In Usonian houses was it typical that brick or block walls were built before the floors were poured, or vice versa? And would that have been the same in the case of terraces as well?
Our interior floor slab stops before it meets the walls and is filled at the expansion joint. That would seem to indicate that the walls were erected first and the floor filled in afterwards.
Others?
Walls or floors?
It's hard to imaging a floor slab, atop a crushed-rock base, being poured without something to contain it. A logical sequence, in the absence of a (redundant) perimeter form, would be to begin the brick wall and take it a couple of courses beyond floor level, then pour the slab, then complete the wall.
Photos and text on p 113 of Sergeant, however, show the sequence at the Richardson Usonian (1951); the hexagonal-grid slab is completed in two pours (base and color layer) before the brickwork is begun. This conforms to Wright's suggestion that the trades be brought to the site in discrete order.
SDR
Photos and text on p 113 of Sergeant, however, show the sequence at the Richardson Usonian (1951); the hexagonal-grid slab is completed in two pours (base and color layer) before the brickwork is begun. This conforms to Wright's suggestion that the trades be brought to the site in discrete order.
SDR
From the construction pictures I have seen of other Usonian houses, there is often a photo of a pristine slab with no walls or partitions on it, through it, or beside it.
At Sweeton, the CMU (interior and perimeter) is laid ON the slab, which is contrary to common construction practice of laying up the masonry to a level above the finish floor, then pouring the slab to the masonry wall at the slab's edge, usually with an expansion joint or insulation between the two. This detail requires the foundation to be laid up, the slab poured, then the masons resume work, trying to avoid dropping blocks on the floor which might chip it. My guess is that Wright liked the clean appearance of no joint between wall and floor other than the masonry bed joint....otherwise the edge strip of expansion joint filler or insulation at the slab edge would need to be covered with a base of some sort.
At a location I excavated next to the workshop, I found that Sweeton was constructed with a course of CMU carved out allowing its outer face to be the edge form for the slab. The detail poses a moisture infiltration issue at locations where the slab is below grade as I have learned at the Sweeton workshop. I excavated and installed a fluid applied membrane, sheeting, drains, even a concrete grade gutter beneath the roof drip line and still I get about a cup to a pint of water flowing through the CMU bed joint with the floor in the NW corner during a downpour (dry otherwise).
The as-built condition was a deviation from the drawings which had the slab turning down 4" in from the outer face of wall and becoming a "course" of masonry beneath the 4" thick facing block. I'll send the appropriate drawing sheet to SDR for posting if he will indulge.
At Sweeton, the CMU (interior and perimeter) is laid ON the slab, which is contrary to common construction practice of laying up the masonry to a level above the finish floor, then pouring the slab to the masonry wall at the slab's edge, usually with an expansion joint or insulation between the two. This detail requires the foundation to be laid up, the slab poured, then the masons resume work, trying to avoid dropping blocks on the floor which might chip it. My guess is that Wright liked the clean appearance of no joint between wall and floor other than the masonry bed joint....otherwise the edge strip of expansion joint filler or insulation at the slab edge would need to be covered with a base of some sort.
At a location I excavated next to the workshop, I found that Sweeton was constructed with a course of CMU carved out allowing its outer face to be the edge form for the slab. The detail poses a moisture infiltration issue at locations where the slab is below grade as I have learned at the Sweeton workshop. I excavated and installed a fluid applied membrane, sheeting, drains, even a concrete grade gutter beneath the roof drip line and still I get about a cup to a pint of water flowing through the CMU bed joint with the floor in the NW corner during a downpour (dry otherwise).
The as-built condition was a deviation from the drawings which had the slab turning down 4" in from the outer face of wall and becoming a "course" of masonry beneath the 4" thick facing block. I'll send the appropriate drawing sheet to SDR for posting if he will indulge.
Probably - but I was not there that day. Check the July 28 pictures for the "beginning" of what I saw. It appears the red concrete was poured, cured, and then a perimeter was laid around that to hold the exterior walls.
My photos: http://www.flwright.us/facultyhousingarchive.htm
Michael Maguire's website may have covered this in more detail, as he was more interested in the construction process: http://www.buildingtheusonianhouse.com/
My photos: http://www.flwright.us/facultyhousingarchive.htm
Michael Maguire's website may have covered this in more detail, as he was more interested in the construction process: http://www.buildingtheusonianhouse.com/
The first few courses of the exterior walls become the form for the interior floor and exterior terraces. At least that was the situation at Dobkins. And the result is walls are on footers and the pour of the concrete mat is independent of the walls on gravel. When the floors move, and they will, they will not heave the walls.
Peter, that is why I suggested, that your parapet walls should have been built first. Pour of terrace and steps independent of masonry walls.
I'm curious about your wall and step detail.
Footers were poured for the steps here at Dobkins, and they are independent of the parapet walls.
