Best way to attach a porch to the home ?

Started by schiada, August 26, 2018, 03:01:22 PM

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.

schiada

Thing about how to attach the porch to my 20 x 40 1 1/2 story home. Will be using a cement block basement.
Can I use the rim joist ? That is to bolt the porch joist to.
Or could I use the unused area of the cement block that is under the home? Push the sill plate inward or go to a sill plate that is 2 x 10 and rest the deck ledger on it and bolt to the house joist ?

I ask because the deck will need to be built after the basement and main floor is in. This will be to be able to work from it. The long side of the basement will be open. Building on a slope.

Randy

akwoodchuck

Yep, attach deck ledger directly over rim joist after you sheath and housewrap, using LedgerLok screws or similar:



"The lyf so short, the craft so long to lerne."


MountainDon

A porch or a deck?  To many people possibly a matter of semantics, but according to the building codes there are differences.

A porch has a roof and a foundation that follows the same requirements for the main structure as laid down in Chapter 4, IRC. "Foundation construction shall be capable of accommodating all loads in .... that is included in concrete mixtures for garage floor slabs and for exterior porches,"

A deck has no roof and may have foundations such as that listed in the AWC publication AWC-DCA62015-DeckGuide-1804. These may differ from the foundation required for a porch.

LedgerLok are generally accepted by codes but for whatever reason the IRC does not list them. Code officials here accept them without comment. The is a Technical Evaluation Report that does state, "5.4. When installed in accordance with the spacing requirements of Table 2, LedgerLok Ledger Board Fasteners provide equivalent performance to 2009 IRC Table R502.2.2.1."    There are many of these TER's on a whole host of products that demonstrate the products meet code when used as tested in the reports.


The roof for a porch also needs a lot of design thought to avoid problems during later construction.
Just because something has been done and has not failed, doesn't mean it is good design.

schiada

Don, it will be a mix of porch and deck's. The ends will be deck's and some of the long side will be porch.

How have the "code inspectors " been with using the deck/porch (uncovered) to build from ? I'm trying to get away from using much scaffolding. Thinking it would be safer ?

This is the idea at the house.

MountainDon

Quote from: schiada on August 26, 2018, 08:14:34 PM

How have the "code inspectors " been with using the deck/porch (uncovered) to build from ? I'm trying to get away from using much scaffolding. Thinking it would be safer ?

I can't see how a building code compliance would have anything to say, one way or the other,  about using a deck or porch as a construction platform.  Maybe I'm missing a nuance regarding that. Doing that would have the advantage of a flat stable base to erect the scaffolding on.

An engineer friend once told me I should use a doubled rim joist on sides where I knew I was going to add a deck or porch. He also encouraged me to use through bolts and flat washers rather than lag screws. That was before Fastenmaster had come out with their Lok series of fasteners.  Simpson also makes some special connectors, the idea being that a screw does not offer as much withdrawal resistance as a bolt and washer do. Simpson makes "lateral load connectors"... One of the most common causes of a deck failure can be traced to a lateral pull away from the ledger attaching points. Then the deck side right at the house collapses and pulls the rest of the deck against that house side.

Watch the amount of overhang.. For a deck (no roof loads) the overhang cannot exceed 1/4 of the main span (ledger board to the beam.  There is a table (2) in the deck guide with numbers.  When it is a porch with roof loads that info is not valid. Note that the joist hangers have a load bearing capacity for downward vertical loads as well as a separate and lower uplift value. The longer the overhang to more the potential for uplift. The hangers also have a lateral load value in the Simpson specs.
Just because something has been done and has not failed, doesn't mean it is good design.


MountainDon

A light bulb just clicked on   ;)

If you meant, what would building code inspectors think about working on a skeleton of joists with no deck material in place, while they might not like it (danger?) that has nothing to do with the code that covers design or construction. If that was a for hire work site OHSA rules would be violated by that. OHSA does cover DIY though.

If I was doing that I would try to use some temporary sheet material as a temp deck. Maybe reuse it later on a roof or whatever.
Just because something has been done and has not failed, doesn't mean it is good design.

schiada

Thanks,I will finish the decking first then move up.

Is this what you suggested ?


MountainDon

Yes, a doubled rim joist gives the screws more to grip into and also gives more for the sub floor sheathing to be attached to.

