Today I’d like to take a look at timber framing as a building system, and subject it to similar scrutiny as I did for stick framing yesterday. By ‘timber framing’, I will be referring to the North American version, especially the modern revival of this building technique, a revival which has been gaining some steam since the mid 1970’s. I say ‘some steam’, since timber frame buildings account for a very tiny slice of the construction industry, so it can hardly be said to be widespread or even growing in influence. I think it is likely to remain a minor niche in the building market, as it is a more costly form of construction than building with 2x material and sheet goods, and economy, as ever, is the main driver in this system.
By ‘timber framing’, I refer to framed structures of timber in which the pieces are connected to each other with wooden joints. To draw a distinction, many buildings have timber components assembled within them, components which are joined with lag screws, timber screws, threaded metal rod, metal plates and brackets, etc – this I call ‘post and beam’ and shall be ignoring that branch of construction in this discussion altogether.
Okay, so I’ve set the terms, and I trust the reader has a somewhat clear idea of what I mean when I talk about ‘timber framing’. Timber frame building methods came to the New world from Europe at the time of the earliest white settlers (I’ll ignore the West coast long houses of the Coast Salish and other indigenous groups for the moment, as these have had virtually no effect on present-day building practice). The earliest immigrant groups to form extensive permanent communities were the English, and thus English timber building methods, especially those of the East English region of Essex, came to have a large influence on building in the New World, especially in the North East of the US. It was not a homogeneous picture however – the English Quakers who settled the Delaware Valley were from the North of England and had a greater preference for massive stone walls in their buildings, and settlers to the Virginian colonies were mostly wealthy elites from the south of England, and their style of building was somewhat different again, and adapting to local conditions. Influxes of Dutch settlers, especially in New York, also contributed a strong influence on the wooden building tradition that developed in that area, and German settlers, primarily in Pennsylvania, also had a strong influence. Some of these influences later cross-fertilized and created new innovations, as carpenters had opportunities to learn from outside cultural traditions to their own.
Regardless, a key point to bear in mind is that the settlers of the 16th and 17th centuries were coming from a largely denuded and barren Europe, where wood, especially long straight pieces of wood, were relatively scarce. Prior to the advent of coal, wood was the primary energy source, and the consumption of wood only accelerated dramatically with the coming of the iron age. Many places in England had very severe laws against taking wood from public lands – er, the King’s lands.
In response to the cost and scarcity of wood, building traditions in many parts of Western Europe came to feature stone or brick wall systems and confined timber to use in the roof structure. This pattern has continued to this day in Europe, where an all-timbered house is something of a rarity in modern construction. Study for the German or French timber framer of today, as in the past, revolves most heavily around the geometry and construction of the roof. In England, timber buildings of the 1400~1700 period tend to feature a predominance of crooked, bowed, and twisted timbers (that’s what they had to work with), and the development of techniques to join short timbers into longer lengths – scarfing – reached a zenith of technical development in England at that time. Clearly, this happened due to a dearth of longer timbers to work with. The methods of timber carpentry of that time, to summarize, were very much in response to dealing with the materials they had on hand.
In the New World, of course, things were very different. Suddenly the carpenter had 200 and even 300′ tall trees with dead straight trunks to work with, and unlike, say, what the English found in Australia, many of these timbers were well behaved when cut and proved to be excellent for building. Thus, framing techniques changed from what they had been in England to take advantage of the bounty of perfect materials available.
Family farm-holdings tended to be much larger in the New world than they had been in Europe, and the larger farms led to larger barns, also made possible by having long timbers readily available. While many settlers might throw up a rudimentary shelter, or log building after their arrival, when they could manage it, they spared no expense or labor with the most important building in their lives: the barn. The barn was always timber framed. Unlike in many parts of Europe, due to the extra space and abundance of materials, the pattern of the the farming family living within the barn, along with their animals didn’t become the normal pattern in North America. Instead, the residence was a separate building, though often quite close to the barn.
The barn, being often a huge structure of -typically- three stories height, did not lend it self to being assembled in a piece-by-piece manner as was common in English residential building. Instead, a method was used whereby the building was constructed of a series of ‘bents’, each a framed section much akin to a slice of bread to the loaf, assembled and pegged while on the ground, and then raised as a unit, with the aid of sometimes hundreds of people, masses of rope, pike poles and so forth. This was the good old fashioned ‘barn raising’. While this technique had been employed in England with cruck-framed buildings, this was a relatively minor aspect of the overall building tradition there at the time. It can certainly be said that the technique of raising buildings composed of bents developed to a great extent in North America.
