 |
 |
By Bruce W. Maki, Editor
Introduction:Our old house had a problem with the roof structure on one of the porches. A previous owner had replaced the load bearing beam with a smaller piece of wood, and the roof was sagging by about 1½" in the middle of the beam span. Besides being too small, their replacement beam didn't even come close to matching the architectural detail found on the other porches. So in the midst of re-shingling that porch roof, we decided to spend a couple of days fixing the structure. This repair involved:
Compare And Contrast:
|
Back Yard -
|
Front Yard -
|
|
 |
Understandably, the back porch was built simpler, typical of houses built in the early 1900's. We figure that the back porch once had the crown molding (just below the edge of the shingles) that is found everywhere else on this house. |  |
 |
The soffit had a weird molding, nothing like the bed molding used elsewhere to cover the gap where the beam meets the soffit. |  |
 |
Of course, whoever repaired the beam also ripped off the ornamental corner trim (or "gingerbread"). |  |
 |
They just stuck a horizontal block of wood below each end of the replacement beam. That wide flat piece at the bottom of the good beam is almost exactly the same size as a modern 5/4x6 radius-edge deck board. Hmmm... |  |
| Note how the crown molding meets at the corner of the
roof... it doesn't. ALL of the corners of the roof are like this.
This is a geometry problem more than anything... if you follow the lines of the sloping crown molding, you'll notice that they don't intersect the corresponding lines of the crown on the horizontal fascia. |
 |
The only way to make these crown moldings join properly in the corner is to tilt the horizontal crown forward by the same angle as the roof. But that would change the appearance entirely.
Later I'll show you a sneaky solution to this little dilemma.
Rip And Tear:
 |
When I stabbed the fascia with a small pry bar, it sunk in easily. The wood was more rotten than I expected. |
| This beam repair was done in conjunction with a re-shingling job. Often combining such projects brings some major benefits. |  |
For instance, just before we nailed down that sheet of OSB visible in the last photo, I removed some old roof boards, crawled inside the tiny attic space and did two quick projects:
 |
1. I fished some electrical cable from the wall
cavity below, to connect to new outdoor lights. This house has never had
decent outdoor lighting.
2. I added some insulation to the outside of the wood sheathing. I used old fiberglass suspended ceiling panels, which are flexible yet can be nailed almost like foam insulation. |
 |
Using pry bars, we removed the fascia and soffit. |
| 1. The fascia had only a minor degree of rot damage.
2. The soffit was almost completely rotted. This was probably caused by the leaking shingles on the porch roof, a condition that we've ignored for a few years. Proper ventilation in this attic space would have reduced the water damage by letting the wood dry out faster. |
 |
Raising The Structure:
 |
There was a gap where the porch met the house. While the porch was in no danger of breaking away from the house, this gap is ugly, gives bugs (especially wasps) a place to hide, and could allow rain to get behind the wood. |
| We placed a long 4x4 post under the beam and used a hydraulic jack to raise the porch roof slightly. |
 |
 |
The hydraulic bottle jack must sit on a wide sturdy surface, such as this pile of wood blocks. Otherwise the jack will sink into the soil as pressure is applied. |
Wait A Second... The exact size of these boards is something that I cannot easily predict for everybody else's house-raising projects. Experience has taught me how big an area I need to distribute the weight across. Here we used 3 layers of 2x12 scraps, about 12 to 16 inches long. This small porch roof (about 13' long by 5' deep) might weigh between 500 and 1,000 pounds. The lack of shingles makes it lighter than usual. This corner post is only supporting one-fourth of the total weight, however, so the jack is not under much load. This is consistent with the effort required to pump the jack... this was a 12 ton jack (way bigger than needed) and it took almost no pumping effort to lift the porch roof. I could have used a one ton jack. Nevertheless, soil compaction is a potential danger when raising parts of houses. The weakest soils can support about 1,000 pound per square foot... if the soil has never been disturbed. But dirt around most houses has been disturbed, so it's not safe to rely on the soil supporting 1,000 pounds per square foot. I have no idea what the bearing capacity of your flower garden soil might be. The point is: pack down the soil before jacking upon it. Walking on the soil is a good start, followed by a tamper. You can buy these tools or make one from a block of wood screwed to a 2x4. Whenever I raise part of a house with a jack, I wait a few minutes to see if the soil packs down. I've seen some frustrating situations where the structure was raised up to close some gap, and then a few minutes later the gap opened again. That's because the soil packed down. So keep an eye on things, and don't walk away from a structure that has just been jacked up. |
| We wedged a supporting 2x6 board under the edge of the side beam, as close to the corner post as possible. |  |
 |
The gap closed up almost completely. Further jacking would not close the gap any more... there must be something else behind there that's holding the gap open. We'll just caulk this gap later. |
| We installed a series of vertical support posts, one under
each rafter. At the top of each post is a short block (making a T-shape),
and each post rests on a long 2x8 we laid on the ground. This is the usual
method for supporting a roof while the walls or beams are worked on.
