Anybody interested in this website will also be interested in John Perlin's new book Let It Shine: The 6,000-Year Story of Solar Energy. The book is an impressive work and a rewarding read. Among the many interesting topics that Perlin surveys in the book, one section that caught my attention concerns the ancient Romans' use of solar heating in the monumental bath structures (p.27-33).
Perlin wrote: "the Romans usually glazed the whole south wall of their bathhouses."
Baths of Caracalla model (Rome, 212-216 CE), showing large, southwest-facing window glass.
Perlin also noted: "Seneca wrote that these giant windows trapped so much solar heat that by the late afternoon, bathers would 'broil' inside the baths." Here is where Seneca writes about the baths:
To show the Romans' facility for solar heating, Perlin points to a 1996 study, "Windows, baths, and solar energy in the Roman empire," by physicist James W. Ring (American Journal of Archaeology). I tracked down and read the original article.
Ring concluded: "the Romans deserve high praise for their use of solar energy." His numbers showed, however, that at noon on the winter solstice, a typical 'solar room' in a Roman bath would lose slightly more heat than it would gain---not very good. He made a low assumption of a 30˚F outdoor temperature and a high assumption of 100˚F inside. We may infer that at more moderate temps, or on days with stronger sun, the windows would likely offer enough solar gains to outweigh the losses.
Ring also wrote: "The sun alone on sunny days could provide most of the energy to maintain the 100˚F temperatures. Indeed, even with fires reduced on sunny days, there would probably be some thermal energy [from the sun] stored in the doors and walls that would maintain the temperature as the sun [went] down. On days where the sun [was] obscured by clouds, the hypocaust with reduced fire, or turned on only part of the time, could by itself easily maintain the temperature [100˚F inside] even with the temperature at 30˚F [outside]."
I find the Ring study to be well-done and informative. The paper's calculations are good and the assumptions are sound, although he did not take into account thermal mass lagtime effects apart from the comment above, and he did not discuss daily or yearly totals of gains & losses. He also did not discuss one of the biggest problems that modern solar architects discovered, as I emphasize in my book, the potential for summer overheating. (A curious point is that Ring gave a lot of attention to the question of whether solar heating would have worked in the baths with unglazed openings---obviously not!)
In the Roman baths, the solar heat assisted the hypocaust system of 'mechanical' heating. The technology of the hypocaust is fascinating. The baths included boiling rooms with tanks where hot and warm water was produced by burning wood. The heated rooms within the baths had raised floors and hollow walls, so that hot water and hot air were circulated in these cavities to create radiant heating in the rooms. This is conceptually the same as today's technology of radiant floor heating, which of course works well in combination with passive solar. (Interestingly, a 1956 scholar quoted in Ring's paper said of the Roman baths: "the principles of radiant heating ... made the open rooms possible and, to date, we have not matched them in a modern building.")
And if you're thinking 'they must have burned a LOT of wood', you're right. In fact, the Romans probably turned to solar heat out of economic necessity---they were running out of wood, and prices rose steeply. (Which of course brings to mind the famous quote: "Those who cannot remember the past are condemned to repeat it.") A link below has more on the importance of wood, and the scarcity of it, in ancient Rome. And Perlin, in another book (A Forest Journey: Wood and Civilization), said that heating the caldarium of one bath consumed 114 tons of wood per year.
I like words, so all of this prompted me to look up the etymology of the word hypocaust, which is usually translated as "heat from below." The Ancient Greek hypo- means "under," although in medicine it connotes "deficient" and in chemistry it means low-in-oxygen. And caust- means "burnt," as in caustic.
Also to note: the large divided arch windows seen above are known as "Diocletian Windows". Ring noted that Romans made window frames from both wood and metal.
Finally, I'll mention that a quick look at Google Earth shows that the two most 'important' Roman baths, those of Caracalla and Diocletian, were not oriented orthogonally to the cardinal directions. Instead, they were built so that the major glass walls face southwest. (Which brings to mind the fact that Villa Rotonda and Chiswick house were later oriented diagonally, as was Villa Savoye.)
Roman Baths – Precursor for Modern Temperature Control at UrbanEmergence
Deforestation of Italy in Roman times discussed at Environmental History Resources
Diocletian Window at Wikipedia