Residential tied roof

[MyCupboard]

I have been in residential structural engineering for about 8 years. I know there are some things that are a “big deal” and other things that can SOMETIMES be approved simply because of redundancy and oversimplification of the system.

Tied roof concepts tend to be a good topic with varying opinions. This specific question is related to when roof rafters (sloped) and ceiling joists (flat) meet at a flush bottom beam.

In typical wall details for the house, the rafter laps the ceiling joist and is fastened accordingly to resist the thrust. So where they come together at a flush beam I just thought they would align the beam appropriately so that they overlapped, get fastened, and both hang off the beam.

During a frame walk today, I came across this little detail. I’m curious to hear everyone’s thoughts. Mine range from (there is still a bunch of nails scattered into the lapped portion), to (should I have called out separate hangers because of the sloped condition and flat condition?), or (couldn’t they have located the beam so that there was more of a “seat” at the bottom of the roof rafter that could then hang into the beam with the double hanger?

I am by no means an all knowing expert, but considering this builder is extremely confident and proving to be not at all competent, I’d like to get some wisdom before I address this line item.

For clarification, they scabbed an angled member to the side of the ceiling joist to create a flush bearing condition at the hanger. That’s my guess at least.

 

[ANE91]

This thread will be fun.

Looks like your LVLs will be doing the heavy lifting in terms of resisting rafter thrust via weak-axis bending.

The angled/lapped wood doesn’t do anything for me. Creative, though!

Do those hangers calc out? That’s my main concern.

 

[MyCupboard]

This thread will be fun.

Looks like your LVLs will be doing the heavy lifting in terms of resisting rafter thrust via weak-axis bending.

The angled/lapped wood doesn’t do anything for me. Creative, though!

Do those hangers calc out? That’s my main concern.

Oh my goodness. Duh! I’m embarrassed I didn’t think of that. It was a late night with lots of errors to review clearly.
Thank you for your reply. As far as method for hanging these members, does this funky angled scab seem ok? I really thought in a case like this they would shift the lvl back and have more of the rafter bear on the hanger.

 

[phamENG]

The weak axis bending isn't a good load path to rely on. It might work, but you'd have to check it. And also check the deflections. These are the sorts of things that can result in cracked drywall and angry home owners. And don't forget - these systems are two sided. You need something to take it at the other side AND you need to resolve the horizontal reaction of the beam at each end - one of which will be in the middle of your gable, probably overloading the connection for the single rafter at the end.

Scabbing that on is fine. You should have detailed how to fill out the hanger (Simpson catalogs say shim by designer). How would you have done it differently?

How many nails are there between your rafter and ceiling joist? Enough to resist the thrust? If not, I'd be strapping the rafter to that scabbed filler to get the load back down into the tension tie.

other things that can SOMETIMES be approved simply because of redundancy and oversimplification of the system.

Not so much. A violation of the building code is just that. If you can't identify and verify the capacity of a load path, then shrugging and saying "it should work" is not engineering - it's being irresponsible.

 

[XR250]

Common detail for me. As long as the rafter is face nailed to the ceiling joist to resist the thrust - which it appears to be.

 

[XR250]

Here is a clever one I saw yesterday in a similar vein. Using a double hanger to support an angled, single I-joist. Fortunately, the joists were only 8 ft. long. But it actually may work fine with heavier loads.. they added subfloor adhesive for extra capacity I suppose.

 

[MyCupboard]

The weak axis bending isn't a good load path to rely on. It might work, but you'd have to check it. And also check the deflections. These are the sorts of things that can result in cracked drywall and angry home owners. And don't forget - these systems are two sided. You need something to take it at the other side AND you need to resolve the horizontal reaction of the beam at each end - one of which will be in the middle of your gable, probably overloading the connection for the single rafter at the end.

Scabbing that on is fine. You should have detailed how to fill out the hanger (Simpson catalogs say shim by designer). How would you have done it differently?

How many nails are there between your rafter and ceiling joist? Enough to resist the thrust? If not, I'd be strapping the rafter to that scabbed filler to get the load back down into the tension tie.


Not so much. A violation of the building code is just that. If you can't identify and verify the capacity of a load path, then shrugging and saying "it should work" is not engineering - it's being irresponsible.

Yes, we specialize is high end custom homes (the real kind $20MM+, not the high end custom homes that everyone and their brother claim to build), so dealing with this low budget builder was not ideal but trying to meet their needs within their budget.

We do check weak axis for these scenarios (when the rafter tie is not present) and deflection and all that good stuff.

Performance is our priority in most houses (instead of code minimum or prescriptive designs). Again, when presented to the builder they opted for a lower end design. We of course needed to be good with it, but performance was on the chopping block and the homeowner signed off on it

 

[MyCupboard]

The weak axis bending isn't a good load path to rely on. It might work, but you'd have to check it. And also check the deflections. These are the sorts of things that can result in cracked drywall and angry home owners. And don't forget - these systems are two sided. You need something to take it at the other side AND you need to resolve the horizontal reaction of the beam at each end - one of which will be in the middle of your gable, probably overloading the connection for the single rafter at the end.