I’m necessarily simplifying and truncating a richly diverse historical pattern here, for the topic at hand is not historic timber frame building patterns in the New World, but modern ones.
It is the barn to which we return when it comes to discussing the timber frame revival of the 1970’s. While timber framing had been employed for a wide variety of buildings over the years, especially meeting halls and churches, it was the barn which found focus among some builders in the revival. Thus, when you look at modern N. American timber frame buildings, what you see primarily is the barn adapted for use as a dwelling. Artifacts from the barn include a elemental near-rectangular plan, 3~4 bents, simple gabled roof, the use of bent-construction methods, the crane now doing the work of the community, rope, and pike poles, and the vaulted space, a.k.a. ‘the great room’. More on that shortly.
The barn is a place to store agricultural products, agricultural equipment, and/or practice animal husbandry. Thus, for purposes of utility, the barn needs to be a lofted, open space. The spaces between bents in a timber structure are referred to as ‘bays’, and in a typical barn these bays might typically be 12′ apart. The bays do not act as partitions as such, however if one wants to divide the barn into discrete rooms, they are a convenient place at which to frame walls. A room for the horses, a room for the pigs, a room for the tractor, etc. The hay, which needs to be stored in a dry place, goes up above, and the barn typically has a second floor of rough, spaced planks, and a means of hoisting bales of hay/straw up to the second floor. Many barns were constructed with an earthen ramp at one end so a cart could be wheeled in directly to the upper floor.
Barns did not typically feature windows, nor were they insulated, nor was thermal performance much of a concern. They were utilitarian structures that were well-crafted of quality materials, the largest material investment of wealth and energy for the farmer.
So, in contra-distinction to 2x platform framing discussed yesterday, the loads transmitted downwards from the roof in a timber frame are supported on relatively few points. Thus, timber structures are said to have higher point loads. If one of these point load supports fails, say the foundation subsides in the area immediately underneath a post, then the effect upon the whole structure is much more acute than for structures featuring the dispersal of loads through numerous points. For a typical timber structure with bents on 12′ centers, the timbers used to support the roof might be 10″x10″s. Crossbeams might be 10″x16″ – I’ve seen much larger actually. When you space the elements in a wooden structure widely apart, the point loads increase proportionately, and thus quite large sections of timbers are required to adequately carry the load.
The barn is typically sheathed in planking. Tobacco barns in New England in fact have the vertical planks on the outside hinged, or slide-able, so as to maximize cross-breeze drying after harvest. Resistance to shear loads on the structure, such as may occur when the wind blows against the end wall, or snow is blown so that one side of a roof has a lot of snow while the other is nearly bare, are resisted in timber structures by the use of bracing. In the interest of giving space to easily move around unhindered, braces in the interior space of the barn were often shorter – these are short pieces are called “knee braces”. Knee braces, while giving better headroom, also act as fulcramits when the structure is loaded in shear, and can lever nearby joinery apart. This has been amply demonstrated in controlled testing.
The shorter the brace, the closer it is to the joinery connection, and thus the more effective a fulcrum it becomes. Longer braces are to be preferred structurally for strength and resilience, generally speaking.
Knee-braced timber framed structures, composed of bents, are now the normal timber frame in North America. Such structures, with their ‘great rooms and exposed timbers are very pleasing to a lot of people, and unlike the concealment inherent to stick framing, the structure is very apparent and therefore ‘real’. What you see is what you get, right? Good old, time-proven traditional joinery of pegs and wedges, to create a stout yet airy structure, built with great craft, that doesn’t try to hide what it really is – what could be wrong with this picture? Is this the way forward?
