through the center stud bears down on the weakly supported center of the window. On Davenport this was a bigger problem on the ground floor ribbon windows because they carry more vertical loading than second floor ribbon windows. Another problem is that where a horizontal load is applied to the wall the wall tends to hinge at the top or bottom of a band of ribbon windows because of the lack of structural continuity. The horizontal loading can contribute to inward or outward bulges at ribbon windows.
_________________
Paul Harding FAIA Owner and Restoration Architect for FLW's 1901 E. Arthur Davenport House, the First Prairie School House in Chicago

allwrightythen1 · Joined: 26 May 2006 Posts: 41

Mr. Harding,

Is a book on the Davenport House restoration in the works?? Seems like you have all the makings for a good one!

pharding · Posted: Tue Apr 10, 2007 9:36 pm Post subject:

I have been approached by an architectural book publisher who heard about the project. Toward that end I am accumulating content. I find the whole process and the project to be really fascinating on multiple levels.
_________________
Paul Harding FAIA Owner and Restoration Architect for FLW's 1901 E. Arthur Davenport House, the First Prairie School House in Chicago