Residential Framing Questions 1

[markreme]

Everyone,

I have a friend who is building an addition to his home. The framing contractor did a few thing that the building inspector is questioning. As a structural engineer (I do not have too much experience with residential framing though),I am looking for some additional opinions.

Here are the problems:

1. The contractor staggered the roof rafters 8" o/c, and thus dd not provide collar ties. He also did not cut a few of the rafters at the proper angle to correctly meet the ridge board, and placed shims in the space. The building inspector is concerned over there being no collar ties, and the shims.

2. The inspector is also questioning the stagger in the rafters (says they should be a max of 6.5").

3. In the basement, due to not placing the basement wall in the correct place or something, the contractor notched the ends of all of the floor joists where they meet the sill. The ends of the rafters are notched at a depth of 2.5" and 7" long, with only 4" bearing on the sill. The max span is 16' (the other end of the joists are correct.)Joists are doubled/tripled under bathroom and partition wall.

What does everyone think? I thank everyone in advance for their help.

Mark E. Reme, PE

 

[diarmud]

Well,
From where I am,
1. Depends on the design but generally, you would need collar ties. No shims.
2. He's probably right, must be in his building code. I don't believe they should be staggered at all (personal preference). NYS requires they meed the ridge board and collar ties or gusset plate. Is this a ridge board or a ridge beam?
3. From the NYS building code, "notches in solid lumber joists, rafters and beams shall not exceed 1/6 of the depth of the member, shall not be longer than 1/3 of the depth of the member and shall notebe located in the middle 1/3 of the span. Notches at the ends of the members shall not exceed 1/4 the depth of the member."
4. 4" bearing is fine.

You really need to look at the building code in your state.
Sounds like your inspector is correct.

It doesn't matter if it will work, it must meet the building code, which your inspector has apparently read.

Regards,
Dermott

 

[Ron]

None of those are good. Depending on where you are, and since you mention a basement, my guess is that you have some level of snow load on the roof, there will be a fair amount of lateral thrust, so the collar ties are necessary.

Poor fit-up at the ridge beam is not good either. The shims are useless and likely won't stay in place unless fastened. Depending on load, you could exceed the allowable stress on the wood perpendicular to the grain. Further, the attachment would not likely be good. You didn't note whether he used rafter connectors or just toe-nailed them.

Joist notches likely exceed code allowable. Further, they are at the point of highest shear. Not good in either case. Also, since they bear on masonry, are they either pressure treated or isolated?

 

[msquared48]

I agree with Ron here.

Depending on the design, collar ties may or may not be required. If there is a large ridge beam, no, but if it is a ridge board, yes.

Generally, I do not like notching the ends of any members as it incurs horuizontal splitting and a corresponding shear and section modulus reduction, both affecting member strength. The deflection will also be greater.

With what the building inspector found, I would do an on-site inspection without the contractor there - as on the weekend. You will find more things I'm sure.

Mike McCann
MMC Engineering

 

[hokie66]

I hope your friend hasn't paid the carpenter. He is lucky to have a good building inspector. Give the inspection report to the carpenter with a notice of termination and find a competent builder.

 

[Ron]

While I agree with the result of hokie66's comment, you have to be a bit careful how you go about it to prevent a lawsuit for non-payment of the contract (yes...you can get sued by the contractor for non-payment even if he has thoroughly screwed it up).

First, get a hard reference for violations of the building code or accepted practice for each of the inspector's deficiencies, not just the fact that the inspector doesn't like it. That's not good enough. I'm not sure where you're located, but there's a building code that applies. In addition, the building code references other documents that are incorporated into code by that reference, such as publications of the American Forest and Paper Association that show proper wood framing techniques, span tables, etc.

The collar tie is a little different. That's a design feature. If plans were drawn for the structure that show no collar tie, then the designer has to justify that by proving that lateral thrust is accommodated, either by essentially no deflection of the ridge beam, by lateral resistance of the wall, or black magic. Even a rigid ridge beam does not remove all the lateral thrust and it could be argued that you still have sufficient lateral thrust to cause problems without a collar tie or ceiling joists.

As for the offset rafters...that makes no constructive sense...yes, theoretically it divides the load path from one side to the other, but who cares? It also induces lateral bending in a member (ridge beam) that is not designed for lateral bending or torsion, as would occur.

 

[bdSE]

The best answers require the best problem description, you did pretty well, but we are still lacking sizes of members to help with specifics.

As Dermott suggested, the roof question can best be answered if we know the size of the ridge member, plus how it is supported (thus a ridge beam or board). You can rest easier if it is a ridge beam (properly supported at each end by properly sized posts.) If it is a ridge beam of appropriate depth and thickness I think the only thing that needs to be repaired are the poorly cut rafters. The contractor will likely figure out how to reuse these members if possible, take note of how he repairs this roof area. A well done shim doesn't concern me as much as a missing collar tie if it is needed. If the bearing area of the rafter is in full contact with the use of shims, I think that would be alright... but I didn't see it in person. It might be difficult to cut the correctly angled shim to really give full contact of the rafter to the ridge member. If you can slip a dollar bill behind the connection request that he do it again.

At the first floor joists above the basement wall, the 2.5" notch info is good, but not the whole picture. We need to know how deep the original joists are (a 2x8 will actually measure 1.5" x 7.25"). If the notch is too deep for the joists used and loads applied it is possible to repair with the use of a wood (pressure treated) ledger on the inside face of the basement wall attached with epoxy or expansion anchors at a calculated interval. The acceptable notch depth can be taken as code depths or can be determined by an engineer. A structural engineer's calcs can override the prescriptive code in certain situations... I'm sure that goes without saying.

