More on Timbrel Roofs

Readers really liked the previous post on Timbrel Roofs. This comment is from Paul, one of our readers.

“Your post a few days back about timbrel vaulting grabbed my interest, and I began to look into it. I found some interesting PDFs on Guastavino.net, in their resources page. Click on the Texts button. It contains several texts written in the 1895-1905 time period on fireproof building construction. I haven’t read them all yet, still working on the Prolegomenos on the Function of Masonry. Part II, 1904 text. I will read the earlier texts once I have finished this one.

It goes into some detail on why masonry is the ideal construction material, particularly for roofs, and why other materials (stone, wood and metal) are less than ideal. It also discusses the classical architectural advances of the Babylonians, Egyptians, Greeks and Romans at different stages of development.

I think you will find these texts of value.”
Paul

“Welcome to guastavino.net
The Guastavino Project at the Massachusetts Institute of Technology is dedicated to documenting and preserving the tile vaulted works of the Guastavino Company. In the late 19th and early 20th century, Rafael Guastavino Moreno and his son Rafael Guastavino Exposito were responsible for designing tile vaults in nearly a thousand buildings around the world, of which more than 600 survive to the present day. The remaining buildings are found in more than 30 U.S. states, and include major landmarks such as the Ellis Island Registry Hall, the Oyster Bar in Grand Central Terminal, and the Boston Public Library.”

Source: Guastavino.net

Follow-up email from Paul:
“Hi Owen,
It took some researching, but I found a clean (and free) copy of Rudolf Gustavino’s book on Cohesive Construction. This is the second edition and is complete, with all of the plates, etc. I printed a copy of this 167 page file for my research library. The new Adobe software has some tricks to it. It will automatically print double-sided (if your printer supports this) but can be disabled for printers that don’t have this feature.

When you read this file on your computer, it shows two pages at the same time, but when you go to print, it does a ‘flattening’, which I have never run into before, but what it does is to split the pages into one page per sheet. There is not a lot of print per page, this must have been a pocketbook, so that it could be taken to a job site and consulted.

Here are my thoughts on the subject. Gustavino used baked clay tiles, and even held a patent on how to make the edges staggered so that if a joint should fail, the tile would still be held in place by the shape of the tile alone. This is a really good idea, and this patent, which should be expired by now (the company shut down in 1965) so that this shape should be able to be used. Is it possible to use geopolymer to create these tiles, and possibly for use as the mortar? Some of Davidovits’ videos would indicate that this is indeed possible. Gustavino recommended tiles fired to a minimum of 2000 degrees F, so that if a fire should occur in the structure, the tiles would not expand significantly. This would prevent distruction of the structure, and seems to have worked, as the only Gustavino buildings that have been destroyed were done so deliberately, and not accidently by fire.

With earthbags for the walls, and geopolymer tiles used to create both floors and roofs, a building should last several generations. The only thing to ensure this would be aesthetics, for an ugly building is ugly forever, and is not likely to be kept around.

Gustavino used the arch in many different forms, but preferred the dome where possible, as it distributes the tension more evenly than a barrel vault will. He has a short discussion in his book how a barrel vault, built to his system, will stand up even if cracked diagonally from one corner to another. I found that point interesting. He liked to use arches (domes) even for ceilings that comprise the underside of floors, by putting up vertical ribs between the two surfaces of the appropriate height, generally 24 inches apart. To keep moisture from condensing between the two surfaces and to permit the running of pipes, wire, etc, he sometimes made them partially or completely hollow. This of course did not take into account insulation, but I suppose this could be installed before putting the floor down, as the arched ceiling is installed first.

I have seen some videos where modern test structures were tested to failure. However, every one of these were only one layer thick whereas Gustavino always built a minimum of two tile layers thick and up to 4 layers thick. As the tiles that he used were only 1 inch thick on the average, this makes for a very thin and light structure, yet one in which was of sufficient strength that after only a day or so, workers could walk across the structure and apply more layers of masonry.

All in all, I find the subject fascinating and full of promise.”
Paul

Image source: New York Daily Photo