Monthly Archives: September 2008

What good is prefab…?

From what I understand about Chad Ludeman, I am instinctively his fan. His work as a developer with postgreen and his vision with the 100K House project demonstrate his values and commitment. Now he’s written an interesting, well-considered critique of prefabrication that has been bouncing around and getting attention the past few days. His title isn’t subtle: Prefab is Not the Answer to Affordable, Modern & Green Homes. It’s worth reading. Do so, but please come back. I have some comments, including agreements, corrections and rebuttals.

My title in response is similarly direct:

What good is Prefab…if it is Not the Answer to Affordable, Modern, Traditional & Green Homes?

Ludeman means to strike down what he outlines as four broad and untrue assertions about prefabricated homes:

1. Prefab is more Affordable
2. Prefab produces less Waste
3. Prefab takes less Time
4. Prefab is more “Green”

He makes some very good points in his analysis, but I’d like to dig in a little. The subject is interesting and I think ultimately an important one, considering where things may go next for the homebuilding industry.

First, I dislike the term “prefab.” I don’t know what it means. Are the manufacturers providing precut parts, or panelized sections, or modular volumes or a combination? What percentage of the house is completed when it arrives on site? There’s no good definition of the term. Most manufacturers are only precutting and/or preassembling a small percentage (15%-40%) of the whole house, leaving some of the most difficult and time consuming aspects to be accomplished on site. Others are doing more complete buildings, but they are really providing fully-formed modular volumes, which I think of as a different construction species, generally just called “modular.”

I also don’t understand why prefab is presumed to be modern. Isn’t Prefabrication a method of building, not a stylistic outcome? What is the correlation? Is it that progressive design should be (or must be) connected to progressive building techniques? Ludeman seems to think so:

“Like many, I hoped that prefab would be the answer to bringing modern architecture to the masses in the US and beyond. I thought that finally, modern home design would be attainable by those of us who aren’t pulling in lofty six figure incomes.”

But really, where’s the connection? Why would a contemporary house be built differently than a traditional home? The only reason I can think of is to amortize the design fees over numerous projects, allowing the architect/designer to get an income stream, as opposed to the typical architectural design fee. Presumably, this is a technique to make “high” design more affordable, but the real motivation might be just a potentially better business model for modernist architects. Who knows? What I do know is that although there is sometimes a symbiotic basis for the relationship between modern design and prefab, the marriage is more an outgrowth of intention than style. The manufacturing process doesn’t care.

I also don’t like the prefix added to fabrication. When parts and pieces are aggregated into larger and more complete components before arriving at the building site, this is simply fabrication that happens in a different place. The term, therefore, only applies to the component itself, not the building. Windows, cabinets, doors, fixtures, and most of the heating, cooling and appliance equipment come to building sites already fabricated, but the buildings these elements become a part of aren’t automatically prefabricated. Similarly, when wall, floor and roof elements are also fabricated elsewhere (to one degree or another), they are still just wall, floor and roof panels that help to improve the quality and efficiency of the home’s construction. Despite my objections, I acknowledge that the term is pervasive and I won’t fight it for this post.

So, back to Chad Ludeman’s challenge of prefabrication’s trumpeted attributes. To begin with, the four points are actually two. “More affordable” and “Less time” are inseparable, as are “Less Waste” and “Green.” More time, more cost; more waste, less green.

I assume that as a developer, Ludeman has direct experience with prefabrication, but I am intrigued by his claim that site built is faster.

“If you scour a prefab company’s FAQ’s, call them or ask others that have gone through the process you will find that 6-12 months or more is not uncommon from start to finish. With site-built homes often going up in 4-5 months where is the time savings in prefab?”

If that’s true, it explains the relative high cost. Ludeman says that prefab homes would typically cost between $250 and $400 per square foot and says that the site built equivalent would be a lot less. Well, if he’s right about the time, he’s also right about the money. And if he’s right at all, it’s more a critique of the particular companies Ludeman has checked out (or used) than it is a condemnation of the concept of prefabrication.

Ludeman is right that prefab has to overcome its costs (overhead, shipping, crane, etc.) to be competitive and affordable. This is hard to do if the product has the costs of prefabrication without the benefit. I’m afraid Ludeman is looking at some poor examples.

