Wright Chat

Last Sunday I toured the Meyer May house in Grand Rapids. When we stepped out on the porch which fronts Logan Street, the docent mentioned that there was some doubt as to whether the roof was intended to cantilever from the main part of the house or whether it was to be supported by the bricks piers which exist now.

Does anyone know of a published plan for Meyer May which would date from the time of its construction? I do not believe it is in either the Wasmuth or Wendingen publications

http://www.mcnees.org/architecture/prai ... e_remc.jpg

According to these plans, the docent is right. The unusual partition between the central gallery and the living room -- and the pier opposite --
would have had structural functions. The whole ground-floor plan is arranged around this major cantilever -- which in the end was abandoned ?

From Taschen I:






Was the addition of a tile roof the straw that would have broken the camel's back ? Or did the local building department chicken out -- after
approving the project as designed ? Mr Wright must have been livid . . .

Investigation of the crawl space or cellar would show whether or not a massive concrete ballast block was cast, to be tied to tension members
rising through the first-floor partition and pier. If present, the casting would indicate that the change in verandah roof support was made after
construction commenced ?

SDR

The photograph appears to show a veranda roof identical in plan to the one on the drawing. If completed as drawn, would this have been the most dramatic cantilever of Wright's early career ?

SDR

Thanks for the floor plan.

Is there an elevation of the porch ? As built, the piers on the porch are supported visually by brick bases, which would seem not needed if the roof is cantilevered.

Right. No, I have no elevation drawings. The piers are based on plinths built into the veranda balustrade, as you say; these would show on the plan, which has no such features. Note also the central extension of the veranda past the roof, another feature missing from the plan. It would be most interesting to know when the change was made. Another possibility is that the roof was framed and partially completed when sufficient deflection appeared to cause the addition of the piers -- a less likely scenario, as I see it, than that the piers were added before construction commenced. Only construction photos, or a detailed account of the procedures, would answer that question, I suppose.

If construction of the roof as a cantilever had begun, the hold-downs I described earlier would be in place. Who do we know who could investigate the property for such evidence ?

The conservative rule for cantilevers had been, traditionally, that one-third of a beam could be cantilevered while two-thirds of its length was secured behind the fulcrum. Wright seems often to have worked with a relaxed observance of that formula.

If the beams supporting the veranda roof, here, extended from the end of the roof to the location of the pier and wall in the middle of the house, the proportion would be approximately 50/50. There's nothing present on the second floor of the house to provide a down-force there, which is why I propose a counterweight in the basement.

If the beams were long enough to extend to near the entrance of the house (and they were deep enough to resist any bending at all between their further anchor and the fulcrum at the balcony walls), the one-third/two-thirds rule would have been satisfied -- and perhaps the pier and partition in the middle of the house would have been unnecessary, except to support the weight of the beam. Perhaps beams that long were unavailable or impractical ?

Thanks for raising a most interesting subject !

SDR

SDR,
2/3rds rule is as you say conservative. This is something I'd love to study in depth (archive research), but my sense of Wright w/ cantilevers is that he was not conservative and usually reversed the 2/3rds rule by stacking stuff on top of the fixed end (Fallingwater) or tying it down with tension rods (carport Pope-Leighy).
I'm not sure but he might have been able to get buy with a cantilever here at the porch for Meyer May without counter weights in the cellar etc...

The as-built looks too carefully detailed to have been an afterthought or quick fix brought on by bureaucratic requirement. I suspect the plan was altered before construction began. It seems suitable from a design standpoint. Perhaps FLW decided the extravagant canitlever wasn't called for, considering the added expense.

How would the extent of the cantilever compare to those of Tomek (earlier) and Robie (later)?

Using Storrer's drawings and scales, Tomek had a cantilever length of 14', while both May and Robie had (or would have had) a length of 20'.

Section drawings show that it is the inner pair of piers which support the Robie roof; I assume that is the case at Tomek as well. The May plan makes clear the location of the pair of roof support beams.

Tomek is an interesting example, because its porch-roof cantilever was altered only a couple of years after construction, sometime shortly after the Tomeks moved in, according to owner/author Maya Moran. Only the porch off the living room was so altered; a similar cantilevered roof at the other end of the house remains without these added supports. Moran suggests that the clients may simply have felt the space was more "porch-like" with the addition of the columns -- which are widely spaced and do not intrude on the floor space.




