Your thoughts on situation 3

[shacked]

I am the EOR for this custom home that has already been framed and recently the trades; electricians, plumbers and fire sprinkler installers went through to install their wires and pipes. Since the floor joists are 1.75x9.25 microlam lvl joists I have a typical detail for the allowable hole size, spacing and location. Obviously the contractor was not there to supervise the trades since I showed up to the site last week and I see a whole section of joists with 2 large holes located at one end. See the attached picture & sketch of joist span showing holes.

This is a multi million dollar home and I do not believe that the owner should pay for a sub standard performing floor system, therefore I told the contractor that they would have to sister a 2x10 joist to the side of each of these joists, after the wires and pipes were removed. He agreed, but a week later I am now getting push back from the Architect and they do not want to do this.

The floor dead load is 27psf and the calcs for the joist at the reduced sections are still working, but my fear is that the weakened joist floor system may have issues in the long run.

Since it is difficult quantify the performance of the floor system of engineered wood joists with a lot of holes drilled through their length, I think that my fix is reasonable considering the cost of the home. What are some of your thoughts?

 

[JoshPlumSE]

Have you run any calculations? What's the shear demand on the joist on this end? I'd compare that to the shear capacity of the joist using may half the joist area. If that's good, then I think you could leave it as is.

If you're still not satisfied, I would think there would be other ways to strengthen this as well.... Maybe adding horizontal straps above and below the openings? A partial length, partial depth sister?

 

[dik]

Any issues with the high dead load... nearly 3x that of a normal house?

Rather than think climate change and the corona virus as science, think of it as the wrath of God. Feel any better?

-Dik

 

[shacked]

Josh, the shear is well below the allowable, 44% of capacity. My concern is the performance of the joist over time, or potential cracking of the wood around the holes, and this is only based on intuition.
dik, what is high dead load?

 

[Agent666]

Based on experience the ones close to the centre wouldn't seem that bad (to me) in terms of requiring reinforcement based on the photo/elevation.

However, there is more to it that simply checking the shear to determine if the member is sufficient for openings in timber.

This design guide might give you some ideas.

But typically it is very simple and effective to reinforce these types of penetrations with some long fully threaded Spax screws or similar driven each side of the openings. If designed correctly they act to prevent the longitudinal splitting that naturally wants to occur.

If preformed holes in the factory are utilised on a project it is often easier to do these screws or other types of rienforcement in the factor than on site. As an alternative, ply reinforcement is effective as well as a reinforcement method. Simply sistering the joists seems overkill and not all to effective unless the goal is to effectively replace the original compromised joists. If the numbers work out regarding any reinforcement that might be required and the design of such reinforcement is undertaken following similar methods to which I linked to then I wouldn't really see any issue long term.

https://engineervsheep.com

 

[Eric C.]

I believe I would contact the LVL manufacturer and send them all the pertinent information and ask for their opinion. Its possible that there could be a warranty issue if the joists weren't installed in accordance with the manufacture's/engineer's instructions, even if the capacities check out.

 

[dik]

27psf is a little higher than what I've usually encountered... I've had greater... thick set tile, etc. I was just wondering if there was a problem with the sustained loading.

Rather than think climate change and the corona virus as science, think of it as the wrath of God. Feel any better?

-Dik

 

[phamENG]

This paper should prove useful for you.

Design of Reinforcement Around Holes in Laminated Veneer Lumber Beams

Be very careful with these. Wood beams really don't fail in vertical shear. If you stay within flexural stress limits and bearing stress limits, you essentially preclude vertical shear. The shear value for lumber is for transverse shear - the kind that holds the plies together and prevents them from slipping while in bending. The NDS does not have an allowable shear value for vertical shear (V=wl/2).

The first failure mode is usually a tensile rupture perpendicular to the grain at the top "corner" of the hole, similar to a bad bearing notch.

I've seen opinions that vary quite a bit. Some say that you have to assume half of the moment is carried above and half below the hole, so if not centered you take the smaller remaining section, look at it in bending with the max moment expected at that section, and calculate your transverse shear stress based on that. This can be really high. I agree that this approach is appropriate for rectangular holes, but for round holes with negligible damage on the peripheries can be checked based on flexural stress amplification from holes. Some experimental results are available in this paper, but note that the equations overestimated the new section capacity - until I can prove otherwise I take the result of the equation and then multiply it by a factor of 0.8 to reduce the capacity to something lower than this limited sample determined.

Let me know what else you find. After starting it 6 or 7 years ago, I'm finally going to finish my Master's and got approval yesterday to study this problem in dimension lumber joists - I've seen a spike in calls for certifications/fixes of large holes in 2x8 and 2x10 joists since the pandemic started. Lots of DIY (or not) plumbing projects, attic conversions, etc...

 

[shacked]

Thanks for your help guys. Pham, also I appreciate your help. I briefly reviewed the pdf and it looks super helpfull. I'm glad someone is researching this specific problem.

 

[phamENG]

My pleasure. It struck me as odd that the only coherent guidance on the topic for dimension lumber is the set of limits in the prescriptive code. Digging into those, I find they stretch back to the earliest model building codes with no explanation or reason behind them. It's worked for a long time, so it's what we do.

To me, the point of being an engineer isn't just to tell everyone how we've always done it, but to find better ways of doing it and solving the otherwise unsolvable problems. This seems to fit that category.

 

[JLSE]

I tend to agree. Owners paid for a standard. Id want to put it back on the contractor, have them build it per the approved documents.... so, replace them in-kind. dont sister joists, you dont want the extra weight. insist they comply with the approved detail for allowable holes. this isnt a huge deal, and this issue should be well known. its their fault. I would insist... its not a good opportunity to get tricky... its an opportunity to insist on a standard.

