Getting it Wright, a series for building stewards and preservationists, offers advice on restorations and maintenance of Wright homes citing real world case studies.

By Steve Sikora, Owner, The Willey House

Prior to breaking ground on her new home, Nancy Willey flirted with a fancy new idea. It was called “air conditioning.”  In a letter to Frank Lloyd Wright, dated March 21, 1934, she wrote that in two of the three bids she solicited, heating contractors did not favor air conditioning. Still, she asked Wright, “If the cost is the same, how would you feel about it?”

Wright did not openly oppose the idea. Nancy’s house was, after all, a departure in every way from homes of the past, and presumably one deserving of an advanced climate control system. Particularly if it could be inconspicuous, or “integral” as he liked to put it.

Mrs. Willey mentioned three bidders, but there may have been as many as eight heating contractors who ultimately proposed on the project. Based upon the surviving record, none of the initial bids came close to Wright’s overly optimistic estimations in cost. Some proposals prescribed a state-of-the-art, forced air furnace. Others favored oil-fired, hot water boilers, though coal was still a widely-used fuel. Some of the proposed boilers and furnaces included a water jacket, to provision year-round domestic hot water to the house, which is one possible answer to the puzzling question from an earlier article, “What kind of hot water heater did the Willeys have?” Later, there was even talk of electric hot water, which is what we use today. Truth is, it’s still a mystery as to what was originally installed.

In the 1930s, Wright was attempting to eliminate bulky radiators, or was at least inclined to disguise the baroque, space-consumptive finned cast iron contraptions. “Poetry crushers” is what he called them in his letters to Nancy. His compulsion for their elimination eventually led to the in-floor radiant heating system he innovated at the Jacobs House two years later in 1936. But for the Willey House, the solution was a heating oil boiler that fed cleverly-camouflaged, streamlined radiators that, for the most part, were concealed inside cabinetry, tucked under shelves or buried in deep pits with floor grates above them. However, prior to arriving at that boiler as a best available solution, Wright did briefly entertain the idea of “air conditioning.”

The heating system that was realized at the Willey House deserves an exploration of its own. But right now, let’s focus on this “air conditioning” business.

Interior of Willey House showing damaged ceiling.

A roof leak and fallen plaster in the low entry soffit opened up unimagined possibilities.

Contractors confused Nancy by equivocating over “air conditioning” in estimates for her heating needs. On April 26, 1934, Nancy Willey sent a note to Wright concerning the latest batch of proposals including this commentary on “air conditioning.”

Therefore I felt justified in looking into the air conditioning proposals. The cost for them ranges from $1050 to $1498. (Excluding the Lewis system, $2005, which is way out of line). I have sent their proposals by separate mail to you…my notations, with each.

There is a difference of opinion about the advisability of placing the inlets at the floor line (base board) or above the breathing line. Both camps array formidable authority for their judgments. My choice for appearance would be floor line, simply because it seems to me that to make them seem integral in a prominent portion of the wall is quite a problem, especially when the problem was not introduced until after the plans were drawn. But perhaps you will be equal to it. Any company will be willing to change its convictions if requested, so, again, please decide.

Contractors added further to Nancy’s confusion over direct or indirect systems.

On the whole the indirect system (“split”) (all except Brown Sheet Iron & Steel and Waterman-Waterbury) seem to be better than the direct. There is a safety factor with not having the hot air we breathe directly in contact with the firebox; this may be chiefly a “psychic” safety. Also, with the indirect systems the arrangements for hot water are comforting; you’d never have to think about it, winter or summer, a 50 or 60-gallon tank would be automatically heated by the same furnace that heats the house.

Again there may be no real advantage over the other however. Of the direct systems I favored the Brown Sheet Iron & Steel, for lower cost. Of the split systems, the Ridler. But the Sutherland seems like the smoothest most tried and comfortable set up.

