Rafter to Wall Connection

[sema79]

Hi everyone,
My company is investigating the failure of some roof framing after the recent snowstorms. The rafter does not connect directly into the wall plate. Instead the joists bear on the wall, continuous 2x4 runs across the top of the joists and the rafters bear on the continuous 2x4. There is a rim joist that also takes some of the load from the rafters. As you would expect the rafter forces are pushing the rim joist away from the joist. I'm from Florida and have never seen framing like this in 5 years of construction and 10 years of engineering. When I saw this my response was why would someone do something like that and think it was ok. I have attached a sketch of the condition. Please take a look at the sketch and let me know what you think. Is this detail something that is or was acceptable in the north?

Thanks in advance!

 

[mike20793]

Is there a ridge beam supporting the rafters on the other end? My guess is there is, hence no rafter tie. Excess deflection may have caused thrust leading the rafter to push the walls out since there is no tie holding them together. I use collar ties to help take the small thrust caused by unequal roof loading or excessive deflection of the ridge beam.

 

[manstrom]

I don't think anyone has ever properly designed a rafter roof. If it doesn't have a ridge beam, it's probably built wrong or engineered wrong.

When I am working on a problem, I never think about beauty but when I have finished, if the solution is not beautiful, I know it is wrong.

-R. Buckminster Fuller

 

[msquared48]

I have seen this sand scenario where there is an attic floor and intermediate 4 foot interior pony bearing walls from the rafters to the floor joists. Usually has a high collar tie too for head clearance.

Yes, the rafters would tend to kick out if not anchored sufficiently to the plate over the joists.

Mike McCann, PE, SE (WA)

 

[XR250]

Just saw this on a house today. It is pretty common around here and have only seen a few instances with the thrust kicking out the band. I agree it is not a great detail. It works better if the attic floor is completely sheathed and a 2x6 is used for the bearing so it can be fastened better.

@Manstrom - Seriously?
A standard gable roof where the rafters lap the ceiling joists and the ceiling joists lap each other and are fastened properly is significantly stronger than a roof with a structural ridge.

 

[EngineeringEric]

The sounds like a false plate used in roof framing. It is my understanding from historic building that is was the default for Mid-Atlantic building. It went through numerous changes and then was eliminated and merged into what we have today. Haven't seen it used on anything built to any 'code'.... unless you are designing a 18th century building for today's use, for historical reasons.

It can work, just not as good. would never use it today on anything i design.

 

[DaveAtkins]

The detail would be OK if there is a ridge beam which can support load by carrying the load to supporting walls or posts across the middle of the attic. My guess is this is not the case, and the connection between the rafter and the 2 X 4 plate was not designed for the outward thrust due to snow load.

DaveAtkins

 

[Tmoose]

Just to confirm, did the roof collapse when the rafter tails pushed outward, taking the 2X4 and rim joist along with them?
The 100 lbs or so rated lateral strength of each 16d nail in a more traditional joint, or securing the (missing) collar ties gets used up quick.
Depending on how things actually failed all that is missing are some Simpson connectors.


That detail is sure a lot easier to insulate nicely and ventilate than when rafters and joists both rest on the top plate.

 

[SteelPE]

This post kind of hits home for me. I recently has a long discussion with a junior engineer who wanted to use this detail without a ridge beam. I spend too long trying to explain to him that the bottom connection needed to be designed to accommodate the thrust from the rafter. He argued that the load could be taken through the toe nails. I don't remember what the conclusion was.... But it involved some sort of Simpson strap or anchor at the base and be hitting the pub early.

Kind of stinks when your arguing with someone who's father writes your paycheck.

 

[KootK]

The detail could be made to work as there's definitely a load path there. The rafter thrust just needs to be transferred to the ceiling joists in a way that works by the numbers. As shown, there's not a lot of meat to work with.

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]

Here's another way I have done it in the past…(see attached)
I can't understand so many Engineer's obsession with ridge beams. I always try to find ways not to use them if it is not too complicated. If you use a ridge, you have the costs associated with the materials and chasing the loads all thru the house. Maybe even have to add a crane to the mix and you end with a roof that is not as stiff, IMHO, as a series of well constructed triangles.

 

[mike20793]

I agree with XR250, lap the ceiling joist with the roof rafter so the joists act as the rafter tie. I usually try to do this first. I also add a collar tie for unbalanced loading.

 

[EngineeringEric]

I attached a photo of a similar condition i saw for a house built in the 70's... We added clips and some of the long Simpson hold down screws.

 

[KootK]

Are the rafters usually aligned with the ceiling joistS? I like that a bit better.

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.

 

[sema79]

Thanks everyone for all the response!

This part of the roof DID NOT have a structural ridge beam. It only had a 2x12 ridge board that was used for rafter attachment.

Tmoose, you hit the nail on the head(no pun intended). The outward thrust forced a failure at the joint causing a roof collapse. There were collar ties every third rafter. These were installed about 1/3 down from the ridge. In some locations there were shear failures of the bottom of the rafters at the birdsmouth cuts.

XR250, Nice detail. That's the way this should have been constructed in my opinion.

I wasn't sure if this was a one of a kind structure to be built this way or if it was reasonably common for the area.

 

[msquared48]

In essence, since the 2X12's span between the collar ties, they are your ridge beams. It just consists of five or six beams that have shorter spans.

This type of construction is very common in older buildings, particularly residences.

Mike McCann, PE, SE (WA)

 

[manstrom]

This type of construction may even meet the prescriptive code. The IRC has some roof details that don't add up when it comes to statics. Specifically, the collar tie connection and spacing.

I'm not surprised that these are seeing failures.

When I am working on a problem, I never think about beauty but when I have finished, if the solution is not beautiful, I know it is wrong.

-R. Buckminster Fuller

 

[RFreund]

I don't think anyone has ever properly designed a rafter roof. If it doesn't have a ridge beam, it's probably built wrong or engineered wrong.

Can you elaborate on this?

EIT

http://www.HowToEngineer.com

 

[wannabeSE]

Sema79,
Is this the house that was in the news because the roof slid off.

http://www.concordmonitor.com/home/...ial-rafter-joint-likely-cause-of-sliding-roof

 

[FSS]

So, was the roof overloaded and you were able to identify the failure mechanism? Before we get too harsh about the detail (which is OK if detailed and installed correctly) what kind of loading did we end up with? With enough snow eventually every roof can collapse.