 

[Mike Mike]

shacked,
If your joist calculations considered the reduced cross section, accounted for stress concentrations, followed the building code, and found only a 44% stress ratio, no repair is needed. If you're concerned about performance over time, check them for vibration and creep. A repair will not give the owner a noticeably better product. Every job has holes in joists. If the numbers still work, don't fix them.

phamENG,
Thanks for the reads. I didn't get too in depth, but it seems the math in the second paper isn't considering stress concentrations at holes, so I find it odd their observed/predicted values in table 4 weren't lower.
In the first paper, I like the idea of adding screws each side of penetrations but I didn't understand how I should calculate the screw reinforced shear capacity.
Is section 5.4 telling me the ratio of maximum shear to average shear is 6.4 if the hole is half the depth? In comparison to a 1.5 ratio for a joist without notches?
Also, please elaborate on vertical VS transverse shear. I was under the impression the "shear" equations in NDS chapter 3 were for vertical shear.

 

[phamENG]

Mike Mike - don't have time to dig back into the papers at the moment to follow up on the first few, but I'll jump on the last one (since my NDS is in front of me...I sure know how to party on a Friday night...)

References are for the 2015 edition. Open up to 3.4 Bending Members - Shear. The first section, 3.4.1, is titled "Strength in Shear Parallel to Grain (Horizontal Shear)". So there you have it...it's not vertical shear. To dig a little deeper, take a look at equation 3.4-1, f[sub]v[/sub]=VQ/Ib. If you've ever designed a built up steel shape (like a crane rail or similar with a wide flange beam and a channel on top), this should look familiar - it's a shear flow calculation. It's what you use to size the welds on that built up member, because it's based on the horizontal shear between the two members. In the wood beam, each layer of grain is a "member" and you're checking to ensure that the lignin holding the wood fibers together can transfer the horizontal shear between members.

When I have some time to spare I'll try to address some of the other questions you have.

I'd be careful in your advice to not worry about the holes. The way you seem to be talking about it is as though you're discussing an isotropic material - which wood most certainly is not. The holes in the picture are going to act more like large notches, which can introduce atypical failure modes that can only be precluded with an appropriate repair.

 

[Mike Mike]

phamENG,

I hope you had a good party last night. That makes a lot of sense. Thanks for pointing that out. I had always imagined wood beam shear failure looking similar to concrete beam shear failure. And whatever the material, the equations describe a vertical cross section so I guess I just figured the failure plane was somewhat vertical.

If the concern is anisotropic behavior, incorporate that behavior into your finite element analysis. If the concern is your analysis isn't able to account for real world atypical failure modes, do a full scale test or look up the results of similar full scale testing. My point is unreinforced joist penetrations are extremely common, and often fall outside the limited guidance of the NDS. Instead of reinforcing all of them, let's take the time to develop an understanding of their unreinforced abilities (if we have the time, and I assume if you're here reading this you have some time). If the penetration has been accounted for and the stress ratio is still only 44%, let it slide.

 

[phamENG]

Yeah. Real exciting. Let me tell you....

Glad it was helpful. It took me an embarrassingly long time to realize what that section was saying.

The existing limitations for sawn lumber are based more on rules of thumb than actual experimental results (they haven't changed since the 20s - or at least that's the oldest reference I've found). The LVLs, on the other hand, are based on testing carried out by the manufacturers. Since the holes in the OP fall outside of what the manufacturer says is allowable, I don't think it's a good idea to call it okay without doing something to reinforce it - though I agree there is some room for judgement there. This doesn't feel like a good place to dismiss those guidelines, though. I also don't think anyone who is doing a lot of residential repair designs has the means to conduct a full finite element analysis of a joist, much less do it and accurately model the nuances of wood behavior.

When I start down the hands-on research road on this I might get access to it from the University, but I'm not about to spend the money on that for myself. Besides, I find FEM of wood to be something of a fools errand. You may be able to model a clear specimen with some accuracy, but the permutations of grain size and angle, knot size and location, etc. would severely limit the applicability of any results in a real world setting. It's like the old adage of measuring with a micrometer, marking with a crayon, and cutting with an ax.

 

[thebard3]

If I were the EOR,
1- I would note that trades should know that you drill holes as close to the center of structural members as possible.
2- I would note that the holes drilled are much larger than necessary.
3- Typical building codes include verbiage regarding good workmanship. I would suggest that I believe this does not meet that criteria.
4- I would consider whether the 'typical detail' that I supplied should be considered as a specification or not. If it is a specification that was not followed then I would leave no alternative but to remediate the situation to my satisfaction.
5- If a contractor did that to my house I would not like it.

Brad Waybright

The more you know, the more you know you don't know.

 

[HouseBoy]

Agent 666, regarding this:
"reinforce these types of penetrations with some long fully threaded Spax screws or similar driven each side of the openings."
Just to clarify for me - you're talking about screws that would be installed vertically in the joists ... (is that right?).
I like that concept but I always wonder about making the screw sufficiently small diameter.I like the Simpson SDWC15600 for something like that (5/32 didm and fully threaded). Sometimes I add plywood on the sides as a way to apply "external shear reinforcing" to help resist splitting.

Just wanting to make sure I understand what you are suggesting.


AG666 and others - Thanks for the references. Now I'll have a party on some Friday night!!

 

[Agent666]

Yeah vertical screws, generally driven each side of penetrations, the design guide I posted earlier has a design method for evaluating this arrangement and some other guidance on ply reinforcement.

https://engineervsheep.com

 

[phamENG]

The vertical screws are great...if it's new construction. In retrofits it can be hard to implement since there's often a finish either above or below the work that nobody planned on touching.