The Lewis Air Conditioning Co. has just come in with a plan quite different and quite out of line in price, $2005. They want to place individual units (copper coil, blower, fan, filter, washer etc.) in each room. I don’t think much of it anyway. One of the things I like about the air conditioning is getting away from radiators, but the Lewis system would substitute units of that size in each room. However, I suggested they finish the plan they had begun, to send you. They had reasons for rejecting the ductwork plan, and perhaps these reasons should be studied. Anyway, I thought it would interest you. But I consider $2000 silly.

Unfinished interior of space above garage

The unfinished space above the garage which Wright labeled “storeroom” was a perfect location for air handling equipment. The soffits running on either side of the house were ideal for carrying ductwork.

Ducts running into wall in space above garage

The unfinished space above the garage which Wright labeled “storeroom” was a perfect location for air handling equipment. The soffits running on either side of the house were ideal for carrying ductwork.

View of unfinished roof from outside, showing ductwork running inside soffit

The unfinished space above the garage, Wright labeled “storeroom” was a perfect location for air handling equipment, The soffits running on either side of the house were ideal for carrying ductwork.

1934, it seems, was the apex of a technology boom. When it came to innovative thinking in creature comforts–heating and cooling systems, residential hot water heaters, kitchen appliances, washers and driers and what fuel sources powered them–there had never been so many choices available to dangle before the dazzled consumer. Coal, oil, gas and electricity all were vying for domination. At this point in time one could find appliances that would seem bizarre today, such as a natural gas-powered refrigerator, and also find cook stoves powered by fuel sources as diverse as coal, electricity or natural gas, all on the same sales floor. This chaotic lead-up to our contemporary utility standardization creates an understandable difficulty in interpreting the past, particularly when the vocabulary describing new technologies is so open to interpretation.

Coal was king until the mid 1930s when fuel oil became a safe and reliable alternative. Around that same time, in 1934, Minneapolis began running natural gas lines into residential neighborhoods. It is clear in the Willey correspondence that new developments in heating were as abundant as the array of fuels to power these imaginative new systems. One of the companies bidding, the Waterman-Waterbury Co. of Minneapolis, had a catalog that listed: Heating systems, coal, gas, and oil furnaces; gas conversion burners, furnace fittings and accessories; humidifiers, room heaters. Notably absent from their capabilities list was actual “air conditioning.”

Photo of inside air handler.

The air handler (interior) and the compressor (exterior) are the two component parts of a high velocity air conditioning system.

Photo of exterior compressor.

The air handler (interior) and the compressor (exterior) are the two component parts of a high velocity air conditioning system.

Nancy Willey weighed the benefits of a multitude of climate control system types, boilers as well as furnaces, including direct and indirect systems. Both terms, in this case, reference forced air furnaces, not hot water boilers. Here are the differences between them:

  • In a “direct system” air is directly heated by flame. That heated air is directed through ducts to the various parts of the house. Today, we might see this in a portable propane heater for a garage, but never in a home.
  • In an “indirect system” open flame does not come in contact with the air. The flame is sealed inside a metal burn chamber. The super-heated chamber transfers heat to the surrounding air and is directed through ductwork into the house.

Because the Willey House was built slab on grade, the decision between types of heating systems was a rather pressing issue prior to construction, as it was a primary consideration in the preparation of the site. One proposal included a service tunnel large enough for a man to crawl though. It would also carry plumbing and electrical to the bedroom wing at the far end of the house, a wonderful idea. While open to it, Nancy Willey’s greatest concern was cost. She was constrained to a fixed budget.

On April 26, 1934, only two months prior to groundbreaking, Nancy urged Wright to help her decide.

Need I mention time? The gloomiest sight I know is the calendar. Piping, steel and iron, going up daily. Brown Sheet Iron & Steel Co.

This is the least expensive of them all. The complete cost would be:

Specified equipment $675

Oil burner $250 (or $300)

Hot water heater elec. $125

Total $1050

Add to this the cost of tunnel construction, which would be present in all of the others also. And the cost of extra excavation, which is also present in all except Sutherland. This tunnel looks especially expensive, is it?

