Splice in wood joist

[ryholde]

I am working on an old masonry building where the wood joists are pocketed into the masonry wall. I have several joists that are rotten at the bearing points. The joist are simple span 15'-9" from wall to wall and are uniformly top loaded by a roof top terrace. The joists have approximately 3" of bearing in the wall pocket. I am proposing the following repair.
1) remove the old damaged joist.
2) Install a new 2x wood joist into the pocket in each wall with a splice located at mid span so the joists can be slid into the wall pockets.
3) Sister a new 2x member to each side of the spliced member. The sisters would be approximately 1" shorter than full span.

I have already verified that the single 2x wood member bearing on the wall can resist the shear and compression perpendicular to grain. The (2)-2x sister members are sufficient to resist the applied moment and meet deflection requirements. I plan on designing a cluster of nails near the end of the (2)-2x sister members to transfer the shear from the sister members back into the single 2x member that bears in the masonry pocket.

My question is what forces should the splice at mid span be designed for. I do not need the spliced member to resist the moment at mid span but the assembled member has a hinge at this point that needs to be stabilized. I had this conversation with another engineer in our office and we had different answers to the problem. I attached a sketch of each proposed solution. I would appreciate any input.

 

[woodman88]

I would use just two 2x members 12' or 14' long. Insert one end into the pocket and lap as occurs. Nail the lap section at the ends with clusters per the 1/2 the full span reactions and 12" oc between. If the numbers work.

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.

 

[ryholde]

Woodman88, are you saying to make one of the pockets 1 1/2" wider so that the faces of each member can be aligned?

 

[woodman88]

No. Take two short joists. Stick one into the bearing pockets at each wall. Slap the faces together and nail.

_____________________________lap section_____________________
|_________________________________________________________|__________________ bearing
bearing |_____________________________________________________________________|


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.

 

[RHTPE]

Woodman88 has the right idea. I would jack at the mid-point of the lap before fastening the 2 members together to insure that they are working when the jack is released, otherwise the new spliced joist doesn't see much (if any) load until the adjacent joists deflect.



Ralph
Structures Consulting
Northeast USA

 

[chicopee]

I can not see how woodman88's illustration will align with the pockets. Why not cut out several feet from the defective end and replace with a new piece of wood and use double splice plates for argument sake 6' long each splice plate; nail or bolt the splices.

 

[jayrod12]

Because double splice plates are hard to actually make calc out in wood. We rarely splice that way in wood where as in steel it's common.

 

[chicopee]

So using woodman88's illustration, the new beam will be slightly askew.

 

[jayrod12]

Correct. Slightly askew (1.5" over almost 16 feet) with a really long splice (basically full length).

 

[jfmann]

"Sideboard" splice should work fine with adequate connectors......though length of splice can be reduced.......and splice-board on one side only may be adequate.

Consider free-body of each member. At joint, end of each inner (main) member must be supported by upward vertical force Va (provided by splice member) near end of main member (not too close). Of course that force also acts downward on splice member.

Now picture deflected (curved) shape of main member (acting as unified beam)........but with splice member straight (not curved). Vertical force (Vb) must now be provided to bend end of splice member upwards to match curvature of main member. Of course such force also acts downward on main member. Moment resulting from force-couple (Va & Vb) must equal moment at joint in main member.

For midspan moment, M = wL^2/8........forces will equal M / s, where s is distance between connectors Va and Vb.

For SPF wood, shear capacity of 10d common nail is 100 lbs. For span, L, of 16 feet, uniform load, w, of 50 psf (total).....and distance between Va and Vb of 4 feet........four (4) nails at each location (Va, Vb) would suffice. This should allow for use of only one splice member (on one side of main member), though 2 is fine and should be considered if there is any doubt about quality of main member.

Nail slip must be carefully considered also......especially for relatively few nails......so more nails should be used (say 6 to 8) at each location. Wood screws might be better for this purpose. Of course through-bolts and lag screws can also be used, though bolts (especially) are likely excessive.

Although I much prefer the sideboard splice design.........it is of interest to compare with splice plate at bottom of joist. Top splice might be provided by floor sheathing and-or compression between inside ends of joists (on each side of joint).........or 2x4s nailed to side face of joist along top.

For 2x10 joist (9-1/4 inch deep).....use 8-inch deep moment-arm ("d") distance. Tension in splice is then 2000 lbs......which requires 20-10d nails on each side, using 14-gage (0.075") x 1-1/2" steel splice plate (or 2x4 wood).




John F Mann, PE

http://www.structural101.com