Connection Design - Load Perpendicular Fastened to Top of Member

[RFreund]

See very crude sketch attached.

Does the NDS cover this condition or is there a design value for this condition?

Thanks.

EIT

http://www.HowToEngineer.com

 

[woodman88]

Yes, you use the single shear connection values from the NDS Tables or do the calculation for the values yourself. If you have 4D edge distance (see 2001 NDS Table 11.5.1B End Distance Requirements) you can use the full value. Or you can reduce the value for a 2D to 4D edge distance.
Typically, it is assumed the connection of the sheathing to rafters/chords perpendicular the force to be zero. As the rafter/chord will more easily deflect in the direction of the force than the sheathing.

Garth Dreger PE - AZ Phoenix area
As EOR's we should take the responsibility to design our structures to support the components we allow in our design per that industry standards.

 

[msquared48]

I agree with Garth here, but just remember in your sketch to show the nail driven down, fully seated onto the plywood or decking. Contractors will sometimes take sketches literally. I've seen it happen.

Mike McCann
MMC Engineering

http://mmcengineering.tripod.com

 

[dhengr]

Rfreund:
That’s just shear loading of the nail in or through the side grain of members, with one faying surface. This is a favorable nailing orientation, the strongest orientation. In the joist member there is no difference btwn. edge nailing and side nailing. The big difference (reduction, actually not allowed structurally in pull-out) is when the shank of the nail is parallel to the grain, or end nailing. Of course, you have to pay attention to spacing, edge and end distance, splitting, and penetration into the holding member, etc. In the deck board or deck sheathing you might have this load being applied parallel to the grain or across the grain, use the Hankinson formula. Usually for a given shank size (dia., penny wt.) and min. penetration there are two shear values given in the NDS; one for loading parallel to the grain (lower value) and one for loading perpendicular to the grain (higher value), and the Hankinson formula interpolates btwn. these two. Also, look at the newer NDS eds. for Yield Limit Equations for Connections for some insight into how these joints can fail.

 

[RFreund]

Hmmm, I'm not sure why I thought this would be different. I thought the connection would act more like a loading that acts perpendicular to the grain and would be subject to splitting of the wood.

The sketch is terrible, sorry again.

Just to verify I have this correct:
You could have a situation where the joist/rafter is say nailed to a wall and the plywood is nailed on top of it. If this plywood is in tension you would have the loading direction as I attached earlier.

EIT

http://www.HowToEngineer.com

 

[dhengr]

Rfreund:
The edge and end distance limitations are intended to prevent splitting from driving and loading. More smaller common nails (and staggered, closer spacing) might be appropriate in some cases. And, I believe there is some special edge dist. criteria for large dia. and highly loaded fasteners and their nearness to the loaded edge, when in tension perpendicular to the grain. RE: your sketch, it sure doesn’t show what you are saying now. Do you mean; reverse the direction of the force arrow; then trim the plywood flush with the right side of the rafter/jst.; and install studs to the right; and the rafter/jst. is then nailed into the studs, kinda like a ledger? Smaller nails, and the edge dist. in the plywood and rafter should be manageable. You could develop considerable tension in the plywd. and into the rafter/jst. But, now the weak point is the nails into the studs in pull-out, and literally pulling the ledger off the studs.

 

[RFreund]

Dhenger - thanks I think you have answered my question. I was originally not considering the plywood/rafter/ledger situation (thus I didn't show this configuration in the sketch) but I brought it up just to make sure I was interpreting the advice given. You answered the question though by saying explaining end/edge distances and noting that the ledger would fail first.

I'd like to summarize this just for future reference.
I have a sketch with forces in x, y, z at 3 locations on a framing member. Side location, Top location, End location.
To summarize

Top
Faz=Fax Both designed with lateral loading equations and theta=0d eg using NDS (parallel to grain loading condition)
Fay => Capacity determined using withdrawal equations in NDS.

Side
Fbz => Capacity determined with lateral loading and theta = 0 deg (parallel to grain loading condition)
Fby => Capacity determined with lateral loading and theta = 90 deg (perp to grain loading condition)
Fbx => Capacity determined with withdrawal procedures in NDS.