Scabbing that on is fine. You should have detailed how to fill out the hanger (Simpson catalogs say shim by designer). How would you have done it differently?

How many nails are there between your rafter and ceiling joist? Enough to resist the thrust? If not, I'd be strapping the rafter to that scabbed filler to get the load back down into the tension tie.


Not so much. A violation of the building code is just that. If you can't identify and verify the capacity of a load path, then shrugging and saying "it should work" is not engineering - it's being irresponsible.

maybe you’re not in residential design. Or maybe you don’t deal with inspections ever. No problem, my short stint in commercial engineering had the same outlook.
The residential building code leaves a lot to the imagination. There are a plethora of things that don’t work but the code deems is okay (don’t get me started on king studs and jack studs).
To be clear for everyone else, by no means am I saying that I shrug anything off. When things work for a less than obvious reason, I know we need to find the reason that it works. And if you hold up a job for doing something that is industry standard but not explicitly written in the code, then I hope you enjoy your short stay in engineering or are independently wealthy.

 

[Eng16080]

I would check:

1. The load path resisting the vertical rafter load by looking at the nailed connection between the lapped portion of the ceiling joist and rafter and then the capacity of the joist hanger to the beam.
2. The load path resisting the rafter thrust based on the nailing between the lapped area. If that's adequate to resist thrust I wouldn't worry about resisting the thrust through negative beam bending (as others are suggesting).

Assuming those conditions are met, I don't think I'd be overly concerned with this solution. I also wouldn't blame the builder on this one, considering it sounds like you didn't detail the connection.

If 1 and 2 don't work, I'd look into adding more fasteners (nails).

 

[MyCupboard]

Common detail for me. As long as the rafter is face nailed to the ceiling joist to resist the thrust - which it appears to be.

Ok cool. This is about all I was looking for haha. I appreciate the lack of condescending remarks. I may not have 40+ years of experience but I’m not going to lose my lively hood to save any builder’s ego

 

[KootK]

We do check weak axis for these scenarios (when the rafter tie is not present) and deflection and all that good stuff.

I don't like the weak axis beam path here. Among other potential deficiencies, the ends of of the beam don't have proper lateral support at the bearings. At best, this path is about using the diaphragm on the other side of the LVL's to resist thrust and just using the LVL as a load distribution element. Sure, this path does exist to some degree. But it's unlikely that you will have explicitly designed this path.

On balance, I agree with @XR250. If you've got enough nails in the lap to deal with thrust, I'm pretty happy with the existing condition.

I like the scabbed wedge for two reasons:

1) Fills out the hanger a bit. I doubt hangers are tested for being half loaded. Minor point. And the scab doesn't wholly solve it.

2) If the rafter delivers its vertical load through the scab in bearing, then that's extra load that the nails in the scabbed joint don't have to deliver. This assumes that the scab is tight to the rafter which is not the case here. Still, you probably get some real bearing transfer at the very end. And it wouldn't take much nail slip to generate more.

maybe you’re not in residential design. Or maybe you don’t deal with inspections ever.

You must be new here. @phamENG is regarded as an expert on residential construction in our community.

 

[phamENG]

@MyCupboard - I reread my post, and I see where it probably came off as confrontational. Not my intention - I probably shouldn't be allowed on here until I've had my 3rd cup of coffee in the morning. Sorry about that.

KootK is too kind - I'm far from an expert, but I do almost exclusively residential nowadays. I don't know you at all, so when you say something like "other things that can SOMETIMES be approved simply because of redundancy and oversimplification of the system" I don't know if you're using a hand-wavy cop-out or if it's because you have a deep understanding of the load paths in a light frame wood structure and have checked those load paths before and know what does and doesn't work. Philosophically, I'm opposed to giving voice to that either way. To me, it cheapens what we do in the eyes of the people who are constantly giving us grief for doing our job. Also, especially in the really high end structures, those redundancies have a tendency to go away in the interest of architectural grandeur....or something.

I don't care if the building I'm designing is residential or commercial. I'm going to give it the same rigor either way. In my area, "the big one" will be a hurricane. Where are people most likely to be when a hurricane hits or we get a freak snowstorm that exceeds roof live loads? At home. So when people tell me "it's just a house," I usually reject it out of hand. Plenty of contractors hate me for it. But there's a handful of contractors that do good work and appreciate what I'm trying to do, and I've tapped into a community of architects that are of a similar mind. I'm not going to rich, but I have no shortage of work doing it this way.

 

[MyCupboard]

I would check:

1. The load path resisting the vertical rafter load by looking at the nailed connection between the lapped portion of the ceiling joist and rafter and then the capacity of the joist hanger to the beam.
2. The load path resisting the rafter thrust based on the nailing between the lapped area. If that's adequate to resist thrust I wouldn't worry about resisting the thrust through negative beam bending (as others are suggesting).