The only point of off-site fabrication is to make consistently better products that can be assembled on site more quickly. If it does that first, then the price points will be variable depending on production runs, types of materials used and customization. Windows, for instance, are almost never made on the construction site. If they were, it would take much more time and the quality/price ratio couldn’t possibly compare with what can be accomplished by the window manufacturers. When the window arrives on site, it is installed in the opening and done. It doesn’t require more work. The same is true of the manufacturing of prehung doors and cabinets. They are better, they save time, and you can choose from very expensive to very affordable. You don’t have to choose between time, quality and cost. You can have all three.

This is not true of many companies that are trying to get on the prefab bandwagon. Since there is no accepted definition of prefab, anything goes. A kit of raw materials, with some essential parts precut, makes the grade; so does a package that comes with open wall panels, roof trusses and bundles of unprocessed materials that have to be cut, shaped and installed on site. Ludeman’s story about his friend having a pile of OSB left over from his building project suggests that the prefab in question had raw materials, even at the building’s sheathing level. No wonder it was slow to construct and expensive. Prefabrication without any added value (cutting, shaping, assembling) is simply a hoax.

Even with much of the package incomplete, it’s hard to imagine any credible reason why a prefab building would take longer than conventional construction. Not only would the company have provided no added value with its package, but it would have added some difficulty factor that must have confounded the guys who were trying to put it together. Perhaps the directions were in Greek. Kidding aside, I don’t get that.

On the other hand, I do understand why costs might be higher. One of the big reasons is that the contractors hired to put the package together often have no motivation to be efficient, and furthermore, have a very convenient excuse if things take longer and cost more: it’s not their fault; they are the solution, the prefab package is the problem. In addition, the subcontractors who bid on the portions of work not accomplished with the prefabrication tend to charge 15-25% more for the same reason. It’s very often not whether the work will be more or less difficult, but just the fact that it’s different. It’s the out-of-my-comfort-zone tax, and it’s one of the hidden costs of any alternative form of construction. Too often, there’s simply an up-charge for having to think.

The secret to successful prefabrication is to make the components as complete as possible. Anything that isn’t done will take longer than it should to accomplish on site. How do I know that? Experience. The lesson learned is that the building can’t be called prefabricated unless most everything, including some of the mechanical systems and finishes, are delivered to the site in a mostly completed state and can be installed very quickly. Since Ludeman has made the point, I’ll raise the ante and say that if the product takes as long or longer to build on site than its conventional counterpoint, then it doesn’t deserve to be called anything other than a pile of expensive materials. There’s nothing fab about it.

There’s an alternative, and it’s similar to Chad Ludeman’s hybrid proposal. We call it a “Shell Package.” It’s not a prefabricated house; it’s the structural shell. Typically, the shell goes up in one to two weeks and includes all the structure from the foundation to the roof and results in a complete, enclosed, weather tight, insulated (closed panel) building. It’s not the finished product, though. It’s just a good jump start. It advances the construction schedule by 2-3 months, but it leaves a substantial amount of work to be done the old-fashioned way on site. The distinction is perhaps subtle, but important. We didn’t promise a car and sell a chassis. With the Shell Package, we promise and sell only the chassis, and then put a lot of effort (including a twice-yearly intensive course) into helping homeowners and contractors understand the most efficient methods for installing and applying mechanical systems and finishes. The Shell Package has good value, reduces time, and is detailed to allow systems and finishes to install easily, but it isn’t priced or marketed as a prefab house. That’s something else entirely.

Let’s define that “something else”: a complete prefab home package should have 60-75% of the above-foundation construction done before arriving on site, and it should be completed a similar percentage faster than on-site construction. This raises the bar, but it doesn’t make sense that a home that is marketed as prefabricated is actually only 15-20% complete when the fabricated elements are in place. This is why I think Ludeman’s gripe is real, but it reveals an immature industry and consumer confusion, not a revelation that prefab can’t work.

A fully prefabricated house is one in which the panelized sections come with finishes in place; wiring, plumbing and heating systems are substantially installed off site; “pods,” as Ludeman calls them, are used for bathrooms and mechanically intensive areas; and millwork comes cut and finished and ready to install. Of course, it’s easy to describe, but harder to do. Despite the reach and challenge, we have decided to make this process our standard product because it is our best way of controlling quality, time and cost. I’ll say more about this process in future blogs, but here are a few photos of recent projects to pique your interest. And since I can almost hear the question, yes, we are building complete homes under $200 p.s.f. (a current project will come in at around $150 p.s.f.; another that is close to completion will be around $185 p.s.f.), and these are fully custom homes, not stock plans. We can’t wait to find out how cost efficient we can be when we find the opportunity to build multiples.
Shop wall buildingExterior walls in process with finishes

Unity PrototypeUnity House nearing completion

Panelized, shop-built homePrivate residence, all components fabricated off site
Prefab private residence

Regarding the green/waste issue, the first point I have already made. Ludeman decries the waste inherent in prefabs by telling this story:

“I recently walked through a delivered prefab with the owner and he offered me pallets of free OSB because he had so much extra and had no idea what he was going to do with it. I’m sure every prefab company is not this sloppy, but it is another indication of waste in an industry claiming extreme efficiency.”

If raw materials such as this are coming to the site, this hardly fits the description of a home that has been prefabricated. Chad’s right: this is a waste of both material and money and there’s no good reason to ship in a raw material that can be purchased at any lumber yard. It also doesn’t fit my definition of prefab. Neither does the rest of his description about the waste issue. He’s describing modular building, not what I think of as a much smarter and better way to accomplish off-site fabrication.

“Most prefab companies are still using loose, batt insulation that is often falling out of the framing by the time the modules reach the site. Batt insulation alone is not what is making the better green homes out there more energy efficient. Also, in many cases, all of that house wrap that is supposed to be sealing up the home from the outside is tearing apart and falling off of the exterior by the time the modules reach their final resting spot.”

This is a fair critique of the typical modular box one sees barreling down the highway, but it doesn’t apply to prefab, in which components are mostly panelized and flat-packed for delivery. Also, the typical fiberglass batt insulation that’s used in most modular homes isn’t either green or energy efficient. Whether installed on site or in a factory, it’s not up to any basic standard for a well-built 2008 home. Other forms of insulation, such as dense packed cellulose or spray foam, can be installed in factory conditions effectively and with greater control than one would typically be able to achieve on the construction site.

The extra framing Ludeman alludes to is also an issue for the modular boxes, not panelized prefabrication, so I’m not going to spend time with that. I do want to address this comment:

“There is something called “value engineering” which can be done effectively on site-built homes and the polar opposite called “over-engineering” that is often done with prefabs. Guess which produces more waste and costs the client more money.”

This just isn’t true. Value engineering can be done on site or in the off-site production facility. The difference is that, in off-site production, the decisions and cutting are accomplished entirely with software and CNC cutting equipment. It is precise and nobody is spending any time on a chop box. The decisions are made by the engineers, who can balance sturdy results and cost-effective framing solutions.

My biggest problem with site construction is that it often results in “devalue” engineering. Frequently, the people on the site think they know more than engineers and they reduce the framing schedule, reduce the fastening patterns, don’t install all the clips and tie-downs and otherwise save themselves effort and time by reducing the structural quality of the building. My start in the building industry was with tract home construction in suburban developments. Before I knew anything, the shortcuts and the flimsy buildings were appalling to me. We sometimes forget that the standard of building in America isn’t the custom homes built by the good builders, or the good prefabrication efforts by responsible companies. That vast majority of homes are tract homes built by low-skilled, unqualified labor.

We can do much better. We have to do much better. When the economy rebounds and large volumes of homes will need to be built for the average American, we need a viable alternative to out-of-date, out-of-sync site construction. I’m firmly dedicated to the concept that the future is in really good quality building components being assembled in controlled conditions, in both small shops and large factories, leaving the site for efficient, effective assembly only. This will lead to the answer for affordable, green homes, both modern and traditional. This could be called prefab, but in the 21st century, I will just call it “smart building.”

Homes without people

In Monday’s By Design blog, Allison Arieff challenges the focus and relevance of the Museum of Modern Art’s (MOMA) Home Delivery exhibit, making the obvious and salient point that cutting-edge technology is always interesting, but when it comes to homes, people matter.

Home Delivery has tons of cool stuff to look at, but it really does feel odd that a show about homes has so little to say about the experience of actually living in one.”

It’s a fair criticism, and I made a similar point-out in my own thoughts about the exhibit. I mused that it would have been instructive if the MOMA exhibit would have tried to make their presentation like a living history museum, such as Old Sturbridge Village, in which actors play the part of the period village inhabitants. It would have been interesting (and fun) to see people trying to live in the little neighborhood of prefab homes on display at MOMA. I think it would make life in an early American village look pretty good. They may not have had amenities back then, but they had space; and they may not have the blessing of modern structural materials to create high perches in metal, plywood and glass, but they had the blessing of being on a firm foundation, living in a dwelling built with solid, natural materials, and an intention of endurance, not with a built-in plan for a fast demise.

Since I agree with Arieff that the lives of people are what really matters in any new housing idea, I’m a little confused as to why Arieff references the Whitney Museum’s Buckminister Fuller exhibit and seems to suggest that, compared to the Home Delivery exhibit’s architects, Fuller was correctly motivated and on the right track, even though he ultimately failed. If that is truly her point of view, I disagree. Even though I found fault with the houses displayed at Home Delivery, I applaud the fact that most of them are attempts to create innovative ways in which buildings could be built, rather than an attempt to create a perfect solution for how people should live.

Fuller’s focus was on people, yes, but he wanted to engineer their lives as much as he wanted to engineer a method of building. To go with his grand alternative, disrupting vision about structure, shape and space, he also had a parallel vision to reshape the daily lives of humans. The Dymaxion House, his most ambitious home-focused project, wasn’t just an innovative building. In Fuller’s view it was THE single solution to the big post WWII housing shortage. He saw Dymaxion as the way every family should live and assumed it would be heralded by all, and would wash across the country overnight, finally making all domestic life neatly organized and efficient, while also completely transforming the architectural landscape with a pox of his round pods. He may have been a genius, but he was also completely full of himself and wrong.

There are few things as personal as one’s home. To remove the inhabitants from the process of defining and organizing the spaces in their home is unfortunate; to postulate that people should submit to a pre-canned and identical home for everyone is both arrogant and ignorant. We are living mammals and after tens of thousands of years of evolution, our need to create our own nest is only more deeply refined and crucial. It is our home that defines us, makes us and remakes us, as we in turn define it, shape it and reshape it, in the constantly unfolding drama of our lives.

Therefore, the only good homebuilding solutions are the ones that create and expand potential, giving increased opportunities for the inhabitants to have the opportunity to self-express, self-create and constantly change. Conversely, bad homes are those that lock the inhabitants into a preconceived notion about who they are and how they should live, and gives them little opportunity to do otherwise. Fortunately, most of those attempts, like the Dymaxion House, fail pretty miserably. Some that exist, like the block buildings in the Communist countries, are evidence that even architecture is capable of being a high crime against the human spirit. Others, like the urban tenement housing projects of the 60s, were imploded to cheering crowds, demonstrating that the human spirit will eventually prevail over bricks and mortar.

But architects keep trying. The house as manufactured widget is still compelling, it seems. But it won’t work. Humans are boundless creatures. The best homebuilding ideas won’t attempt to tie them down; homes will be designed and configured to let their imaginations and lives soar free.

Hopeful promise or looming threat?

I don’t know how the Museum of Modern Art (MOMA) team chose the five full-scale buildings to be in the Home Delivery exhibit, but I applaud their choices. Erected on a vacant lot, the mini-village offers a diversity of perspectives about how new approaches to fabrication might have an impact on architecture and lifestyle, and in turn, how technology might have an impact on fabrication. I used the word might instead of will, because while the projects give a sense of the great force of ingenuity we can expect in the coming years, they each pose as many questions as answers.