End
Fcx => NDS end grain reduction of 0.67
Fcy => NDS end grain reduction of 0.67
Fcz => Varies - Lag screws reduce 25%. Wood screws, nails not allowed (too variable),


EIT
www.HowToEngineer.com

 

[woodman88]

RFreund - "You could have a situation where the joist/rafter is say nailed to a wall and the plywood is nailed on top of it. If this plywood is in tension you would have the loading direction as I attached earlier."
I would never do this, on top of a wall, as you are creating a moment in the joist/rafter. I would transfer the tension/compression force from the sheathing to the wall with full depth of the joist/rafter blocking. Using a Simpson (or equal) H connector at the top plate to blocking and three or more nails from the sheathing to the blocking. Space the blocking at the o.c. spacing required for the force transfer.

Garth Dreger PE - AZ Phoenix area
As EOR's we should take the responsibility to design our structures to support the components we allow in our design per that industry standards.

 

[RFreund]

Woodman - I should have attached as sketch of that situation as well because I do agree with you. However I was referring more to a ledger condition.

Any comments the second sketch as far as using NDS equations for capacity and the orientation of forces.

EIT

http://www.HowToEngineer.com

 

[woodman88]

Sorry should have looked at the second pdf. The answer still stands for the OP/first pdf.
As for the second pdf, don't. I would use lag screws (or Simpson SDS) with washers and the edge of the washers within 1/4" to 1/2" of the ledger/sheathing connection. If a Simpson H (or equal) can not be used between the ledger and studs at the ledger/sheathing connection.

Garth Dreger PE - AZ Phoenix area
As EOR's we should take the responsibility to design our structures to support the components we allow in our design per that industry standards.

 

[dhengr]

Rfreund:
I shot a bull in my first post @ 17:23, where I said “... NDS; one for loading parallel to the grain (lower value) and one for loading perpendicular to the grain (higher value), and the Hankinson formula interpolates btwn. these two.” in the next to last sentence. It should read... NDS; one for loading parallel to the grain (higher value) and one for loading perpendicular to the grain (lower value), and the Hankinson formula interpolates btwn. these two. Think of these dowel type fastener bearing stresses/situations like we think of studs and sill plates and of compression parallel to the grain (higher value) vs. compression perpendicular to the grain (lower value) and crushing. Here grain can be thought of as the wood fiber direction, and the wood being stronger along the fibers and crushing more easily across the fibers.

I don’t have the last Eds. of the NDS in hand, but I think your summary looks about right to me, with the following two exceptions for the way you itemized things.

For the top edge condition, Faz is not the same as Fax. And as I said earlier, there is no difference btwn. the jst. side face and its top edge as relates to the fiber or grain direction. Thus,
Faz => Capacity determined with lateral loading and theta = 0 deg (parallel to grain loading condition)
Fax => Capacity determined with lateral loading and theta = 90 deg (perp to grain loading condition)

And, for the end face, note that Fcx and Fcy are both related to perp to grain crushing condition, and then the reduction.

Otherwise, I agree with Garth, you shouldn’t try to attach a ledger the way we have been talking of doing, you’ll pull it right off the studs/wall.

 

[RFreund]

Garth, DH - Thank you, makes sense.
For completeness sake here is the updated summary:

Top
Fax => Capacity determined using lateral loading equations and theta=90 deg using NDS (perp to grain loading condition)
Fay => Capacity determined using withdrawal equations in NDS.
Faz => Capacity determined using lateral loading equations and theta=0 deg using NDS (perp to grain loading condition)

Side
Fbz => Capacity determined with lateral loading equations and theta = 0 deg (parallel to grain loading condition)
Fby => Capacity determined with lateral loading equations and theta = 90 deg (perp to grain loading condition)
Fbx => Capacity determined with withdrawal procedures in NDS.

End
Fcx => Capacity determined with lateral loading equations and theta = 90 deg (perp to grain loading condition) NDS end grain reduction of 0.67
Fcy => Capacity determined with lateral loading and theta = 90 deg (perp to grain loading condition)NDS end grain reduction of 0.67
Fcz => Varies - Lag screws reduce 25%, withdrawl. Wood screws, nails not allowed (too variable),

EIT

http://www.HowToEngineer.com

 

[RFreund]

Quick revision

Top
Fax => Capacity determined using lateral loading equations and theta=90 deg using NDS (perp to grain loading condition)
Fay => Capacity determined using withdrawal equations in NDS.
Faz => Capacity determined using lateral loading equations and theta=0 deg using NDS (parallel to grain loading condition)

EIT

http://www.HowToEngineer.com

 

[RFreund]

I should also say that for nails the 90 degree case is the same as 0 deg (as previously mentioned by Garth) as long as the edge distance requirements are met.

EIT

http://www.HowToEngineer.com