I think not, after having looked at the matter for a number of years. I don’t think the barn is a good archetype to choose for building residences. A home isn’t simply some utilitarian structure. A barn isn’t heated or insulated, has no windows, and has a lofted open space to aid in storing hay (or w.h.y.). You see, a braced timber structure is actually not a rigid structure – in fact no engineer will sign-off on a timber structure for adequate performance against sheer loads. To have such niceties as modern windows and air-tight doors, you need a building that is not going to move appreciably in shear. If the wind blows and the building leans slightly, it might burst the seal on the expensive double pane glass windows, or even break the glass. The reason that the timber structure can’t be adequately stiff does not relate so much to the braces, it relates to the pegs that connect them – these cannot be stiff enough to impart adequate rigidity. If you make the pegs out of steel, then they are vulnerable in case of fire, which is in fact a good argument against timber structures using metal connecting plates. One benefit of large timbers is very good performance in fire, as the exterior chars and thus insulates the inside of the timber. Many timber buildings have suffered extensive fires and remained standing.
Now, getting back to my point, this inherent flexibility in a timber structure I tend to think of as an asset, generally-speaking. Flexibility makes little difference in a barn or other structure without windows. Yet, lots of timber structures – houses I mean – are going up, so obviously those engineers must be satisfied somehow, no?
Well, here’s the little secret: the common method of resolving the problem of inadequate stiffness in a timber frame structure is to enclose it within a rigid envelope – say ‘hello’ to the Stress Skin Panel (SSP) – or Structural Insulated Panel (SIP) as they are more commonly referred to these days. These panels are composed of a sandwich of materials, typically off-gassing oriented strand board (OSB) on the outside, some sort of foam board, often styrene, often off-gassing, for the core, and often the gypsum board, off-gassing formaldehyde, is applied to the inside as well. These panels, coming in sizes of up to 20’x40′, are manufactured in a factory, and delivered by truck and placed on the building with a crane, installed in a matter of hours. Right off the bat, even if ‘greener’ versions of these panels are made (and some brands tout their products low-off gassing), they feature high embedded energy in their manufacture, transport, and attachment to the structure.
Leaving the embedded energy point to the side, the SIP is also vulnerable to termites, which have no issues tunneling through the foam. It is difficult to run electrical, plumbing and HVAC systems though that sort of wall envelope, and difficult to modify down the line if the building owner’s requirements change. It is not a form of building with the future in mind. At the end of their – unknown – lifespan, SIPs do not lend themselves to recycling or re-use. They are a worse problem by far for the landfill than simple sheetrock. They are NOT a gift to future generations.
Further, the sizing of timbers to carry the loads, 8×8, 10×10, 12×12, and so on, often are produced by taking a tree and slabbing it to produce a single timber. This timber has the heart-center within, and is thus referred to as ‘boxed heart’. Timber framing is typically done with green boxed-heart timbers, since they are much easier to work (but not to lift, I can assure you). The practice of using green timbers makes great sense for unheated structures like barns, but in a house, where heating is constant in the winter time, the green timbers lose moisture quickly. The loss of moisture from boxed heart timbers in particular, leads to the development of very uneven shrinkage tensions between outside (where the amount of shrinkage is greatest) and the inside (where is is minimal) of the timber. These shrinkage tensions invariably resolve themselves by the timber warping and developing large cracks (termed ‘checks’) on its exterior. While some laypeople apparently consider these ‘checks’ and twists to be ‘charmingly rustic’, it cannot be claimed they do anything but impair the integrity of the timber structure – especially if such checks perpetuate through joints. Many times as well, the shrinkage of one piece will be so severe that it will crack or break the connecting joints, especially at such connections of a central beam and the floor joists that attach to it every few inches or feet. If the tenon on one piece has only a small amount of wood remaining beyond the peg location – termed the ‘relish’ from the French relais, meaning ‘remainder’), then this piece of tenon will often be sheered off by the shrinkage of the receiving timber. A similar effect is common on brace tenons, which usually have very little relish beyond the peg. It is fortunate in such cases that the SIP is keeping the building together, not the frame.
And what of the craftsperson? These SIPS are a factory-made product, which through the use of division of labor, must reduce any craftsmanship at the factory end to a bare minimum. The builder can of course help attach these SIPS (unskilled work) to the timber frame that they spent their time laboriously crafting. That is, unless the timber frame was produced largely by an automated timber cutting machine like a Hundegger, in which case they aren’t doing much crafting at any stage, simply assembling. That is increasingly the model these days in the industry – it’s all about efficiencies and through-put. And what of the forest base?
