Horizontal Deflection - How much is acceptable?

Started by redside, September 07, 2017, 04:58:22 PM

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.

redside

Me and my wife decided to go with parallel chord trusses for our 16x24 cottage.  The truss plant is able to get the horizontal deflection of the top plate down to .66 inches with a horizontal thrust of 465 pounds.  Keep in mind this is at 110 lb total load, so your worst case scenario.  Under normal dead load, the deflection and thrust is minimal.  The thrust of 465 pounds can easily be connected at the top plate with hurricane ties; however, this does not keep the walls from pushing out.  Our place will have a 12x16 loft dropped down 2 feet below the top plate which will keep the walls together over half the building; however, the other half of the building will be vaulted and completely open.  The trusses are 24 inch OC, so I figure that there will be approximately 2,790 pounds of thrust uniformly loaded at the top plate of the open cathedral half of the house.  This is a lot of thrust to be restrained.  Will the gable wall and loft be enough to restrain this force?  Does it even matter since the deflection is only .66 inches; how much can a wall push out before it becomes an issue?

I figure worse case scenario, I can tie the top plates together in the open cathedral portion of the home with a beam at the 6 foot mark.

The truss manufacturer doesn't think the deflection is anything to worry about due to it being so minimal.  Simpson also makes sliding connectors at the top plate that allows the truss to deflect without pushing the top plate out; however, this weakens the wall to roof diaphragm since the connection isn't pinned like a normal hurricane tie would be.  I would like to use pinned connections if possible.

By the way, I was thinking of doing this with rafters and there is basically no way to safely retrain the horizontal thrust without rafter ties.  The parallel chord trusses, however, do significantly reduce this thrust and make it possible to have a vaulted ceiling without a ridge beam!!!  Of course, a ridge beam is better since it effectively eliminates horizontal thrust, but if the building is strong enough with the trusses, then we saved a lot of foundation work, will be able to have an open floor plan, and have the beautiful vaulted ceiling look without having to figure out a way to safely install a thousand pound ridge beam!  I am hopeful this will work, but I defer to the experts on this forum......I can always install rafters and rafter ties and still have a cabin in the woods if trusses are a bad idea.


Don_P

Under full snow load the way I'm seeing it that would blow a normally connected top plate at the end connections to the gable walls. If you are notching in a ledger for the loft floor I wouldn't be surprised if that split the studs... imagine the stud strapped horizontally to a table notch up at the edge of the table and setting a drum full of water on the end of the 2' projection. Will their engineer weigh in?

If you can put a center post in under the ridge a double 11-7/8" deep LVL would work from my understanding of the problem with about .4" deflection at the ridge, that stock weighs ~6lb/foot. A double 14" at ~7 lb/ft would deflect about .25".


redside

The studs will likely be doug fir #2 or better and 2x6 16 inch OC.  They have enough strength to support both the vertical and the horizontal loads.  However, a 2x4 does not and I guess notching out a 2x6 to hang a loft on may weaken the stud enough to risk a failure as Don P pointed out.  A better option for the loft is to have it bear on a sister stud and leave the original 2x6 intact.  Many thanks to Don P for pointing this out. 

I am not sure if I agree that the connection at the top plate where the gable wall intersects will fail.  The reason I feel this way is because the walls have a plywood shear panel that provides additional strength.  In addition, where the walls intersect should easily be able to withstand the horizontal forces applied.  Am I incorrect in that it is not just the top plate connection that should be considered, but rather the entire wall to wall connection?

A ridge beam is not feasible due to the remote location and wanting an open floor plan.  The next best option is having a beam tie the top plates together over the open portion of the vaulted ceiling.  I don't see why this will not work with trusses since it does with rafters.  We are only talking about a conservative 1300 pounds of horizontal thrust at the gable wall intersection, without a beam.  If I add a cross beam then the force will be about 1/4 of that amount, or 325 pounds at the gable wall intersection.

I do see the concern; but the trusses seem like a great option.  For example, even if a person uses rafters and raises the rafter ties to the upper 1/3 limit above the top plates, the forces will exceed what I have proposed hear....and that passes code.  Maybe I am not understanding things correctly, but rafters with rafter ties still exhibit horizontal thrust, albeit is restrained greatly.  Or, maybe the rafter ties convert 100% of the horizontal thrust into a vertical load?????  If you look at my numbers you will see the scissor trusses have less horizontal thrust than a code compliant rafter does. 

akwoodchuck

I'm confused...you're too remote for a ridge beam but not for trusses? Lol
"The lyf so short, the craft so long to lerne."

redside

I also wanted to mention that Simpson makes special top plate anchors that allow the truss to move horizontally.  This is to keep the walls from pushing out.  From what I read, some people will have a slip connector on one end of the truss and a hurricane tie on the other.  They will then alternate sides so that there is a least some transition from the wall to roof diaphragm.  I don't know if this is a good idea, but it seems like it is and solves the problem of wall spread.


redside

#5
I think everyone will agree a ridge beam is the best option, but I will tell you it is not the best option for me in my situation.  I am not hauling the trusses out whole.  They will be in halves and I will perform a field splice to connect them.  They are much easier to install than a ridge beam and will save me much foundation work as well.  This post is about using trusses to get a similar effect to a ridge beam.  I would like to keep it there if possible.  I have my reasons for not wanting to use a ridge beam....the main reason is I don't want a beam coming down in the middle of an already small 16x24 footprint........nothing to do with being remote....LOL

akwoodchuck

I'm pickin' up what you're layin' down.... 8).....well in your place I would likely mess with neither trusses nor beam....I'd stick frame the roof, extend the loft joists out to tie into the rafter tails (boxing in the soffit quite nicely), and throw a couple rafter ties out in the vaulted portion....easy peasy lemon squeezy.....
"The lyf so short, the craft so long to lerne."

redside

#7
Great suggestion on using rafters, but I will not be able to get the room in the loft I am wanting.  Also, I am concerned about having a rafter tie every other rafter pair over the open area (the forces on the rafter to top plate connection are pretty high when you eliminate a rafter tie......I think that might be a reason why code is every rafter pair). 

I talked to the truss company and will get an opinion from a designer regarding the trusses.  The truss company said the gable end walls will have zero deflection due to them holding the walls together.  As DonP pointed out, the potential forces are large where the walls intersect and I will make sure to restrain these forces when the building is built.  The middle of the wall could bow a little, but the loft floor will hold the walls together somewhat and that the deflection would be well within the limits of code (wood can bend a little).

I will post back once I get word from the designer.

redside

I spoke with both Mitek engineers and the local truss plant regarding the drawing.  The truss drawing has two bearing points, one of which is pinned and the other on a roller bearing.  The purpose is to show the max deflection and horizontal reactions.  When the building is constructed, the truss will be connected to the top plates and the deflection at each of the two top plates will be about half of what is shown on the truss drawing.  For example, if the truss drawing shows a 1/2 inch deflection, then it will be about 1/4 inch at the top of each wall.  The same goes for the horizontal reaction. 

The code is about 1.25" horizontal deflection for these trusses (not that I would ever want to build with one that deflects that much), but it sounds worse than it really is.  The truss drawing is showing the worse case scenario per the load case the software runs, so the reality is that the truss will likely never experience the max reaction.  Also, the truss drawing does not account for sheathing the top and bottom chords which would limit the deflection even more.

So, in the end, it is up to the builder if he can live with the deflection.  A wall will easily bend 1/4 inch so I am fine with it.....others might not be. 

It is a judgement call. 



Don_P

Good deal, a quick conversation with the designer or the truss plant engineer has cleared up a lot of my questions before.