The man who designed this layout is probably a good engineer. But it is rumored that the company is in receivership. Just what importance that has to the business I am not quite sure of. I am however inclined to have confidence in them and take a chance.

This system was planned on the theory that it was necessary to have a tunnel big enuf to permit access in case of trouble. (With however all the usual assurance that there can be no trouble.) This advantage to accrue also to the water pipes and wires. The tunnel is laid under the fill because a contractor convinced the engineer (against his will, I think) that this would remove the danger of uneven settling of brick floor because of the tunnel. The others haven’t worried about this possibility anyway.


Agnostic to system preference, Nancy wrote interchangeably about furnaces and boilers, though the modernity of air conditioning was tantalizing. Her conversations with Wright, as preserved in the letters, are a bit deceptive to our modern ears. The thing we understand as “air conditioning” today, was not at all what Wright and Mrs. Willey were discussing. In 1934, other than a handful of commercial installations, those at landmarks such as the White House, the Houses of Congress in Washington, and of course, in movie theaters scattered across the country, refrigerant-based, centralized systems didn’t really exist, certainly not in private homes. In formal proposals and conversations about them, the phrase, “air conditioning” seems to have been interjected with little specificity. The verbiage and the ideas behind it were probably as intoxicating and perplexing then, as “artificial intelligence” or “blockchain,” seem to us now, inviting misuse. It appears, in this case, that the expression “air conditioning” so casually bandied about was really used to describe a marriage between another advancement of the time, “forced air central heating”  and the age-old evaporative cooling.

Willis Carrier is the man credited with the invention of the electric air conditioner in 1901. He became a pioneer in the burgeoning field of AC, and established the Carrier Corporation during the Dust Bowl years, promoting his company with the slogan, “Weathermakers to the World.

In 1929, Frigidaire debuted a split-system room cooler shaped like an oversized radio cabinet. Although it fit inside a home, it was extremely cumbersome and required a separate condenser unit (the split). Early rate of adoption was low, but the concept was taking shape. The following year, General Electric patented 32 prototypes for improved self-contained room coolers.

Carrier Corporation, initially formed to serve commercial industry, also recognized the need to develop “unitary equipment,” a term they coined for factory-assembled air conditioners that could be sold as standalone appliances to serve private residences. In 1931, Carrier introduced the Atmospheric Cabinet, a room cooler with a fan, cooling coil and filter enclosed in a wooden cabinet, roughly 5 ft high by 4 ft wide by 1 ½ ft deep (proudly described as being half the size of its competition), with a refrigerating compressor located outdoors (the split). So the first, true residential air conditioners were bulky cabinets scaled like a free-standing wardrobe in every room to be “conditioned.”

1931 also brought about the next innovation, when H.H. Schultz and J.Q. Sherman developed the first window-unit air conditioner. They were similar in design to today’s room air conditioners, a box that straddles a window ledge and cools one or more rooms. One year after their invention, these miniaturized room air conditioners were available for purchase. In 1932, the cost, as reported by Popular Mechanics magazine, was a whopping $10,000 to $50,000 per unit. To put that into perspective, the entire construction budget for the Willey House was $10,000.

In short, these earliest forays into domestic air conditioning did not yet include whole house systems. Those did not fully take hold until the 1950s. But manufacturers and suppliers were dancing around the edges of a breakthrough by the year 1934. “Air conditioning” was in its infancy and was still described in hyperbolic language such as “weathermaking” at the time the Willey House was being erected. Close scrutiny of the fine print of the proposals sent out by the companies touting “air conditioning” is telling.

The Brown Sheet Iron and Steel Company provided a proposal entitled “Specifications and Proposal for Air Conditioning System for Mr. & Mrs. M. M. Willey residence.” Under GUARANTEE the proposal promises,  “The contractor shall guarantee to heat the building, excepting the garage, to a temperature of 70 degrees with an outside temperature of minus 20…” a total absence of cooling. Under CONDITIONING UNIT the proposal describes an oil-burning furnace, calling out a “multi-blade blower with a capacity of 2000 CFM,” “spun glass filters,” and “a bank of 3 spray nozzles,” to affect humidity. What they were describing might have been what today we’d call an evaporative cooler or swamp cooler. A swamp cooler employs the same principle that cools your wet body after a swim. But swamp coolers are most effective in arid regions, where water evaporates quickly. Minnesota, Land of 10,000 Lakes, is not particularly dry.