This porch was framed with 2x4 rafters spaced at 32 inches on center. That would not meet code today. |
 |
Note that these temporary supports are as close to the beam as possible. The ideal point of contact would be the ends of the rafters, but several of these rafters were rotted at the ends. These posts mostly support the ceiling joists and ceiling material. If the rafter ends are firmly secured to the ceiling joists (and not damaged from rot) then these posts will also support the weight of the roof.
| While driving around town we noticed this stone porch being rebuilt. Note how the contractor has propped up the roof with a half dozen 2x6's. These temporary supports are on quite an extreme angle... but they work because they are held in place with wooden stakes. |
 |
These supports bear directly against the beam, so this technique would not apply in our case, but if the corner columns or the ends of the beam had to be repaired, this method would be an excellent choice for temporary support. Just supporting the sides would cause the roof to collapse if the front beam was removed.
Back At The Ranch:
 |
I took a picture looking along the edge of the roof sheathing. There are still some ups and downs in this line, but it was much worse before we installed the four temporary supports. |
| I used a reciprocating saw to cut the nails that held the ornamental trim in place. |  |
 |
I pried the side panel away from the end of the beam and cut
all the nails that I could find.
The idea is to get access to every face of the beam and remove or cut all the nails. |
| I jumped down and let the homeowner do all the heavy lifting. Taking
pictures is much easier than doing this work.
Note how the beam had a 2x6 nailed to the top... the part that she is holding with her right hand. |
 |
I should point out that beams like this can be quite heavy, and after the nails are cut they can fall down with little or no warning, so be careful. It may be a good idea to attach some sort of brace to the beam to hold it up. Large beams may need to be cut into pieces and dropped to the ground one piece at a time. Having some helpers is crucial. Or a forklift.
 |
The site after the beam had been removed.
1. The point where the beam connected to the wall. 2. The ceiling joists. There were 6 of these, one at each end and 4 in between. These were just 2x4's laying flat, spaced 32" apart. The joists extend all the way to the fascia... the fascia was nailed to the ends, and the soffit was nailed to the underside. |
Note how the T-post supports are very close to the site of the old beam.
| A closer look at the end point. The beam rested on the
turned post (half a turned post, actually) which is nailed to the wall.
The beam was also nailed directly to the wall, but most of those old
spikes were rusty and loose.
I added the red line so you could visualize the top edge of the rafter that is nailed to the side of the house. Luckily, this rafter had no water damage. |
 |
 |
This is a device for measuring and duplicating angles. The wing nut allows you to tighten the steel blade. |
| I used this device to record the angle of the rafters. It turns out that the rafters are exactly 25 degrees above horizontal, which is less than the 6-in-12 slope that I had guessed. |  |
 |
I cut a block of treated lumber at this angle, to fit snugly between the rafter and the ceiling joist. |
| But before I nailed this wedge in place, I nailed a slightly larger triangle of plywood to it. The plywood will allow me to fasten everything more easily. |  |
 |
1. The plywood-and-lumber wedge filler. After I
pounded the wedge in tight, I just nailed the plywood to the rafter and
the ceiling joist, because I couldn't reach inside to nail or screw the
wedge itself.
2. I installed a new "outrigger", or end of the ceiling joist, to replace the rotted wood I cut away. I used treated lumber. |
In fact, it wouldn't be a bad idea to build all porch structures from treated lumber, given the risk of leaks and water damage.
| The red arrows indicate the two rafters that I repaired in this manner. The other rafters and joists were okay. |  |
At this point we were ready to cut the lumber for the beam.
These drawings describe the differences between the old and new beam construction methods:
|
Note the empty space between the soffit and ceiling boards. The beam should have filled this space. Instead of supporting the ceiling joists directly, they pinched the edges of the soffit and ceiling T&G strips with their 2x6. I don't know what they were thinking when they made this repair, but a 4x6 is woefully inadequate for a beam that spans almost 13 feet, and tacking a 2x6 on top of such a small beam does not add enough strength to it. |
|
|
The New Beam:
In order to have the tallest possible beam, we carefully measured the space between the bottom of the old ceiling joists and the supporting points that the beam had to rest on (i.e. the top of the turned post). But the bottom one inch of this distance was to be occupied by a horizontal 5/4 x 6 radius-edge deck plank, to replicate the original architectural detail. The resulting height was about 8.5", so we bought three 14 foot long 2x10's for our beam, which was almost 13 feet long.
 |
2x10's are 9¼" wide, so we ripped them slightly
narrower with a circular saw.