Peter, that is why I suggested, that your parapet walls should have been built first. Pour of terrace and steps independent of masonry walls.
I'm curious about your wall and step detail.
Footers were poured for the steps here at Dobkins, and they are independent of the parapet walls.
As I said previously, we are following the recommendations of John Eifler. Not being an architect, it could very well be that I am not completely capable of explaining our situation...
Are you then saying that the Sweeton detail (but with scuppers and sloping slab...) would be unsatisfactory and possibly faulty in the case of an exterior terrace?
From what you are describing, it seems that Dobkins was built the same way Lamberson (failed terrace and all...) was. Did you end up rebuilding your terrace in a different manner from what was originally built, or did you duplicate the original construction method? I remember you saying that you needed to add a drain because the original scuppers failed and destroyed the walls...
Our steps are also independent of the slab and masonry walls.
Are you then saying that the Sweeton detail (but with scuppers and sloping slab...) would be unsatisfactory and possibly faulty in the case of an exterior terrace?
From what you are describing, it seems that Dobkins was built the same way Lamberson (failed terrace and all...) was. Did you end up rebuilding your terrace in a different manner from what was originally built, or did you duplicate the original construction method? I remember you saying that you needed to add a drain because the original scuppers failed and destroyed the walls...
Our steps are also independent of the slab and masonry walls.
-
outside in
- Posts: 1338
- Joined: Sat Jul 29, 2006 9:02 pm
- Location: chicago
Building the walls first may be tied to sites that are not level, requiring a slab that may be elevated above grade. The detail works well when there is a roof overhead, but at the exterior, a slab that abuts a parapet wall creates a vertical joint, allowing water to enter and eventually causing failure. I suspect that is what probably occurred with the first terrace. By pouring the slab on top of the foundation a more desire able horizontal joint is created by placing bricks on top of the slab, and water can be directed towards scuppers. Kraus is built this way, and performs much better over time.
I failed to mention the parapet terrace wall and about half of the terrace mat was re-built during Dr. Dobkins time. Date of the restoration unknown.
When replacing building sealant at floor and parapet joint It was discovered that the new and the original terrace did not come in contact with the wall. I had to tear out the backer rod before I could replace the old building sealant along with new backer rod and 100 percent Dow Corning silicone sealant.
I'm sure part of the problem that developed at the Dobkins terrace was improper installation of thru wall scupper copper inserts.
Period photos show 2 scuppers at each wall.(a total of 4) So to eliminate that problem in the future they installed a centrally located drain in the terrace mat.
Both interior and exterior floors were steel troweled and the joints were hand tooled.
Since the Dobkins re-build I had the rowlock removed and re-built once again. Sorry to report water is still entering the parapet. I'm sure we will need to re-build in the future with yet another flashing system.
This has been a perennial problem. Again early pics of the house always reveal recent tuck pointing on the parapet. Do see Storrer's FLW Companion,(pg.388) the terrace prow has new head and bed joints. Was it partially re-built again, or just tuck pointed? I do not know. It is just a matter of time...I will need to address it again.
Every mason that has seen the problem points to the deeply racked bed joints and the rowlock. They basically ask who designed this? & inform me it will continue to be a problem. They leave with smiles on their faces as the say "good luck".
When replacing building sealant at floor and parapet joint It was discovered that the new and the original terrace did not come in contact with the wall. I had to tear out the backer rod before I could replace the old building sealant along with new backer rod and 100 percent Dow Corning silicone sealant.
I'm sure part of the problem that developed at the Dobkins terrace was improper installation of thru wall scupper copper inserts.
Period photos show 2 scuppers at each wall.(a total of 4) So to eliminate that problem in the future they installed a centrally located drain in the terrace mat.
Both interior and exterior floors were steel troweled and the joints were hand tooled.
Since the Dobkins re-build I had the rowlock removed and re-built once again. Sorry to report water is still entering the parapet. I'm sure we will need to re-build in the future with yet another flashing system.
This has been a perennial problem. Again early pics of the house always reveal recent tuck pointing on the parapet. Do see Storrer's FLW Companion,(pg.388) the terrace prow has new head and bed joints. Was it partially re-built again, or just tuck pointed? I do not know. It is just a matter of time...I will need to address it again.
Every mason that has seen the problem points to the deeply racked bed joints and the rowlock. They basically ask who designed this? & inform me it will continue to be a problem. They leave with smiles on their faces as the say "good luck".
-
outside in
- Posts: 1338
- Joined: Sat Jul 29, 2006 9:02 pm
- Location: chicago
The masonry knee walls will perform better if flashing is installed under the row lock course. The vertical joints should be individually filled with a backer rod and sealant recessed at least 3/4 of an inch below the mortar- or one could apply clear sealant to the vertical mortar joints to prevent moisture intrusion. Finally, it is also a good idea to lay the rowlocks with a slight pitch to encourage the water to run off the top surface.