Storeroom interior sowing media cabinet rolled away from wall for access.

The end wall of cabinets in the storeroom, now a media center, was built on a trolley that rolls out for mechanical access.

Photo showing air handling unit and ducts hidden behind media cabinet in storeroom.

The end wall of cabinets in the storeroom, now a media center, was built on a trolley that rolls out for mechanical access.

Photo showing storeroom with media cabinet in place concealing air conditioning equipment.

The end wall of cabinets in the storeroom, now a media center, was built on a trolley that rolls out for mechanical access.

The Spenser Cooling & Air Conditioning Co. proposal begins with this overview: “…we are attaching herewith a copy of a set of specifications and plans for an air conditioning and heating system for your new residence.” And they reveal a clue as to what they mean by air conditioning. “We might add that this plant will be able to heat the house automatically. Humidify to 50% Relative Humidity at all times, (40% R.H. is the recommended figure), circulate air in the summer, and at some future time cooling may be added by making arrangements to circulate cold water through blast coils.” Clearly “air conditioning” refers to a forced air system with control of relative humidity. The option of cold water circulated through blast coils was for the purpose of evaporative cooling. That feature could be added later–it literally wasn’t available, or standardized, yet.

Growing weary of the long drawn-out debate, and having heard enough about split systems, humidifying and cost overruns, Wright fell back on the hot water boiler as his preferred solution. He responded to Nancy with:

Mrs. Nancy Willey: Delaware Str.


Dear Nancy Willey: Send all the fellows packing who talk about “air-conditioning” and get a simple Spence boiler to carry a sufficient number of feet of heating surface needed. Say 1000 feet. Use the tile-duct specified and the layout we have on the plan. Use any one of five automatic oil feeders and forget the rest.

The whole installation shouldn’t cost you over $850.– I should say N.R.A.– P.W.A. or X.Y.Z.…whatever.

Perhaps a hot-air furnace with a forced air duct to the rear bed-room and Bath would let you out for four hundred and fifty.

The house is so spread out and air-conditioned anyway that machinery and sales – talk is beside the mark. Get some sensible heating-farmer on the job.

Faithfully yours,

Frank Lloyd Wright: Taliesin,

Spring Green: Wis. April 26th. 1934.

Wright ultimately denounced the notion of “air conditioning,” except for his own newly-introduced meaning of the term. Wright’s “air conditioning” here is descriptive of how solid masonry houses can naturally leverage their thermal mass to regulate indoor temperature. The concept he called air conditioning is based on exploiting solar gain in winter and utilizing shade and airflow through cool masonry in summer to cool the house. He assured Nancy Willey that her “house would be spread out and ‘air conditioned’ anyway.” His claim was true to some degree, judged by 1934 standards at least.

The Willey House is a pretty reliable passive solar design. The house faces due south to capture the low winter sun, which warms the brick masonry floor and walls, in the dead of winter. During the hot summer months, deep sheltering eaves (and native shade trees) provide adequate daylight, but prevent the direct sun from entering. Perforated masonry in front of window openings on the shaded north wall draw cooled air into the house.

Updating the System

Flash forward to the new millennium and our stewardship of the Willey House. The population density in the city had increased many-fold. An interstate highway encroached upon the neighborhood, carrying with it road noise and grime. In the decades since the house was first built, the automobile became the prevailing mode of transportation. Concurrent with the drone of internal combustion, property crimes in Minneapolis rose, raising security concerns. At the Willey House, as in other Usonian homes, ventilation came courtesy of tall screen doors, a bit insecure by the standards of modern city life.

