Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations waross on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Roof Truss vs. Braced Ridge Beam vs. Bolted Collar Ties

Status
Not open for further replies.

the Paper Owl

Structural
Sep 22, 2021
22
Good afternoon everyone, I was hoping for a discussion on the expertise / applicability of choosing a roof system based on structural serviceability. It is my understanding that in situations like the one I am about to describe, the savings in labor / current state of the supply of materials may make a roof truss become the most applicable option.

The structure in question is a simple garage (attached to a 2,000+- single family home) with 24'x24' overall dimensions. For simplicity sake, we will consider the wall that the garage doors are located on as the x-axis, and the direction of the roof ridge as the y-axis.

The input I am hoping for (assuming I am not using a roof truss system for the garage and that the enclosed attic will be for access only, not storage) is if I am able to design a shear connection between the rafters and a 1/3 height placed collar tie as well as a shear connection between the rafters and the garage ceiling joists as a means to satisfy the thrust the roof would want to place on my exterior walls. With this logic, could the exterior walls be sized to mainly support 1/2 of the vertical roof load, and eliminate the need to provide roof braces along the ridge every +- 6 feet. I ask this because with a strong connection between rafters / collar ties & rafters / ceiling joists am I not essentially recreating a framed truss ?

In my experience I have seen (when roof trusses are not specified) vertical roof braces provided along the ridge, bearing on either a center steel beam (along the y-axis) or roof brace point loads bearing on LVLs (in the x-axis) with a center LVL (y-axis) that is hangered into the x-axis LVLs.

In summary I would like to use the connections at the collar ties and ceiling joists to eliminate the need for roof braces below the ridge, is this thinking unjustified?
 
Replies continue below

Recommended for you

The collar ties would only help keep the ridge closed in an uplift scenario. But the type of framing you're describing is a rafter and ridge board system. As long as you have ceiling joists with appropriate connections to deal with the thrust, I don't see why you couldn't rafter frame the roof.

However, 24 foot long ceiling joists may be a bit hard to get to calc out. So you end up needing something to support the ceiling joists anyway.
 
Thank you jayrod12, I appreciate the response. I will read up on the applicability of the rafter and ridge board system. I too see your point with the 24 span on the ceiling joists. This greatly helped.
 
I agree with jayrod. You're describing a traditionally framed rafter roof. The ceiling joists take the thrust, the collar tie fixes the rafter connection at the ridge, opposing rafters support each other, and the ridge board doesn't do much but brace the rafter to rafter connection and make it easier to build. This is the basic construction type described in the North Carolina Residential Code.

If you don't want trusses or a column, you'll need a large ceiling beam to support the ceiling joists. Since you'd be using them to resist thrust, each joist would need to be either lapped over the beam and appropriately spliced or, if a flush beam, run a strap from joist to joist to tie them together.
 
Thank you phamENG, I appreciate the insight.

My hesitation has been that I am seeing a fair amount of plan sets come across my desk where the typical roof system is comprised of a horizontal purlin brace (placed at rafter mid-span) with vertical roof brace placed roughly every 6 feet. I understand that this type of system would allow for smaller rafters (reducing the deflection through the reduction of the clear span that is loaded) but typically results in roof brace point loads that in return create built up columns / stud packs (allowing more thermal bridging sections within the wall system if an exterior wall).

I am of the opinion that if I do not have any ceiling treatments / not worried about the openness of the attic above the space, I should instead look into a framed rafter roof system versus a smaller rafter system but with more purlin braces (hog ties, vertical roof braces).

These assumptions do not insist that roof brace point loads created where valleys / hips intersect ridges are not needed as I understand their applicability regardless of the chosen roof system.
 
Honestly, if they have truss plants where you are, that is the easiest and fastest way. Material price alone is higher, but once you account for labour and schedule it's not even a contest. A good framing crew can put the roof, including sheathing, on a truss framed garage in about 3 hours. A raftered system takes significantly longer.
 
I agree with jayrod and pharmeng

Fox - I would add that from a structural perspective, in a braced rafter system, the optimal horizontal brace (whether you call them ceiling joists, rafter ties, or collar ties) location is at rafter bearing, not rafter mid-span. I would guess the plan sets you see come across your desk are optimized for non-structural aspects.

To answer your original questions, yes, these are not challenging connections to make. You might even be able to justify nails alone. Yes, you can size exterior walls to mainly support 1/2 of the vertical roof load, and eliminate the need to provide roof braces along the ridge every +- 6 feet. Yes, with a strong connection between rafters / collar ties & rafters / ceiling joists, you are recreating some aspects of a framed truss.
 
Mike Mike said:
he optimal horizontal brace (whether you call them ceiling joists, rafter ties, or collar ties) location is at rafter bearing

Have to disagree with you on the last one. A collar tie and a rafter tie are two very different things and are very plainly defined in the code. Collar tie goes in the upper 1/3 and locks in the rafters at the ridge for unbalanced loading, rafter tie goes in the bottom 1/3 and resists spread. I do agree that the most effective place for a rafter tie is at rafter bearing.

I'm 99.9% certain that he's not talking about either of these, though. By "horizontal purlin brace" I'm thinking he just means purlin. So there's a purlin that runs parallel to the ridge at rafter mid span that is supported by posts at 6' on center down to a wall, beam, or some other transfer element. Hence the "smaller rafters" comment. Putting a purlin at the wall where the rafter sits would do...nothing at all.

Also - why does nearly everyone add an 'r' in my handle when they retype it?
 
I guess I just think of pharmacist engineer when I see you I don't know why [glasses]
I think I might see what you mean, like there's two lines of purlins parallel to the x axis with posts between the two parking stalls?
 
Mike - good one. Should have gone into pharmacy. Pay's a lot better.

Here's a clip from the IRC:
purlin_mbr2mx.png
 
phamENG thank you for the code reference and all for the discussion.

I still stand with the opinion that a better design should entail rafters that are sized to carry the full span of the roof section (within reason) in lieu of the possible ability to use 2x6 rafters with purlins and purlin braces.

The added need for the contractor to provide purlin braces with the proper bearing (fingers crossed) as well as the now needed built up columns /stud-packs with associated squash blocks below each point load feels to me like more precision needs to be taken in lieu of running 2x10 / 2x12's at the desired spacing. I've also seen complexities with purlin brace designs when the home owner may go to renovate the attic and now needs to redistribute a slew of roof brace point loads that proof to be in the way (needing to insert built up columns in wall sections where the newly distributed point loads will bear).
 
Paper Fox Collective said:
a better design should entail rafters that are sized to carry the full span of the roof section

Couldn't agree more.
 
I used to be a carpenter. I stick-built conventional roof framing, did site-built trusses and used manufactured trusses. Unless you have a good crew, do the manufactured trusses.
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor