Wood Roof - Thrust 9

[Buleeek]

Hi everyone,

I'm analyzing the TFEC Bulletin No. 2018-12 "Behavior of statically determinate and indeterminate rafters". If I use a collar tie (or strut) system (with no horizontal reaction at the bottom of the rafter)can I assume there will actually be NO horizontal force at the wall top plate? I know that collar ties should not be used to carry the thrust, but struts yes (located higher than top plate level in this case).

I am not sure what situation occurs in a real life. I can't assume that the rafter has no thrust at the bottom, since it is resting on the wall plate and is always notched/nailed/connected.

What do you usually do in such situation? That assumption determines a wall design and how much thrust it will carry.

Thanks,

 

[1503-44]

I think you would use rafter ties, similar to collar ties, but placed lower and nearer to the wall. Like this...

https://www.mathscinotes.com/2010/11/the-mathematics-of-rafter-and-collar-ties/

“What I told you was true ... from a certain point of view.” - Obi-Wan Kenobi, "Return of the Jedi"

 

[JoshPlumSE]

Generally, if you want the tie / strut resist the thrust / displacement at the top of the wall, then it should be located in the lower 1/3 of the joist. It doesn't have to be right at the wall location. But, it should be reasonably close.

I tend to use collar ties / rafter ties interchangeably. Though, many refer to collar ties as being in the upper 1/3 of the rafter. If they are that high up, I don't personally consider them as resisting outward thrust on the wall.

 

[KootK]

If I use a collar tie (or strut) system (with no horizontal reaction at the bottom of the rafter)can I assume there will actually be NO horizontal force at the wall top plate.

Like most things, the load paths that get activated will respect the relative stiffnesses of those load paths which implies that:

1) A collar tie up near the ridge won't shield the walls from thrust at all.

2) A rafter tie at the ceiling level, basically a truss bottom chord, will shield the walls from thrust very well. The walls will only see as much imposed lateral movement / thrust as will be generated by the very stiff axial extension of the tension member.

3) Any tie placed above the ceiling level will impose some degree of thrust / lateral displacement on the wall top plates. If the tie and rafters are designed to work without thrust reactions, then you don't need to worry about providing competent thrust reactions in order to maintain equilibrium. You only need to concern yourself with whether or not the tops of the walls can spread along with the rafter tails without damaging things.

Questions for you:

a) Rafter span from bearing to ridge?

b) Roof pitch?

c) How high up are your ties as a percentage of the ridge to bearing height?

I my experience, with a pitch less that 6:12 and ties located in the lower 1/3, lateral wall spread is usually manageable.

 

[Buleeek]

KootK,

Please see attached. A horizontal length of the rafter is 14.5'. There is no ridge beam (just ridge board). The architect called for a collar tie, which actually is considered as a strut and is located in lower 1/3 of the height. This is a typical roof system I'm dealing with on a daily basis. I always keep the worst case scenario in mind, so the structure is often oversized.

I would like to finally find a definite answer: If there is no ridge beam and there is a strut (in lower 1/3, but not on a ceiling level) should I assume the horizontal reaction at the bottom of the rafter?

If anyone know any articles/literature explaining this situation clearly I would be very thankful.

 

[KootK]

should I assume the horizontal reaction at the bottom of the rafter?

Know that every possible permutation of this has some combination of:

1) A horizontal reaction and;

2) A demand for lateral, top of wall deformation.

The only question is how much of each will you have in a particular situation. With the tension tie designed correctly, the hope is that the amount of wall lateral deflection required is small enough that developing it doesn't damage other things and doesn't generate a large horizontal thrust.

I would like to finally find a definite answer:

Yeah, wouldn't we all.

For the arrangement that you've described, and with pitches 6:12 and less, it is my opinion that it will usually be okay to ignore the lateral thrusts at the bottom of your rafters. This needs to be considered, thoughtfully, on a case by case basis however.