If using through bolts and flat washers the double rim thickness probably does not make as much difference. Rim joists are generally simply nailed onto the floor joists ends. That connection is not very strong in itself... screws into end grain are easy to pull; nails are even easier. Doubling with the extra nails through the subflooring helps. The rim joists at the gable ends probably are secured in place better, and offer more resistance than the rim joists nailed across the end grain of the floor joists across the width.

For reasons I don't fully comprehend when a fastener such as a lag screw greatly exceeds the thickness of the main member (the rim joist in this example), the AWC connection calculator shows withdrawal strength is reduced, sometimes greatly reduced. I think that may be because the threaded portion of the longer length extends through and there is less length that is threaded near the head of the lag. Therefore fewer threads biting into the wood.  That would apply to lederLok as well though the AWC calc does not have data for them. An extreme example would be attempting to use an 8 inch ledgerLok to secure 1.5" thick side member to a 1.5 inch thick main member. It would not grip at all at the 8 inch shank is mostly unthreaded.
Just because something has been done and has not failed, doesn't mean it is good design.

schiada

Thanks ,Don.

I was thinking more about the shear on the fastener at the interface of the rim joist and deck joist. But pulling out is bad also.


Don_P

The best support for that ledger is posts down to the house footings. Whenever I can that is how I support the house ledger... my connections to the house are for lateral only then. I use treated rims where decks and porches will be, it's liable to get wet sometime.

Don, check the results you are getting for reduced withdrawal with the footnotes to table 11J in the NDS, Note 3 mainly
Design values are for a minimum of 8 diameters penetration of the full threaded diameter... the tapered tip doesn't count. (That's where the callout for the longer than necessary lag comes from. With a single rim it must poke thru and fully grab that 1.5" with full diameter threads. So normally the callout is for 1/2" lags and we have a 1.5" thick rim most of the time. That's 6D)

Footnote 3;
When penetration is greater than or equal to 4D but less tha 8D the lateral design values shall be multiplied by p/8D
My note from class under the footnote says never use under 4D penetration!

See if that jibes with the results you're getting.

I've worked from a number of deck frames, I prefer to go ahead and cover if at all possible and then cover that with 7/16" osb. Consider it disposable and use those sheets throughout the job for those kinds of uses. 2 years later one of my clients is still using the random sawmill boards I covered their porch frame with to work from.

Don_P

Pffft  :D I blew that!
I mixed withdrawal that MD was talking about, with lateral.... and 1.5" thick and a 1/2" diameter lag is 3D, code minimum is 4D, that was the subject of some of the ledger research. If the strange numbers you are seeing are for 3D, then it becomes "normal" at >4D it is probably reflecting that research. That was Woeste and Bender, the report or reasoning might be in that series of deck articles they did for JLC and Structuremag.

Withdrawal strength is about depth of penetration, fastener diameter and wood density, table 11.2A in the NDS. The main thing I see is "Depth of penetration shall not include length of tapered tip."

schiada

#11
So should I take the time to use the block foundation and a wider sill plate to support the ledger board for the porch ? Or is it just over thinking ?

Porch Ledger by Randy Wefel, on Flickr

Don_P

I've debated that but it invites water in and locks the deck elevation. I typically set up the deck framing 2.5" below the subfloor surface. This allows a treated 2x to slip under the door thresholds on top of 1" thick decking.

MountainDon

Quote...reduced withdrawal...

What I was noticing about the AWC Connection Calculator was that the writers of that calculator are very smart.  I sort of "knew" that there were standards for lag screw design and manufacturing, but wasn't sure what they were.

One standard is the threaded length. They are threaded for about 2/3 of the total length, until the lags get to be very long. "The minimum length of thread shall be equal to 1/2 the nominal screw length plus .50 inches, or 5.00 inches, whichever is shorter. Screws too short for this formula shall be threaded as close to the head as practicable."

So as a lag screw gets longer the unthreaded portion closest to the head gets longer. When you use way too long of a lag screw into a main member that is only 1-1/2" thick (single thickness rim joist, for example), that unthreaded portion begins to extend into the main member. That reduces the number of threads that engage the wood. Thus the withdrawal resistance is reduced.  The AWC Connection Calculator takes that reduction into account. Enter too long a lag screw and the calculator does not compute but advises choosing a thicker main member or a shorter lag.
Just because something has been done and has not failed, doesn't mean it is good design.