And what of this amalgam between timber frame and SIP? Well, it makes very little sense, except in regards to the insulation performance of the SIP panel. Unlike the stick-framed wall, with it’s many spots of thermal bridging, the SIP has a core of foam, and the only locations for bridging are at the junctions between panels. Thus their thermal performance is quite good compared to an equivalently thick conventionally-framed wall. The SIPS go on a building mighty quickly, and thus are very cost-effective, in the narrow accounting that only considers immediate cost. The SIPs are very rigid in resistance to shear loads, another benefit they bring.
Here’s the catch though – the SIP can be used to build the structure all by itself – no timber frame is required. In fact, the SIP industry got it’s start as an industry devoted to promulgating a new method of building homes, well before the timber frame ‘renaissance’ came along. The new-built home market remains the business focus for the majority of SIP companies, though they are of course glad to have the timber frame industry on board as well. But as I said, a perfectly strong, stiff and well-insulated structure can be made from SIP panels alone. The SIP is what holds the windows and doors and thus it is the SIP which allows an engineer to sign off on the timber framed building, since he knows the reality: the timber frame is essentially decoration in such a system. It might be better called a ‘trim-ber frame’.
What sort of structures are there out in nature that combine a rigid shell with a skeleton? None I’m aware of. Insects use the exoskeleton, and mammals use the endoskeleton. There are no creatures employing both, as they are designs that represent completely different adaptations. Perhaps though there is some science lab out there working on such a creature, after all, they have created fluorescent mice and mice that grow human ears on their backs, so anything is possible – but not ‘natural’. The hand of man, the technologist, comes up with some odd ideas sometimes.
The timber frame does have two functions in these modern forms of construction employing timber frame and SIP:
1) temporary scaffold to mount the SIP onto.
2) ego gratification
I say ‘ego gratification’ since the timber frame, when contained in a SIP box, becomes an extremely self-conscious building element. It is really screaming at you, “hey look at me”, “look what a lot of big wood my owner can afford”. It is NOT integrated. Rigid shell and flexible frame do not integrate, they are like mixing oil and water. A glance at any issue of industry mags such as Timber Frame Homes, virtually any month’s print run in past years, will show prominently on the cover a ‘great room’. This is clearly the major selling point, from the point of view of the industry, and surely they have done their marketing homework in this regard. The appeal is to ego, for surely it cannot be to practicality, a ‘boring’ virtue apparently. There is nothing practical about having a vast lofted space in a residence. It is a gross display of waste. Waste of timber, a precious resource, to frame it, and waste of energy to heat, since that vaulted area is the place to which all the heat will be rising. A ‘great room’ is not great – it is a display of conspicuous consumption, and that’s all. It lacks integrity. The vaulted space might impress the visitor at first (that must be the main purpose), however, as anyone who has spent time living in such a space can attest, the vast majority of time we are not looking skyward at the ceiling, but forward and down, as we interact with our friends, family, books, the fireplace, the tv, etc. And when we do crane our heads skyward occasionally, to rue our high heat bills perhaps, then the ceiling is so far away that any real architectural detail is lost – only the broadest of stylistic brushstrokes suit such a space. Perhaps dollar bill signs, hung as mobiles and wafting in the rising warm air currents, would be an appropriate finish treatment for such a space. Of course I’m being a tad sarcastic here, but I hope I make my point.
There are other options besides SIPS, and there is a growing movement towards alternatives within a segment of the timber frame industry. A common choice is to infill the timber frame with stick framing, sheath it in plywood or OSB, and insulate, sheetrock, etc, in the conventional manner. As with SIPs, the rigidity of the sheet goods applied to the outside give the needed shear load resistance to the structure, and thus the engineer can be satisfied. This solution however, is not altogether well-thought out. The thermal bridging issues inherent in the stick framed wall are only exacerbated by placing large timbers in the mix. And there’s the obvious point that since a stick-framed wall is more than adequate a form of construction to carry any likely roof load, what’s the point of having timber posts? The intelligent thing to do would be to make the walls entirely stick framed, and confine the timber work to the roof structure, just like in Europe. I might add, that in respect to the thermal bridging problem of the large timbers, the other problem they add to the stick-framed wall matrix is an intensification of difficulty in routing wires, pipes, and HVAC equipment though the wall. Compared to the stick-framed structure, itself mal-adapted to these mechanical/electrical systems, the timber frame is much more poorly integrated. Cutting holes in a series of studs, common practice for the electrician and plumber, is mitigated in effect somewhat by the redundancy of having many studs supporting the load – when there are few supporting elements, as in a timber frame, drilling holes becomes a much more critical issue. and yet, the electrician and plumber tend to follow their experience – if there’s a piece of wood in the way, they drill a hole through it without much thought.
