2x8 roof rafters, adding collar ties 2

[DCStructures]

Any tips on strengthening 2x8 roof rafters? Collar ties currently exist on every 4th rafter. I'm thinking that adding collar ties at every rafter could easily increase capacity. Any tips? Suggestions? Warnings? Cautions? Thanks in advance.

 

[KootK]

It depends on your arrangement and the nature of the failure in the existing joists. Sistering new material to the existing joists is a common strategy where the collar tied jousts are not failing and you're just needing help at the joists between. Are the members that are currently failing:

1) those with collar ties?
2) those without collar ties?
3) both?

Tips for collar tying:

1) the rafter tails below the tie work very hard in bending. Position your collar ties low enough to deal with this.

2) the collar tie to rafter connection places a lot of tension perpendicular to grain stress on the rafter. Watch fastener end distances and try to engage most of the rafter cross section with the fasteners.



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.

 

[jdgengineer]

I think you need to analyze the roof system so you can have an understanding of the impact.

I'm assuming you are confusing collar ties with rafter tie. Collar ties occur at the top 1/3 of the rafter and are in place to resist wind uplift to keep rafter from seperarting from the ridge.

See attached for good explanation from fine homebuilding.

Rafter ties allow you to form a field built truss and remove the necessity for a structural support at the ridge. If your rafters are over spanned, adding additional rafter ties won't help you increase the capacity of the rafters but it will help reduce the demand on the rafter tie connection.

If you are referring to adding additional collar ties at the top, fhey won't really help you too much. They will help with some ability to resist unbalanced snow loads and wind uplift. I think the requirement based on conventional construction is to have collar ties @ 48" oc max so adding in additional infill ties will meet that requirement. Aside from large wind events, I don't think you will see noticeable roof framing improvement as a result.

 

[jdgengineer]

If your rafter ties are above the wall plate line the connection does create some serious bending moments as mentioned by KootK. Typically, it's preferred to have the ceiling joist serve as the rafter tie and have them both sit on the plate. This eliminates the additional moment on the rafter.

 

[Ron]

jdengineer....A collar tie is to reduce the amount of lateral thrust at the rafter tail to wall connection so that the wall is not pushed out. Deflection in the ridge beam or the absence of a ridge beam increases the lateral thrust at the top of the wall.

Agree with KootK's comments and warnings!

 

[KootK]

Great article jd, thanks. I wasn't really clear on the terminology difference myself.

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.

 

[jdgengineer]

Ron -- What you are describing is a "rafter tie". And yes, I agree they are necessary when you do not have a structural ridge support (I believe I mentioned this in my response). The collar tie is at the peak (think shirt collar).

It's a common confusion, but the two elements are different and serve different purposes. The article I posted I believe explains it we'll.

 

[oldestguy]

If you think you really need stronger rafters, one way wold be to add a steel strap along the under side, properly secured with screws mostly near the ends.

 

[XR250]

See attached for good explanation from fine homebuilding.

I actually wrote most of that article for FHB. Nice to see it was helpful.

 

[dik]

I've done numerous engineering reports of roof failures and all collar tie failures (upper third) have been compression failures, pushing the ties through the roof sheathing.

Dik

 

[jdgengineer]

XR250 that's great. It's a great article and one I use for reference with homeowners and younger engineers. In full disclosure we pretty much never conventionally frame roofs and generally use structural ridges. But that's probably just a regional thing.

Dik that makes some sense to me. If rafter ties are properly in place the collar ties I believe are actually in compression for standard gravity loading.

 

[KootK]

I actually wrote most of that article for FHB. Nice to see it was helpful.

Always kinda had you figured for a leggy brunette.

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.

 

[XR250]

In full disclosure we pretty much never conventionally frame roofs and generally use structural ridges. But that's probably just a regional thing.

Interesting. Most roofs here are conventionally framed unless vaulted or have big dormers or other crazy stuff.

Always kinda had you figured for a leggy brunette.

Hey, I resemble that remark!

 

[jdgengineer]

Yes, thankfully builders around here are used tonl framing everything around here with structural beams. Most other engineers in our area show it the same way so it isn't a fight.