Robie









http://www.mcnees.org/architecture/prai ... e_remc.jpg

At the May house, a look at the photo linked above shows very different conditions at the two porches or verandas. On the right, a porch off the kitchen has free-standing columns or piers held far back from the end of the roof, while the main veranda at the left has its columns pulled all the way to the end of the roof so that the hipped portion of the roof doesn't rest on them; in an unusually weak move, the flat extension of the roof transfers the load to the massive piers. The columns can only have been placed there so as not to intrude on the usable area of the porch, it seems to me. The result is both inconsistent with the other porch supports and with Wright's usual composition -- isn't it ?

Granted, the complete lack of support for a roof this size would be unusual, even for Wright. At Robie, though the total cantilever may be similar to the proposed one at May, the effect is a bit different because Wright equipped that house with a prow-like extension which falls between the extended beams and does not support them.

One expects to see, on the original plans above, a pair of deeply recessed piers on the veranda echoing those of the kitchen porch. Why did Wright skip those and go directly to a full-length cantilever -- before the final arrangement was achieved ?

Both Tomek and May, as built, might be instances of Wright's preference being thwarted. In both cases a clearly expressed fulcrum, in the form of a pair of engaged piers extending from the body of the house, was to be the visual and literal support of a cantilevered roof; in both cases, for whatever reason, the roof ended up being supported by an auxiliary pair of columns.

SDR

Cool thesis SDR. I think it deserves attention.
(The point of the prow at Robie carries a cross beam that connects to the main cantilevered beams. I think it is probably carrying some load.)

Ah. That could be. If so, the Meyer May proposed cantilever would have outdone the later Robie one, because there's no such auxiliary support there . . .

I'll see if I can scale the original May plan drawing above, to see if it varies from Storrer's as-built one.

(Incidentally, Storrer dates Tomek to 1904, Robie to 1906, May to 1908. His numbers reflect "the date the project first took a form fully identifiable in the final built work.")

SDR

As best I can calculate, the May drawing shows a roof cantilever of 22 feet -- two more than at Robie, and without the intermediate support Tom suggests. It would seem to make sense that Mr Wright would want to outdo himself a bit with each outing . . . ?

SDR

In An American Architecture (1955) Mr Wright presents illustrations of hundreds of his projects, including all of the major Prairie period works. The Meyer May house isn't among them. Nor does it appear in A Testament (1957). In In the Nature of Materials (1941) Wright includes the cost of the house ($16,000) and this note: "(wing added to rear, 1920"). Did he do that work ? Did Steelcase have it removed during restoration ? Was there in fact any addition ? Did Mr Wright disown this house, despite its cost and its admirable degree of finish and detail ? Do any view drawings of it survive ?

SDR

SDR, house web site at http://meyermayhouse.steelcase.com/ has the answers to your questions. Yes, an addition was constructed in 1921, designed by Osgood and Osgood, to the east, incorporating the porch off the kitchen. It included servants quarters on the first floor and two bedrooms on the second. The addition was later converted to apartments, and was removed when Steelcase performed their restoration. I recall being told that care was taken to match all of the materials of the addition to the original house, including the brick and the copper eaves. Removal of the addition provided materials used in the restoration.

More to the point of this thread, in the historical gallery on the web page, http://meyermayhouse.steelcase.com/gallery the fourth photo shows the main porch roof in place without the subject piers. So the piers must have been added during or shortly after construction.

Thanks, Forest. Most useful, and just the evidence I was looking for. I note that the porch balustrade is not built, in the photo you cite. Forced to adopt plan B for this part of the house, Wright may have had the added piers and the extended verandah prow ready for use "in his back pocket," as it were. The original design seems to have called for a simple rectangular porch plan, judging by the plan drawings posted above. The pulpit-like extension, a device he had used before, would help to mitigate the truncating effect of the piers (placed as they are so near the end of the roof) by creating a stepped profile to this end of the house.

Did Mr Wright anticipate that his major cantilever might prove troublesome -- to the structure itself, or to one of the involved parties ? He would have had this incident in mind, years later, when demonstrating to client Kaufmann that a supporting wall added at the west bedroom terrace cantilever at Fallingwater was redundant (Hoffman, pp 49-50). Perhaps we'll never know the reason for the change at the May house . . .

SDR

The veranda/porch at Meyer May reminded me of the Barton and Darwin Martin houses. The sense of shelter is immediate, and is enhanced by the piers. It becomes a contained outdoor space. Robie's "verandas" never seemed like the same space as Meyer May. They become more of an exit to the outside than a " let's sit out and watch the sun set" space.

Could Wright have reconsidered the cantilever after the initial drawings (for cost or other reasons)?