I would recommend 2D FEM modelling a few of your common scenarios to get a feel for:

1) How much deflection you'll get at the top of the wall if you allow unrestrained lateral movement there.

2) How much horizontal thrust you'll get at the top of the wall if you allow no lateral movement there.

The truth will be somewhere between these extremes. Knowing what's going on with this stuff is really a gift that you can give yourself by doing the necessary, exploratory work.

 

[KootK]

This fine home building article is the closest thing to "definitive" that I know of: How it Works: Collar and Rafter Ties

 

[Buleeek]

Thank you. I will do some research on that matter.

One more question. When a roof slope is 3/12 or less you need a ridge beam. How do I attach the rafter to ridge if I want to have a beam 4/12+ ?
What are proper support examples for rafters at the top when a ridge works as a beam (except dropped beam) in general?

Thank again for all the help,

 

[BAretired]

You can't assume a horizontal reaction unless you provide supports capable of resisting it. The stud walls do not qualify, so under gravity load, the collar tie is a tension member, not a strut. Since it occurs above the top of wall, the rafters must be capable of taking the moment caused by the reactions. And the walls will spread laterally.

BA

 

[oldestguy]

Hows come you engineers keep dwelling on rafters and possible thrust, etc. and never mention the plywood sheathing on top. That plywood system is a truss of sorts parallel to the slope and it can carry all the thrust you keep dwelling on.

 

[KootK]

Hows come you engineers keep dwelling on rafters and possible thrust, etc. and never mention the plywood sheathing on top. That plywood system is a truss of sorts parallel to the slope and it can carry all the thrust you keep dwelling on.

It's the same reason that we've told you each of the many, many times in the past that you've brought this up: it is extraordinarily difficult to design, detail, and construct such a system properly such that it would get done right in the field and perform reliably. Additionally, nobody wants to design a roof that will fall down if an owner decides to install a couple of skylights in the future.

I can assure you with absolute confidence that we structural engineers know all about the diaphragm business. It's old news for us. We don't often avail ourselves of the benefits of diaphragm action when considering gravity loads because we know our crafts and feel that ignoring this effect is appropriate.

 

[oldestguy]

OK, it's been termed "belt and suspenders" to take care of unknowns. Explains why the "walls don't bulge out?" with no rafter ties, etc.

 

[BAretired]

But if you remove your belt, you need suspenders or your pants will fall down.

BA

 

[KootK]

Explains why the "walls don't bulge out?" with no rafter ties, etc.

1) Sometimes the walls do in fact bulge out.

2) Yes, I agree with you, in many cases the diaphragm does offer meaningful resistance.

3) What I don't agree with is your repeated implication that we structural engineers are failing to recognize a viable load resisting mechanism. We recognize it just fine; we simply happen to think that it's a bad idea to rely on it in most cases.

My own house has couple of broken truss bottom chords up in the attic. So far, so good. That doesn't mean that I'm going to start advocating bottom-chord-less trusses all over the place.

 

[oldestguy]

I agree with the reasoning. I'll another reason. Ya can't put this in a table and download it.

 

[r13]

Engineers are trained to argue something they don't understand about (Just a joke, no offense intended)

 

[r13]

Am I making mistake in resolving the force below?

 

[oldestguy]

Buleek: For your situation, I'd use plenty of nails for that plywood sheeting, especially for the bottom two rows. Then that "beam effect" of the bottom roof plywood rows can be a distinct advantage. End walls should be sheeting if possible to help take that thrust from the roof.. This of course in addition to beefing up connections for that horizontal tie. Ya can't compute this stuff, but we know it can help.

 

[JLNJ]

Retired13

The key to your free body diagram is what happens to the force at the top of the rafter. You have to have a reaction up there. If it can’t sink, then the forces are all vertical and there is no outward force at the heel.

And I’d indicate gravity as down rather than toward the plane of the roof.

 

[r13]

JLNJ,

No wind?

But, you are correct. The force along the rafter will be taking care by the tie as shear in the connection. Thus, no outward force will occur.