MOMA Prefab exhibitThe ultimate question is not “How will we build?” but “How will we live?” When you visit Old Sturbridge Village in Massachusetts, the outdoor museum attempts to help visitors understand both questions: the reconstruction of 18th century buildings make up a historical village that is occupied by actors who attempt to reenact what life was like in the old-world context. The point is to help one understand what was built, how it was built, how the resulting buildings affected life, and vice versa. I thought of this as I toured the MOMA outdoor exhibition. People acting the lifestyle suggested by the buildings would have been an interesting addition to the exhibit. Architects should never forget that homes are more about function than art and more about life than technology, but they can be remarkable indeed when they are also art and when the technology is deployed for the right reasons.

There are five homes in the exhibit, but I only went into four of them. The smallest one, The Micro Compact Home (foreground), caused my claustrophobic tendencies to surface. I toured it from the outside, peering through the windows. Still, it’s quite a feat. It’s the work of British architect, Richard Horden, who managed to shoehorn two beds, a sitting area, a table, dining space for five, HVAC, a shower, toilet, kitchen, broadband internet, clothes storage and two flat screen TVs in a very compact 2.2 ton cube. So, all the accouterments of modern living are there, but it looked to me like a too-heavy, immobile tent, not a place where one could really live. So, lock yourself in your bathroom and try to fathom what it would be like if this was also your bedroom, your kitchen, your living room and your office. If that sounds good, you’re in luck. You can buy The Micro Compact Home for around $40K-$50K. It would be pretty impressive, except there are better executions of the same idea that float or have wheels.

Burst 008I did walk through BURST 008, the work of Jeremy Edmiston and Douglas Gauthier. As The Micro Compact Home drew out my spatial irrationality, BURST 008 grated on my natural irritation about bad detailing and workmanship. BURST is a cool trick that melds a design concept, parametric algorithms, plywood and laser cutters. It basically works and that should be impressive, but the result looks like a tree house built by a Carpentry 101 class. I didn’t know whether it was the fault of the design, the algorithms, the laser cutters, or the people who assembled it. The cuts were rough, the fits were bad, and I had the sense that the whole unit might not be completely structurally resolved, which made me nervous. I noted there were a few too many people touring with me for my comfort, and quickly reduced the load by one.

It is nearly always obvious and unfortunate when builders try their hand at being architects: all the skills of their real trade can’t overcome their lack of skills as designers. BURST 008 shows what happens when architects and programmers think they don’t need the skills of a builder: their good design skills don’t prevent them from abusing valuable building materials. There are good ideas here, but they aren’t yet well executed, but who knows? With the right design, the right algorithms, the right materials, a better cutting machine and improved building details, the next BURST just might take off. As it exists in the exhibition, it’s only an elaborate plywood shack.

MOMA: Instant houseThe Instant House by MIT Associate Professor Larry Sass is also not quite ready for prime time, but the execution is precise and the result is quite sophisticated. The home model that was built for the exhibit was called, “Digitally Fabricated Housing for New Orleans.” Its design was derived from the vernacular architecture of historic New Orleans. Using exactly the same set of tools as BURST 008 (design concept, parametric algorithms, plywood and laser cutters) Sass and his team developed an interlocking tab-and-slot system. The outcome is a precise and flexible kit of parts that allows them to make everything in the building, including columns, ornamentation, and the complete structural shell, using connections and details that could be applied in large and small scale, and in the creation of numerous building elements.

InstantHouse frontIt’s a good start, but there is some work to be done before one would think of Instant House as a real building system. It creates a structural shell, but it still needs a protective skin on the outside and a more appropriate finish on the interior. It also needs accommodation for mechanical systems, insulation and lots of structural testing. Still, Instant House is a strong demonstration of the potential of bringing emerging software and cutting technology to help solve the big problems in homebuilding. For me, this one was a hit, even if not yet a livable house.

System3, by Austrian architects Oscar Leo Kaufmann and Albert Rüf, is the one you would go to if you were trapped in the exhibit in a snowstorm. It’s a complete home with a tight envelope ready for living. Like The Micro Compact Home, this isn’t a prototype; it’s a product that is being sold in Europe. As such, it displays European craftsmanship and attention to detail. The solid wood shell is made from laminated wood elements that are common products in Europe. This is an attractive, modern home and a bunker. For that, I like it. It’s also a clever system. It’s a box made up of component elements that can be configured in a variety of ways. In addition, the box itself can be attached to other boxes side by side and stacked one on top of another, or both. In this way, System3 can be a detached single family home with various size and configuration possibilities, or it could be a multifamily complex with numerous variations. If fact, it could be a village of boxes.
MOMA: System3System3 is very good. It’s also a proprietary product, and since I wasn’t there to buy a home, but to steal ideas, it bored me. The one great idea I might have stolen is the separation of the complex mechanical areas from living areas. But Kaufmann and Rüf had already stolen the idea from Louis Kahn’s distinction between “servant” and “served,” calling it “serving” and “naked,” instead. In any event, we had already developed our own version of the idea for Unity House in which we developed a “mechanical bar” with kitchen, bathrooms, and mechanical room, that serves the private and public living areas. With nothing more to steal, I moved on.