Most homes built earlier than 1980 in our area are conventionally framed. To this day I have never seen one of these roofs without a "flaw". Typically, at least one (usually more) is present:

1) ceiling joists perpendocular to rafters in some area without additional rafter ties
2) rafter tie to ceiling joist occurs only every 3rd rafter and connection only has a few nails
3) one side of the ridge is partially vaulted but no consideration to the fact that the rafter tie no longer runs plate to plate
4) purlins are overspanned and only have braces at perpendicular partition walls rather than closer spacing required.
5) ceiling joists are not properly connected at splice location
6) no structural support at hip / ridge intersection.
7) collar ties are spaced too far apart.
8) at prulin brace locations no consideration given to impact on floor loading below for concentrated load.

Because these are so common I'm nevervous about contractors properly installing conventional framing. Less opportunities for error with structural ridge. As we don't have snow loads these roof defects tend to perform "OK". Areas with higher snow loads I assume can't get away with these errors and may have them less frequently (I hope at least)[pre][/pre]

 

[jdgengineer]

I have also never really rationalized how conventionally framing works at hips. I understand there is folded plate action and other mechanisms, but on one side of the hip it seems the jack rafters do not have rafter ties (I never have seem them in place) Sometimes the hips are quite large and I'm not really sure how that works.

 

[jdgengineer]

XR250 is the flitch plate article also yours? We use this technique frequently.

 

[XR250]

Yes, thankfully builders around here are used tonl framing everything around here with structural beams. Most other engineers in our area show it the same way so it isn't a fight.

Most homes built earlier than 1980 in our area are conventionally framed. To this day I have never seen one of these roofs without a "flaw". Typically, at least one (usually more) is present:

1) ceiling joists perpendocular to rafters in some area without additional rafter ties
2) rafter tie to ceiling joist occurs only every 3rd rafter and connection only has a few nails
3) one side of the ridge is partially vaulted but no consideration to the fact that the rafter tie no longer runs plate to plate
4) purlins are overspanned and only have braces at perpendicular partition walls rather than closer spacing required.
5) ceiling joists are not properly connected at splice location
6) no structural support at hip / ridge intersection.
7) collar ties are spaced too far apart.
8) at prulin brace locations no consideration given to impact on floor loading below for concentrated load.

Because these are so common I'm nevervous about contractors properly installing conventional framing. Less opportunities for error with structural ridge. As we don't have snow loads these roof defects tend to perform "OK". Areas with higher snow loads I assume can't get away with these errors and may have them less frequently (I hope at least)

I hear ya. I see that all the time on older houses - less so on newer ones. In my experience, I rarely see a problem except where the roof pitch is really shallow or the entire roof is framed with the ceiling joists perp. or the 1/2 vaulted situation or long shed dormer.

I have also never really rationalized how conventionally framing works at hips. I understand there is folded plate action and other mechanisms, but on one side of the hip it seems the jack rafters do not have rafter ties (I never have seem them in place) Sometimes the hips are quite large and I'm not really sure how that works.

Folder plate, diaphragm, sheetrock, helium - who knows. Same as above, they seem to work in our area unless the roof is a shallow pitch.
FWIW, I always support the hip-ridge-valley joint unless the roof is small. Homeowners love it when they want to finish off their attic and there are posts everywhere.

 

[XR250]

XR250 is the flitch plate article also yours? We use this technique frequently.

All I can tell you is that the builder is a client of mine.

 

[jdgengineer]

Oldestguy I'm curious if you have successfully retrofitted a roof condition with your strap method you are describing. I'm assuming a failing rafter is likely one that is deflecting significantly. I don't think a strap would add appreciable stiffness to the rafter assembly. I suppose if you were able to pretension it before install it would help with deflection (similar to what we do sometimes with bowstring trusses)

I'd also be a little nervous of the slip between the strap and the rafter. Although maybe not an issue if the deflection is large enough in the rafter.

 

[HouseBoy]

Great reference articles XR
Now, for the OP - can you tell us what is the nature of the problem?
I'm guessing that since you want to "strengthen" the 2x8 rafter, it might be because the rafter is exhibiting bowing.
Collar ties (in the upper third) go into compression under gravity loading (as dik and jdg mention).
Collar ties will help to brace the rafters but I want to note that the addition of collar ties will effectively increase the thrust at the bottom of the rafters and the connection from the rafters to the rafter ties is sometimes overlooked.