Now, one would think that the timber frame carpenter in N. America would have been on this problem of poor integration with mechanical systems by now, but that’s not the case, leading me to another observation. The timber frame industry, is yet another victim of the Scientific Management ideal of division of labor, and the vast majority of timber frames put up in this country are cut and assembled by companies that specialize in timber framing and nothing else. They cut the frame, either by hand or by machine, they come to the site, where another contractor has established a foundation, then they typically deal with poor tolerances in that foundation (an bigger issue with joined framed structures than for nailed-together ones). Once they’ve sorted the foundation issues (by adjusting the frame), they erect the frame, with help of a crane, and then perhaps put the SIPS on a day or three later. And that’s it, job done. Back to the shop to cut another frame.
What’s wrong with this picture? It certainly makes sense from a specialist-efficiency perspective, no doubt about it. Timber framing requires a different set of skills and tools than stick framing, in investments in those tools demand work to pay for them, which leads to the tendency to specialize. However, back there at the site where the frame was raised, in comes some new carpenters and other tradesmen to finish the job, people schooled in different methods of work and who relate to the building process from that perspective. Like the person with a hammer seeing every problem as a nail to be slammed down, these other carpenters and other tradespeople often treat the timber frame no differently than the structures they normally work with. The timber framer is not present to see how the product they have created works with the other systems that go into making a complete structure. In fact, I would venture that most don’t care – just pay them for the structure they built, and then they are on to the next client and next frame. Any downstream problems are the other carpenter’s concern, not theirs. This situation is unfortunate, and doesn’t serve the client well at all. They end up with a hack-job to a varying degree, and they are told to accept the problems of poor integration as ‘normal’. The timber framer leaves a nice structure with fresh and clean timbers – does he not consider his product later, when it has cracked, warped, and checked, and the gaps produced render the integrity of the joints, dependent as they are upon tightly abutting surfaces, largely moot? Any craftsperson, worthy of the name, needs to visit their past work and learn from it, just as they must learn lessons from builders of the past. Simple imitation is not good enough. Imitating a poor design still results in a poor design. Study to learn which designs came to succeed others, and if it is because they were better, sounder, then imitate that. If the change happened for reasons of economy, then consider carefully, and try to consider the long-term view.
It’s a strange way to do business, this ‘frame supply model’. We certainly don’t buy our cars that way. Can you imagine going to one dealer for a chassis, another for the engine, and another for the body and then finding someone to make it all work? It seems absurd, yet that is the business model in the timber frame industry. It’s thought of as ‘efficient’. It’s ‘normal’.
I well remember a talk given by Len Brackett at the 2001 Timber Framer’s guild Western conference in Asilomar California. He talked about the exact same issue of over-specialization and narrowness in the timber framing industry, and suggested that timber framer’s ought to consider doing more. By that he meant: “hey, timber framer, why not try making the doors and windows? Maybe tackle the kitchen cabinets? Or building the staircase as well on your next project? Or installing the flooring”? Do a little more with each project. Integrate more. Crazy old Len, huh? I fear his wise words simply bounced off the walls of that room – hopefully I’m wrong about that. Hopefully a few people took his words to heart and made it their practice.
The more that the timber framer of today can do, the more he can involve himself in building process in a holistic manner, start to finish, the more integrated the product will become, the better therefore the product will become, and the more fully developed the craft will be. For now though, timber framing in N. America, by and large, aspires to be just a component supplier, as opposed to holistic building practice, and this arranges itself in a competition with other component suppliers, like the SIP industry, the foam-block concrete wall systems people, and so forth. It’s a narrow specialist view.
The Master Builder of yesterday was far more than a timber framer.
I omitted mention of some alternative solutions in the above discussion, to make the timber frame a into part of a more holistic building method. I only dealt with the most common current approaches in the industry. I still believe strongly in the use of joined timbers for building. In upcoming posts I plan to look more at the alternatives, to see what way forward might be found.