Note the ripping guide that attaches to this circular saw (red arrow). This handy feature makes it easy to cut accurate width boards. This saw is almost as precise as my table saw. |
| We set the first board in place and positioned it carefully so the final beam (which will be 3 times as wide) will be centered on the support columns. | 
|
 |
We drove in a few 4" stainless steel deck screws to
attach the first board to the house structure.
A potentially useful fastener here would be the Timberlok brand of long mini-lag screws. Regular lag screws might work too, but they require a lot of pre-drilling. |
Where possible, it's important for the fasteners to reach the wall studs, or at least a sturdy piece of wall sheathing. Old houses like this usually have solid wood sheathing on the outside of the studs. Installing L-shaped metal framing brackets would be an even better idea, but in our case we would've had to remove some siding, which would complicate the job too much.
| We nailed up the second piece of lumber.
Note the hydraulic jack and post that are lifting up slightly on the beam. This is to force the two pieces of wood to lie exactly parallel to each other. |
 |
Since this lumber will be visible, I won't tolerate built-up beams with pieces of lumber that won't line up precisely. By anchoring one end of this 2x10, and forcing the loose end into place with the hydraulic jack, I can completely control the wood. I nailed from the left end to the right end, and I moved the jack along as I progressed.
 |
The third 2x10.
We used a couple of Quick Grip clamps to hold the lumber tightly against the other two |
We mostly used 3½" spiral galvanized nails to fasten these boards to each other, but we also drove in some 3" deck screws to keep the 2x10's from separating over time. It's important to use enough nails and use nails that are big enough. We drove three rows of nails, spacing the nails about 8 to 10 inches apart horizontally.
| The decorative bottom piece:
We nailed up the 5/4x6 deck plank (red arrow). After a little smoothing of the corners with a belt sander, this plank looked just like the original. |
 |
 |
The completed beam. We drove in several 4" stainless
steel deck screws at each end (on the first and third 2x10) to reach into the house framing as much as
possible.
Note the collection of ladders. The yellow ladders supported an extension plank, and the aluminum ladder was used to access the plank. This plank made it easy to reach every point along the beam, without having to move ladders all the time. |
Safe Access:This porch roof is over ten feet above the ground. Getting safe access to such places takes some equipment. The extension wooden extension plank we used was mediocre... it's quite flexible and several times I almost lost my balance. 2x10 or 2x12 planks are often used for situations like this, but lumber can break and send you to the ground, and then to the hospital. Lumber is not rated for use as scaffolding planks, but you can buy OSHA approved plank boards that are rated for this purpose. The safest access method for a job like this is OSHA approved planks on proper steel scaffolding. Scaffolding can be purchased or rented by the day, week, or month. |
| Once the beam was completed we removed the temporary supports. |  |
Installing New Trim:
 |
We installed a piece of treated 1x4 to replace the fascia. We primed the back side before nailing this in place. Backpriming will protect the wood (hardly necessary with treated lumber) and should also make the paint job last longer, because the wood will absorb less moisture. |
I've learned over the years that minimizing the amount of water absorbed by wood will make the paint last much longer, so I try to prime the back and sides of all exterior wood that will get painted, even treated lumber.
| We nailed up a "new" piece of crown molding to the
fascia.
This wasn't really new trim... we salvaged this wood from an old garage that some friends tore down a couple of years ago. |
 |
 |
We proceeded with the roofing job, and installed a layer of OSB over the old wood roof sheathing. |
Some Trim Trickery:
 |
There was no way to make the angled crown align with the
horizontal crown... it's a geometric impossibility. So I cut the crown
moldings short and tacked a thin piece of wood to the end of the
horizontal crown. (Left hand photo)
Then I figured out the angle to cut the crown so it would meet the block of wood. |
 |
Cutting crown can be quite a brain-teaser... this involved holding the crown molding upside down and backwards on the miter saw, adjusting the bevel to a 25 degree angle from vertical, and cutting that on a 45 degree miter.
 |
Once the crowns were installed, I scribed a line parallel to
the edge (using the end of the horizontal crown as a guide) and cut this
with a coping saw.
But that cut, viewed from the opposite side, is not parallel to the sloping crown molding. |
 |
But from the ground, the trim looks good, and there are no gaps where the crowns meet.
