Evaluating a masonry dome

A friend of mine, Dick Fischbeck, recently shared an article with me about concrete domes built at Yerba Buena Center for the Arts in San Francisco, California. Mr. Fischbeck is a follower of R. Buckminster Fuller, and believes –as Fuller did- that weight is an important criterion for evaluating a building. As I discussed earlier in this blog, if a building weighs too much, it is fundamentally flawed, according to Fuller and his followers. So Mr. Fischbeck sent me this article, writing “here’s a bad idea for you!”


I read this article and was quite intrigued. I had to agree with Dick, that this seemed like a bad idea, although I suspect I have different reasons for thinking so. This is an interesting example which shows many of the advantages of the triangular interlocking masonry system which I’ve been describing on this blog, versus conventional masonry assembled with rectangular bricks and blocks.

At the risk of upsetting this large team of engineers, architects, artists, donors, sponsors, volunteers and anyone else involved in this project, I will now offer my informed critique of this structure. It is my intention to show that a better system is available for masonry construction of domes and spherical sections. Given that Mark Sinclair, a principal at Degenkolb Engineering, which donated expertise and staff time to this project, is quoted as stating "Part of the reason I'm excited is that with something like this, you see how it could be applied (economically) to homes and small commercial buildings," I am gently trying to point out that there is a much better way to build masonry domes than what was done here, and to apply this improved masonry method to the economical construction of homes and small commercial buildings.

First, if we look at the assembled structure built at Yerba Buena, the profile of the dome structure is accentuated by undulations and sharp changes in the radial design it attempts to describe. These undulations create undue and unwanted focal points for stress. At these locations, the stresses become focused and serve as points where failure is more likely to occur. A more stable design is provided if the structure is kept truly radial or catenary, such that there is no focal point for stress.

Michael Ramage, an engineer who attended MIT and is currently teaching at Cambridge University in England, designed this dome system at Yerba Buena. Dr. Ramage is quoted as saying "Vagaries of construction are to be accepted ... We let the structural forces dictate what the forms want to be," in explaining the undulations found in this dome system. To me, these “vagaries” are to be minimized, avoided and are not acceptable. To me, they visually detract from the form, and from an engineering standpoint provide focal areas of stress which weaken the structure. To me, the architect, designer and builder should dictate what the form is, not the vagaries of construction. 

Second, a rectangular brick or block is not the ideal unit shape for assembling a dome structure. As discussed earlier in this blog, a triangular block is inherently disposed to conjugate shearing along control joints, so that stress is allowed to be relieved through strain (movement) in a controlled manner; the structure is “pre-fractured” and is thus less likely to suffer a fracture by nature of its being pre-broken.

A rectangular block or brick simply cannot be assembled into a sphere (or dome) without the creation of gaps or spaces between bricks, unless the bricks are custom cut and fitted to their specific location. Conversely, triangular blocks can be assembled into a number of polyhedral arrangements, without creating gaps or spaces between bricks. These triangular unit shapes are interchangeable and do not have to be custom cut or placed at specific locations within a dome or sphere.

If a very large dome is assembled, and mortar is used between bricks, then the effect of gaps or spaces between bricks is minimized. One obvious example of a well-executed dome built with rectangular bricks is the Brunelleschi’s Duomo, which also utilized a herringbone pattern for bricklaying. On a smaller scale (smaller domes, like the Yerba Buena domes), these gaps and spaces between bricks are noticeable, and have an effect on both the visual appearance of a structure and its engineering performance. 

The domes assembled at Yerba Buena were done with two concentric shells of brick; one interior and one exterior. Between these shells, a geotextile fabric was included as a tensile element to help provide some tensile reinforcement to the overall structure. To me, this use of geotextile fabric appeared somewhat sloppy, wasteful and inelegant.

The masonry system I have developed and am attempting to describe on this blog also allows for concentric shells to be assembled, if so desired. Also, the interlocking “DIMP” design allows for a tensile element to be incorporated into the structure, in a more efficient and simple system which involves weaving this tensile element into the blocks as they are assembled. This incorporation of tensile elements is done so that the tensile elements are placed at the conjugate shear planes within the structure, resulting in a stronger, tougher system which utilizes active control joints, allowing for stress (applied force) to be relieved via strain (movement).

This notion of allowing stress to be relieved by strain figures critically into another aspect of evaluating masonry domes regarding seismic stresses. In the Yerba Buena structure, the architects, engineers and designers had to design the structure so that it was suitable for earthquakes which are more likely to occur at this location. Their design dealt with this engineering challenge by providing a rigid dome, which will move as a whole, atop a base isolation system. If the ground were to move underneath the dome, the whole dome is free to move in its entirety; like an upside down bowl placed atop ball bearings. This engineering solution requires an expensive and extensive base isolation mechanism which the structure sits on top of. In contrast, the interlocking triangular block system I’ve been describing in this blog relies on the ability to deform (strain) under seismic forces (stress). This is possible through both the interlocking feature of the block and tensile elements (steel cable, carbon fiber, etc.) woven into the block as they are assembled. Each tensile element is anchored at the base of the dome, and fitted with a spring which dampens the stresses and add to the dynamic flexibility of the dome. Thus the Yerba Buena domes and the domes I’ve developed have fundamentally different approaches to dealing with seismic stresses. They rely on the entire dome being able to move relative to the ground, and my design relies on the ability of the structure to strain along control joints, via conjugate shearing. The design I’ve developed is further advantageous because a dome thus constructed can sit atop vertical walls, and is still free to move; the Yerba Buena dome cannot be built atop vertical walls, unless the entire structure (including vertical walls) is allowed to move via a base isolation system. Again, this requires more extensive and expensive engineering features.

The Yerba Buena dome used lightweight bricks for their construction. This seemed to me an unnecessary feature. With concrete bricks, any reduction in weight is also accompanied by a reduction in strength. One of the fundamental features of a masonry dome is its high compressive strength: there is really no reason to use a lightweight block, unless one is a strict adherent to the principals of Bucky Fuller. It seems to me that this structure was made less strong by using lightweight bricks. The only real advantage to lightweight bricks is that they serve as better thermal insulators. However, this occurs at the loss of thermal mass to the structure, which is a beneficial aspect of masonry construction. I believe (as do others) that a more thermally efficient structure is provided by incorporating a high thermal mass, and simply insulating the outside of the structure so as to maximize the thermal mass benefits.

It should be noted that lightweight masonry units are in fact advantageous for high temperature refractory applications, where thermal insulation benefits outweigh the thermal mass benefits at high operating temperatures, such as in a kiln or furnace. The kilns and furnaces I’ve built using my masonry system did incorporate liquid foam insulation into the cast bricks, to provide a lightweight insulating brick.

If a dome were built with standard manufactured rectangular concrete block, the dome would have the block oriented such that the weak axis of compressive strength is facing the outside, or radial, direction (the weak axis is normal to the axis of compression as the block are made). The triangular concrete block which I’ve developed have the high strength axis (direction of concrete compaction and consolidation during manufacture) facing the outside. This provides a much stronger structure.

This wraps up my evaluation of the Yerba Buena dome system. It is my hope that anyone reading this critique can do so in the constructive manner in which it was intended. I am certainly very happy to see others attempting to build concrete domes today, and to aspire to creative solutions to some challenging engineering problems. If anyone wants to try and build a better concrete masonry dome, please contact me; I may be able to help. I am willing to allow use of my patented systems at no cost for interesting and worthy projects such as this.

Construction is a conservative industry. Within construction, the field of masonry is even more conservative. I hope to advance the state of masonry today, through a thoughtful approach, using good design and appropriate use of materials.