The Cellophane House by KieranTimberlake is the most ambitious and also the most intriguing building in the exhibit. Before I say more, I need to also say that our company built Loblolly, another widely-known KieranTimberlake project, and I will be co-presenting with Steve Kieran at two separate conferences this fall. So, I’m a little close for complete objectivity, but I’ll try.
Cellophane HouseCellophane is a real house, architecturally compelling, and the biggest in the exhibition by far. System3 is the other complete house and could have been equivalent in size, but it would have been rather expensive to import more boxes from Austria. Instant House and BURST didn’t attempt to show how the whole house would work, so they are more demonstrations of an automated system to produce a building shell. In those cases, how the shell would become a livable home still requires some imagination. But Cellophane has five levels and most everything a modern home requires is accounted for: bathrooms, mechanical room, laundry, kitchen, bedrooms and living area. Except for some unresolved building details (big gaps in the building skin, for instance), one can imagine living there, at least on a seasonal basis.

More than anything else, Cellophane House is a physical exhibit of a philosophy. Where Instant House and BURST are particular solutions using software and laser cutters, and where System3 is a particular rectangular building block, KieranTimberlake contemplates an approach to design and construction that could be accomplished in an endless variety of ways, with an endless variety of materials. Their approach has been to develop a process of building that is a temporal assembly of fabricated components that, if logically separated, but strategically connected, can make up a complete house in a manner that is systematic and flexible. The details and the architecture itself celebrate the distinct functions rather than acceding to the notion that buildings are defined by the massing of the intertwined elements of the building shell. Each component type is integrated into the building assembly with connections that are intended to make the building easier to construct in the first place, and easier to demount for reuse or recycling in the future. The idea is that an efficient, sustainable construction method also considers a plan for its sustainable demise. In its philosophical ideal, it’s an Erector Set, constantly regenerating into new forms.

Cellophane House (like Loblolly House) passes two important thresholds in my mind. It is first of all a demonstration of excellent, evocative modern design, while secondly also being a demonstration of an innovative approach to building construction that has few limitations. The KieranTimberlake team is definitely on the right track.

I don’t think I could live in the Cellophane House. Transparency may be a good thing in politics, but doesn’t strike me as a good idea for a private residence where a sense of security is typically an essential quality. But that’s a small quibble and a personal one. There are other Philip Johnsons (the Glass House architect) out there, who don’t mind their private lives being fully exposed. The fact that KieranTimberlake was able to make nearly every aspect of the home with either transparent or translucent materials is further evidence of the ultimate flexibility of the idea.

My main disagreement with the KieranTimberlake idea is that it assumes that buildings can’t or won’t endure. I agree that many parts of a building should be designed to be easily demounted and easily reconfigured or replaced. But I don’t agree with the goal to make buildings more transitory than they already are. It takes a mighty effort to create a house, especially one that can fight off hurricanes and can cut off heating and cooling bills. I’m therefore much more comfortable with the notion that the shell of the building should be relatively permanent, while most other aspects should be designed and built to be easily changed and upgraded.

Even this is a minor complaint, I suppose. I would just change the scaffold idea for a very strong, permanent beautiful structure that could also be dismantled and reused at the end of its long, multi-century life. (Hey, that sounds like a timberframe!) Then, I would wrap that frame with a thick (and opaque) skin that would reduce the energy requirements to a bare minimum for that long life. (Hey, that sounds like the Unity House wall system!) Everything else could be changed, improved and reconfigured at the whim of the homeowners. (Hey, that sounds like Open-Built®!)