Photo of kitchen register, a small round hole in the wall.

The kitchen cool air supply is an unobtrusive hole in the plaster. The return duct and filter are mounted on the kitchen wall. The living room has two unobtrusive slot-shaped supplies.

Photo of kitchen air return, an unobtrusive grate in the wall over a window.

The kitchen cool air supply is an unobtrusive hole in the plaster. The return duct and filter are mounted on the kitchen wall. The living room has two unobtrusive slot-shaped supplies.

Photo of living room air register, a small slot in the plaster.

The kitchen cool air supply is an unobtrusive hole in the plaster. The return duct and filter are mounted on the kitchen wall. The living room has two unobtrusive slot-shaped supplies.

It seemed highly unlikely we could make one fit, but we wanted AC. There were good reasons for it. A climate-controlled environment would be more comfortable, quieter and more secure. Air conditioning would be beneficial to the preservation of the house and furnishings. But best of all, and the green light for us, was when we discovered that, not only was it possible, but we’d be able to integrate a system somewhat invisibly.

It was a stroke of luck that our restorer, Stafford Norris, pulled down some wooden lath in a place where a prolonged leak pooled inside the flat entry soffit ceiling and eventually collapsed the plaster. Peering into the void, he discovered it was a cavernous space devoid of structural elements. Wright had designed a complex network of trusses for this section of the roof, but the builder, in his wisdom, substituted a steel beam, eliminating the many wooden struts. That decision left a gaping tunnel parallel with the exterior soffit. It was an ideal chase to accommodate ductwork! So we decided to install air conditioning. A high velocity unit was the obvious choice because of its flexibility. The innovative UNICO system was ideal for a complicated retrofit like ours. UNICO makes an “open source” system adaptable to a surprising number of conditions found in older homes. It worked admirably here, in a structure that seemed impossible to air-condition.

Cool air ducts in the guest bedroom and study are hidden in the floating decks.

Once the unfinished storeroom, above the garage, was completely purged of the previous owner’s possessions, Stafford confirmed the opportunity was real. Within the framing of the open soffit was a space large enough to run ductwork. A flexible, insulated duct was fitted that ran the length of the garage storeroom, through the floating roofline connecting garage and house, and down to the end of the living room. From there, smaller lines radiated out to the kitchen, living room, study and bedrooms. Registers were carefully hidden in each room. A return air duct was fitted inside a matching soffit on the north side of the house with the return inlet on the kitchen wall.

To make space for the air handling equipment, a false wall of cabinets was designed for the far end of the unfinished room above the garage. The cabinets matched the band of cabinetry original to the dining area. The center section of the wall of built-ins was a custom-made steel trolley that could be pulled in and out as needed to access the air handling equipment.

The compressor unit was placed on the north side of the house, concealed by shrubbery.

Restoration decisions at the Willey House were always balanced between three determining factors:

  1. What Wright originally intended
  2. What was actually built
  3. What would be best for the future preservation of the house

Today we have two systems in the house: a hot water boiler fueled by natural gas for heat, and a high velocity system powered by electricity for air conditioning. The UNICO system has suited us perfectly since being installed in 2004. High velocity air conditioning, with its compact lines and registers, is far beyond what was imagined in 1934. I have no idea if Wright would have embraced it. What I do know is that social and environmental conditions are different today. Personal security is of greater concern than when the house was constructed. In older homes, enhanced home security is a byproduct of the UNICO system. It should be a selling point of its own, particularly for Wright’s Usonian houses.

Ultimately, the long-term preservation of the house remains paramount in our considerations. UNICO provides us with complete climate stability. Wright’s ideas were always modern, relative to the times. He built for the moment. Our job, as we see it, is to preserve that moment.

System specifications:

  • UNICO components: Cooling Module, Blower Module, Flexible Ductwork, 11 Registers (5 slotted, 6 round)
  • Other HVAC components: 2 ton Trane XR11 Condenser Unit
  • Square footage of house is approximately 1365 sq ft

Posted February 12, 2024