 

[COEngineeer]

Im supprised people still do ridge board + collar ties construction. I dont see that anymore on new construction. People either go with trusses OR ridge beam. I dont think I will ever specify ridge board & collar ties.

Never, but never question engineer's judgement

 

[RHTPE]

All of the above comments are valid. It is unfortunate that these things occur and the owner must be forced to accept a product less than designed. Assuming, of course, that the design drawings properly showed the contractor what to build, and did not presume that any contractor would know the right way to construct a wood-framed building.

The 2005 NDS (Eqn 3.4-3) provides the means to calculate allowable shear for a member notched at the bottom. It does not specify a maximum notch - using this equation and diarmud's 1/4 depth maximum, the resulting allowable shear force is 56.3% of that for a full depth member.

I disagree with Mike's statement though ("... and section modulus reduction, both affecting member strength. The deflection will also be greater.") - Section modulus is not as critical at the end of a simply supported member, and I doubt that deflection (controlled by I) will be measureably increased. Horizontal splitting will be a concern IF the member is close to its allowable shear force near the end that is notched. Chances are it is not.

Very few of my wood designs for longer spans have been controlled by shear strength - it's almost always bending stress or deflection controlled.


Ralph
Structures Consulting
Northeast USA

 

[msquared48]

RHTPE:

I stand by my statements here.

In general you are correct, but, depending on the application of the load and the extent of the horizontal crack, the bending and deflection can be affected. Particularly with transfer beams.

Mike McCann
MMC Engineering

 

[RHTPE]

To msquared48:

In this forum we tend to cover this kind of a problem in general terms. My statements really apply to joists in general. I think you would agree that a careful examination of all notched members is in order, particularly those used to carry loads other than uniform loads.

There are many 200+ year old structures in New England still standing that have the ends of members notched to 50% of their depth. A notched end of a joist, while not always desireable, is not necessarily indicative of a structural weakness.


Ralph
Structures Consulting
Northeast USA

 

[StructuralEd]

I think that installing a ridge beam may be a reasonable solution to eliminating the rafter thrust that may be pushing the walls outward. I would hire a professional to design it though; don't rely on the contractor's opinion. Or better yet, have him hire the Engineer.

 

[msquared48]

RHTPE:

I agree.

I don't doubt that such structures exist. They do here too. But then again, you have to look at what is above the top of the notch as opposed to what is needed to resist the applied load to get the true picture here.

Mike McCann
MMC Engineering

 

[Ron]

StructuralEd....a stiffer ridge beam only takes out one component of horizontal thrust...it will not eliminate it. If if reduced it to the point that the wall can handle it...great, otherwise, put in a collar tie or ceiling joists.

Secondly, DON'T have the contractor hire the engineer. Unfortunately there are engineers out there that will tell you what you want to hear. That's a shame on the profession, but it exists. Have the owner hire the engineer so that the engineer has no incentive to provide an opinion one way or another (the owner can win either way...if nothing's wrong, great...if something's wrong, he has a competent, unbiased individual to support his position).

 

[msquared48]

Ron:

If the rafters are notched to bear flat on the wall plate, and the rafters are also notched to bear flat on a joist hanger at the ridge, or the ridge beam itself, there is no horizontal component to spread the walls. To further mitigate this, a strap can be put over the top of the rafters at the ridge beam.

Mike McCann
MMC Engineering

 

[csd72]

markreme,

Just to summarise and clarify the above comments regarding the ridge beam.

There are two options regarding the design of the roof:

A. The roof is designed as an old style conventional pitched roof with a ridge board only and collar ties at the base. In this case it is assumed that there is no vertical reaction on the ridge board and that the reaction of each rafter is taken as axial load in the opposite rafter. As this reaction has a horizontal component then a collar tie is required to resist this at the bottom of the walls.

B. The roof is designed by the more modern method using a load bearing ridge beam at the top and no collar ties. In this case the ridge beam is designed to take the vertical reactions of the rafters and any horizontal components are ignored. There is no collar ties required for this type of design as the reactions at the top of the walls are vertical only.

So in my opinion, if the ridge board is not sufficient to take its share of the vertical load (as per B.) then the design has to be as per option A. and have collar ties and the rafters lining up with each other. If the ridge board is actually an adequate size ridge beam then it is okay as described.

Not sure about the shims and the rafter cutouts as I would need to see them to get a feel for it.

 

[StructuralEd]

Ron,

Provided that the contractor's Engineer is required to seal the drawing, what's the difference?
The owner should not be saddled with paying for the contractor's lack of savy.

And I don't agree that collar ties completely eliminate horizontal thrust on the structure; only from the opposing rafter action, same as the ridge beam can do.
Overall the wind pushes everything, roof and walls, laterally, tied or not, so the base structure needs shearwall capacity to take the lateral loads out.

 

[Ron]

Just because a contractor's engineer has to seal the drawing or recommendation, doesn't mean that he won't treat it as a "risk management" issue. Many engineers will sign things they shouldn't just because the statistics are on their side for failure. The problem is that sometimes, the design conditions get met...such as a hurricane or a seismic event. Then they wish they would have followed the standard of care!

I agree that the owner shouldn't be saddled with extra costs; however, he can ultimately get that back from the contractor it he so desires.

As for the collar tie....you're right if you only consider an unbalanced load or a wind load. Consider the condition with uniform snow load or wind load parallel to the ridge beam...you'll see the influence of the collar tie.

 

[csd72]

StructuralEd,

External lateral loads are taken out by in plane bracing of the roof. This works the same as flat roof bracing except for the vertical components at the external walls and at the ridge(where the two pitches oppose each other).

Shear wall capacity is required to take the loads to ground regardless of the roof construction.