Assuming those conditions are met, I don't think I'd be overly concerned with this solution. I also wouldn't blame the builder on this one, considering it sounds like you didn't detail the connection.

If 1 and 2 don't work, I'd look into adding more fasteners (nails).

Wait, maybe I’m in the wrong. Does everyone on here detail every single condition in a house, no matter how common or simple? This is an extremely common and simple condition. I call out “lap rr with cj and fasten with xx”. I would never think I would need to show someone how to lap a rr with a cj. Now add the hanger condition, you lap your members, then fasten it with the hanger. In every other house I’ve seen (maybe not thousands of houses but hundreds at least), they cut the rafter as if it’s sitting on a wall (but instead the seat of the hanger) and lap it properly with the cj.

Apparently this is not the common method? I will make sure I add this to my standard details list.

 

[phamENG]

Here is a clever one I saw yesterday in a similar vein.

Interesting. Did they do anything to fill the 'wedges' on either side and fasten it?

The glue is pretty normal to me - hangers can lead to squeaks, so most contractors put a dab of adhesive in there to limit that.

 

[phamENG]

Does everyone on here detail every single condition in a house,

Not all of them, but I would say the condition in question isn't very "normal." It seems obvious to us, but for most contractors the normal thing to do with a beam is to put it in line with the wall. So which normal governs? The beam usually does in my market, so I have a typical detail that I can drop into the set if this sort of condition exists.

 

[MyCupboard]

@MyCupboard - I reread my post, and I see where it probably came off as confrontational. Not my intention - I probably shouldn't be allowed on here until I've had my 3rd cup of coffee in the morning. Sorry about that.

KootK is too kind - I'm far from an expert, but I do almost exclusively residential nowadays. I don't know you at all, so when you say something like "other things that can SOMETIMES be approved simply because of redundancy and oversimplification of the system" I don't know if you're using a hand-wavy cop-out or if it's because you have a deep understanding of the load paths in a light frame wood structure and have checked those load paths before and know what does and doesn't work. Philosophically, I'm opposed to giving voice to that either way. To me, it cheapens what we do in the eyes of the people who are constantly giving us grief for doing our job. Also, especially in the really high end structures, those redundancies have a tendency to go away in the interest of architectural grandeur....or something.

I don't care if the building I'm designing is residential or commercial. I'm going to give it the same rigor either way. In my area, "the big one" will be a hurricane. Where are people most likely to be when a hurricane hits or we get a freak snowstorm that exceeds roof live loads? At home. So when people tell me "it's just a house," I usually reject it out of hand. Plenty of contractors hate me for it. But there's a handful of contractors that do good work and appreciate what I'm trying to do, and I've tapped into a community of architects that are of a similar mind. I'm not going to rich, but I have no shortage of work doing it this way.

i appreciate you being the bigger man here...i too need to take your advice. I see you post on here all the time (makes me question how you get any work done), so i should know your intentions by now. part of my anger is residual from the builder essentially telling me he isn't going to build the house the way i design and demanding i change my plans. Not fair to take that out on anyone on here. My apologies to the group.

I see what you're saying now that i've cooled down 3 degrees. Thanks for your input and i will take it with humility.

 

[Eng16080]

Does everyone on here detail every single condition in a house, no matter how common or simple?

My approach is more on the detailed side (compared to other plans I've seen) but it's of course not reasonable to detail everything. I typically would detail this connection though. It's generally better not to assume what the builder might or might not do.

This is an extremely common and simple condition. I call out “lap rr with cj and fasten with xx”. I would never think I would need to show someone how to lap a rr with a cj.

And there you are! The builder met that requirement!

Now add the hanger condition, you lap your members, then fasten it with the hanger. In every other house I’ve seen (maybe not thousands of houses but hundreds at least), they cut the rafter as if it’s sitting on a wall (but instead the seat of the hanger) and lap it properly with the cj.

Based on the geometry of the rafter and ceiling from the picture, cutting a flush bearing seat at the rafter would require notching the ceiling joist. I'd prefer what the builder did here rather than notching the ceiling joist.

 

[XR250]

Interesting. Did they do anything to fill the 'wedges' on either side and fasten it?

The glue is pretty normal to me - hangers can lead to squeaks, so most contractors put a dab of adhesive in there to limit that.

Nope. I imagine this would test out fine for realistic loads though due to some of the joist bearing on the "strong" parts of the hanger.

 

[XR250]

My approach is more on the detailed side (compared to other plans I've seen) but it's of course not reasonable to detail everything. I typically would detail this connection though. It's generally better not to assume what the builder might or might not do.

I would detail it as well

 

[MyCupboard]

I would detail it as well

We have a consensus. I will detail this. Now, more so to my curiosity, i would think that i would hold the beam back slightly (left in the photo) so that i can create a "bearing condition" for the rafter on the hanger. I think someone else commented on how this would require a notch in the CJ, but the way i'm thinking about it it would not - so maybe i'm not explaining this portion well. i'll attach a photo of what i'm thinking.