Wood Moment Connection Using Simpson Straps (or Equivalent)

Had to help a contractor with a field fix recently. Something very similar to Detail 1 was proposed, and I couldn't sleep at night - we twisted an arm to have them swap one of the simpson straps for a full-height flatwise member nailed all the way up. Strap on the opposite face remained. Not the best, but a lot better than Detail 1.

Detail 2: I'm with you...maybe with some careful detailing. Would you feel better using an MSTC48B3 or similar for the Simpson straps?

Yeah, I would feel better with the underside grab. I guess, for me, the big difference between detail one and detail two is that joint success for detail one is highly dependent on joint stiffness whereas, with detail two, joint stiffness would really just be an aesthetic issue. It might be an aesthetic issue culminating in legal action but, still, nobody gets hurt.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

If you are talking standard wall heights (8' and greater) I don't think detail 1 is a good idea a all. Unfortunately, I see this mistake a lot with local framers. Just today a small vaulted section of the ceiling at the gable end probably should have been balloon framed but was not. When the wind sucks the wall outward the 7/16 osb sheathing is going to load up in tension however when the wall sucks outward the drywall will be in tension. Fortunately, in this situation the piggyback wall or pony wall is only about 2' in height so a few straps on the inside will probably suffice. How does one actually calculate this situation? If the shorter moment arm is the thickness of the framed wall it doesn't take a genius to see there is a serious mechanical disadvantage.

I digress slightly but partial vaults in gable end trusses have always caused me some pause. When the roof diaphragm is loaded longitudinally the double top plate of the wall at the gable end will be loaded in tension or compression. If this double top plate is discontinuous because of having to balloon frame under a partially vaulted gable end truss how is this supposed to work. My solution of course would be to strap the top plates together at the pitch breaks with a MSTC40 or something along those lines.

I've never tried detail 2 or seen it done. An application might be a half wall on a overlooking loft? Even if it calcs out and it has a couple of straps on each stud I'm guessing there will always be a slight gap between the wood members and the guard rail will always feel a bit shaky, giving the impression of weakness. A simple summation of moments shows that a 200lbs lateral load at 48" will exert a 2,743lbs moment couple with a 3.5" wall. This amount of force will cause some give in the fastener/wood interface and again the connection will probably not feel very firm, but again this could be a very subjective thing.



A confused student is a good student.
Nathaniel P. Wilkerson, PE

I have used detail 1 before but with straight straps (not coiled) and only when splicing near the top of the wall where the moment is minimal. I have used the bottom detail on residential interior guardwalls. Used LVL's so the cross grain bending is not as much of an issue. They end up being pretty darn stout.

The problem i have with straps like this is they are never installed tight, i don't think it is possible to make it tight enough that load transfer goes into the straps immediately on deflection fo the system.

I will say that i have done similar to #1 with heavy timber columns, step notched cut the splice, and sandwich with steel channels and bolts. I assume it acts similar but i am much more comfortable with those connections.

#1 in a gable end or great room wall has been proposed to me before. I have seen this repair (similar) provided to contractors before. I think it may not fall over from out-of-plane loads, but it will fail serviceability. I also think that it taxes the sheathing in ways that the sheathing wasn't intended.

#2 would not be my choice for this, i'd rafter make the connection where the post extends down to the bottom of the beam (like an exterior deck and Simpson DTT2Z clips)

I doubt you could ever get true fixity but a partial moment connection seems possible.

Professional and Structural Engineer (ME, NH)
American Concrete Industries

In addition to the reservations expressed above, I think cross grain shrinkage of the wood would make the connection too flexible. Just as bolts through posts loosen with time, the straps will loosen with time, and you can't go back and tighten the straps as you can with the bolts.

In order to maintain a moment connection, the straps have to be in tension. With time, shrinkage and creep strain, those straps almost always go into compression, which as hokie66 alluded, will also negate the moment resistance. In short, you'll get a lot of hinge bending before you achieve "take up" for moment transfer. In failure investigations, I've seen enough shrinkage/creep to cause the straps to "bow" out.

Thanks for all of the excellent advice guys. Indeed, it does sound as though shrinkage is the Achilles heel of these connections.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

Shrinkage, fastener slip, and the fact these aren't tensioned while installation tend to make me a non-believer in both of these details.

But at the same time, I've seen it done and it appears to work in practice (even if it doesn't calc out).

I also fail to see the difference between 1 and 2 from a strictly connection standpoint. If you're willing to consider the guard rail fixed at the base using 2 straps, why wouldn't the same detail work for a stud wall. In for a penny, in for a pound.

jayrod12 said:

But at the same time, I've seen it done and it appears to work in practice (even if it doesn't calc out).

Click to expand...

I guess the next question would be do you do it? I'll confess to having stamped a few of the guardrail details already.

jayrod12 said:

also fail to see the difference between 1 and 2 from a strictly connection standpoint. If you're willing to consider the guard rail fixed at the base using 2 straps, why wouldn't the same detail work for a stud wall.

Click to expand...

The difference, in my opinion, is the impact of connection rotational stiffness. On the guard rail, the load is what it is and that's the end of it. With the wall to wall connection and meaningful axial load, connection flexibility will lead to amplified moments at the connection as a result of P-little-delta effects. Shanely would not be proud.



I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

No I have not done either of these details, the guardrails either have posts that extend down and then bolt-couple them to the beam, or it's a decorative guardrail and the connection details are as per manufacturer. I just make sure I provide enough meat for them to fasten to (most often a screw couple at the baseplate).

Anytime I have come across situation 1, I make them cut in full height columns and turn the plate pack at the hinge into a wind beam spanning laterally.

I'd be curious what the test results from both of these details would come back at.

One thing that maybe makes me think the two details are likely similar in final performance. For detail 1, I'm providing those straps at every stud (it's also likely that the studs are at much tighter than standard spacing to work for deflection anyway. The guardrail is likely only done every 4 ft (depending on the guardrail design). The moment expected at each detail 1 would be say .6 k/ft service. The moment at each detail 2 would be upwards of 3 k/ft service. That's almost 5 times the moment going through the same detail.

There is in-plane stiffness that is tough to calculate but must exist for both drywall and wood sheathing (wood sheathing obviously much stiffer) otherwise they couldn't be used for lateral loads.

Not mentioned too is the working of the top a d boot Tom columns on the plates in bearing. That working will create additional looseness in the connection as the perpendicular to grain stress in the plates will probably be exceeded. The working of the connection will crush the plates more and more with time, making the lateral movement seen more and more. I would not use either detail either.

Mike McCann, PE, SE (WA)

jayrod12 said:

Anytime I have come across situation 1, I make them cut in full height columns and turn the plate pack at the hinge into a wind beam spanning laterally.

Click to expand...

Clever. I'll keep that in my back pocket for future use.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

Thanks,

I'd like to take credit for that idea, but alas it was passed down to me from someone much more experienced and intelligent than I. It works pretty slick.