My motto for funky trim is: When in doubt, make it stick out.
| I installed the roof drip edge over this trim. |  |
 |
The view from below. For some reason I could not get the drip edge and the crown molding to line up perfectly. I think the fascia board was warped. Small flaw... nobody will notice. I call that "character". |
Beware The Perils Of Open Attic Spaces:We did this work in the fall of 2001, but we did not have a chance to complete the soffit, so we waited until the spring of 2002 to do the last details. There can be some problems with leaving a soffit open for a few months. In March I noticed some birds flying around the porch soffit. They were starlings, and they seemed interested in building nests in the attic above the porch. At that point I knew I needed to close up this project. Take it from me... you do not want animals making nests in your house. I have experience with that subject on a different house. In early 1996 I was wrapping up the last few details on an addition to my first house, which included a covered porch. I had left an attic access opening so I could store light-weight outdoor items (like Christmas lights) above the porch. But I was also involved in a new business that had me away from home for weeks at a time. During one long weekend at home I installed soffits, fascia and trim, and built a hinged access door to the attic. I noticed some dead leaves sticking out the side of the recessed light fixture in the ceiling, but I ignored the issue. I should have investigated further, because leaves don't get into light fixtures without someone's help. I returned almost a month later to discover that a squirrel had chewed the edges and corners of EVERY piece of wood in a vain attempt to get inside the attic. I climbed up a ladder and removed the leaves and two dead baby squirrels. It took me two long days to patch the squirrel's gnawing with wood filler and repaint the trim. I guess Ms. Squirrel got her revenge. |
| Later we installed a new soffit made from a pair of pressure
treated 1x4's joined together with urethane glue. We couldn't find 1x8
treated pine at our usual sources so we resorted to making wide lumber
from smaller pieces.
We nailed up some new bed molding where the soffit meets the beam. We primed all sides of this trim. When the weather warms up we will give everything a final coat of paint (actually, solid-tone stain). |
 |
We also cut some rectangular holes in the soffit and installed 4"x16" vents. We used short stainless steel screws and spray painted the heads white so they would blend in. These fasteners won't ever leave rust streaks. The soffit was pre-painted indoors so there would be no need to carefully cut-in around the vents later on.
Now all we need to do is re-create the ornamental bracket (or gingerbread) that belongs at both ends of the beam, and the porch roof will be restored to nearly original condition.
Nerd Alert!
Notes On Beam Sizes:You can't just use any 'ol piece of lumber for a load bearing beam. Tell that to the guys who replaced this beam years ago. When replacing a damaged structural member like a beam, joist, or column, there are only two safe approaches.
We used the first approach, sort of. We could have dissected the other porch (which is identical in size) to determine the exact materials used in the construction of the beam, but that would be destructive. I've done repair work on porches on other old houses, and what I've seen is surprisingly light-duty... a single 2x8 or 2x10 that is "clad" with a U-shaped jacket of straight-grained pine 1x lumber. Given the 32" center-to-center spacing of the rafters, I wouldn't be surprised to find such minimal construction in the other porch. Beam design involves some engineering calculations. There are several factors involved:
Given the 30 psf live load, plus the 10 psf dead load, the 6' width of the roof, and the 13' span, I can make some safe assumptions:
The maximum bending moment (a way of describing bending forces) for a simple rectangular beam with a uniformly distributed load is: M = W * L Where W is the total distributed load, and L is the length of the span. This gives us an idea of how big the cross-sectional area of the beam must be, based on this formula: S = M Where Fb is the allowable extreme fiber stress. That is the maximum force (per square inch) that can be safely applied to the worst-stress location of the beam, which in this case are the fibers on the bottom edge of the beam. S is called the Section Modulus, and it is a property of the geometry of the beam, not the material. Fb is a property of the material. Fb varies across different species of wood:
Based on these formulae, I computed a bending moment M of 2535 foot-pounds (1560# x 13' / 8) So S = 2535 lb-ft x 12 in/ft Which yields a Section Modulus of 20.28 in³. But what does that mean? That tells me how big the cross-sectional area of the wood beam must be, sort of. The formula for computing the Section Modulus of a given size and shape of simple rectangular beam is: S = B x D² Where B is the width of the base, and D is the height (or depth) of the beam.
So it's quite possible that the reason the beam sagged after all those years was the fact that they did not nail the pieces together very well. I imagine it would take dozens of big nails to join these boards properly, but nails always seem to loosen over time, based on my experience. Heavy deck screws or lag screws would have been better. If they had just used a 4x8 beam, there never would have been any problem. Fun Facts: Note how the calculation for S depends on the orientation of the piece of lumber. Taller is better. If the height (D) is doubled, the section modulus (and the overall strength of the board) is quadrupled. This is why floor joists are stood on edge... when laying flat a 2x10 is much weaker and much more flexible than a 2x10 on edge. If we turned our beam on its side, the section modulus would be only 28.7, about half. Same material, different orientation, half as strong. Neat, eh? If you read all the way through this, and understood anything, you deserve a medal. Sources:
|
|
Tools Used:
|
Materials Used:
|
|
|
|
|
|
|
|
|
|
Copyright © 2002 HammerZone.com Written April 2, 2002
|