Recently we were hired to frame a new iii,000-square-foot house, simply before we could go started we had to set up some serious problems with the block foundation. The mason who congenital information technology apparently had a broken level, because the interior bearing walls were at least an inch higher than the perimeter walls, which were neither level nor square. The project GC had hired a second mason to remove the top block course and level the foundation, but without much success: When we arrived on site and checked his work with our transit, the foundation was nevertheless out of level by as much as 1 1/ii inches. Broken-hearted to go started, we decided we could correct these issues with some creative framing.

Using the longest foundation wall for reference, we found that one foundation wall was out of square by 2 1/8 inches over about 33 feet, while the opposite wall was out of square by nearly 3 7/16 inches over about 45 feet

Using the longest foundation wall for reference, nosotros found that one foundation wall was out of square by 2 i/eight inches over about 33 feet, while the opposite wall was out of square by most 3 7/16 inches over well-nigh 45 feet

Squaring the Walls

Using the longest foundation wall for reference, nosotros found that one foundation wall was out of foursquare by 2 i/8 inches over about 33 anxiety, while the opposite wall was out of foursquare past virtually iii 7/16 inches over about 45 feet. The short offset wall sections at the front of the house were out of parallel with our reference wall past about i 1/4 inches from one side of the outset to the other .

Since the foundation would somewhen exist covered with human being-fabricated stone veneer, we planned to extend our sill plates out past the 8-inch-wide cake foundation past at least 1 three/4 inches (to account for the thickness of the veneer, sheathing, and siding). Usually we would do this with 2x10 PT stock, but by using wider 2x12s instead, we were likewise able to compensate for the out-of-square walls.

First, nosotros aligned 2x12 plates on the 33-foot foundation wall with their inside edges flush with the inside of the wall. And so we marked them — as shown in the illustration on the previous page — so that there would exist an inch of sill plate extending over the foundation at one finish of the wall (iii/4 inch less than the desired 1 iii/4 inches), and 2 5/8 inches extending over the foundation wall at the opposite end (7/8 inch more than than the desired one 3/iv inches). We snapped a line from mark to marker and ripped off the excess, then that instead of being 2 1/eight inches out of square, the left-paw wall was at present only ane/2 inch out of foursquare over 33 feet.

We used the same technique on the opposite wall, but ended up with a larger 3 1/4-inch overhang at one end. This brought the wall to almost xiii/16 inches out of foursquare (instead of 3 vii/16 inches out of square) — non perfect merely much more manageable.

Nosotros corrected the front offsets in a similar fashion to brand them both square and parallel with the reference foundation wall. Later, the rock bricklayer made upwards for the variations in the sill-plate widths by building up his scratch coat in the sections where the sill plates extended further past the block foundation wall than they should accept, and by cutting down the stone in sections where the sill plates didn't extend by the block foundation far enough.

I calculated the depth of each notch and the height of each shim and marked them on the floor plan.

I calculated the depth of each notch and the height of each shim and marked them on the floor program.

Leveling the Floor System

Instead of shimming the sills level and filling the gaps with grout, we decided that the all-time way to fix the out-of-level block work would be to level the trusses. We would shim the trusses that were depression and cut slots in the sills with a router where the trusses were loftier, somewhen bringing all of them to the same level. We started by setting upwardly our rotary laser to take elevation readings all forth the top of the sills. We institute that in that location were very few floor truss locations that were "right" — almost everything would have to be modified to bring it to level — but we wanted to come up with a plan that would require the to the lowest degree corporeality of shimming.

Double plates. To make sure we wouldn't try to cut slots where ballast bolts were located, we doubled up the plates. Beginning, we marked our truss layout on the lower plates before attaching them to the block wall. Where an anchor bolt coincided with a truss position, we cut off the bolt flush with the top of the block using an angle grinder, and so installed another anchor commodities abreast it that was out of the fashion of our layout.

In one case the lower plates were pulled down snug to the wall, we attached the upper plates, transferring the floor truss layout from the lower to the upper layer. So we took elevations at each layout marker, starting with a section of the cake foundation that was nearest to level and using that as our benchmark for elevation readings across the rest of the floor truss positions. The whole process sounds tedious, but it took but nigh an hour.

Before heading dwelling house, I transferred those elevations to a plan view of the foundation. While analyzing the information later on that evening, I adamant an summit that would require the least corporeality of routing and shimming across all of the other floor truss locations. From that benchmark, I calculated the depth of each notch and the elevation of each shim and marked them on the floor plan.

Notching and shimming. The side by side day, we notched the sills using a Porter-Cable fixed-base router equipped with a Bosch three/4-inch double-flute straight router chip. This router has authentic micro depth aligning, which fabricated information technology easy to set it to the precise depth required for each notch. To accurately guide the router, I built a simple jig out of four brusque lengths of 2x4s. Nosotros ripped shims out of oak for the few places where the trusses needed to be shimmed; none were more than than 5/16 inch thick.

Afterward installing the trusses, we scribed the ii 3/32-inch rim material to fit. To make sure at that place would be enough of begetting, we doubled up the rims where we had corrected for the out-of-foursquare walls.

Cost

We probably did a petty more than shimming and routing than nosotros had to, but the floor organization ended upwards no more than 3/xvi inch out of level across the entire foundation. Information technology took an additional day and a half to foursquare and level the floor organization, at a rate of $xc per hour for three carpenters. While this added $ane,080 to the project budget (in improver to the amount the homeowner had already paid to the 2nd block bricklayer), the homeowners were happy to get their projection dorsum on runway.

Jason Fultz and his brothers David and Robbie are partners in Fultz Bros. Construction in Norton, Va.