Small Circular, Horizontal Openings in Reinforced Concrete Beams 3

[abusementpark]

I am having trouble finding guidelines for small horizontal, circular openings in reinforced concrete beams. I have a job where this is being requested and need to set some limitations for the architect and MEP consultant. I was hoping ACI would have some guidelines, similar to what AISC has, but can't find anything.

Since these penetrations are occurring in multiple locations at varying locations along the depth of the beams, I want to simplify this by establishing a maximum diameter and acceptable locations for the penetrations. My initial thought is that an opening with a diameter that is less than 20% of the beam depth through the middle 2/3 of the beam depth, at a location in the middle half of the beam span could be made with any adverse effects. But this is just a gut-feel. Any thoughts???

 

[KootK]

I like your proposal. I do this all the time for large prestressed girders. 4"-8" diameter max holes well spaced in something as shallow as 24" at times. I don't do any of the fancy analysis or fancy reinforcing schemes for these and, so far, it's worked out fine. If shear is tight, I'll prorate Vc to reflect the lost material in the diagonal tension plane.

 

[Lomarandil]

I'd buy off on that. Maybe add a minimum spacing (X diameters) between penetrations like KootK implied.

----
just call me Lo.

 

[WARose]

The two best references I've seen on this subject are:

'Design of Reinforced Concrete Beams With Web Openings', by: M.A. Mansur, Proceedings of the 6th Asia-Pacific Structural Engineering and Construction Conference, p.104-120

and

'Design Procedure for Reinforced Concrete Beams with Large Web Openings', by; Tan & Mansur, ACI Structural Journal/July-August 1996, p.404-410.

In the first reference, "small" holes are differentiated from "large" ones by behavior: small holes allow beam like behavior.....whereas large ones mean you have a mini "frame" within your beam that requires analysis. "Small" openings (defined in the first ref) are about 40% (or smaller) than the overall beam depth. In such a case, it is just a matter of keeping the opening in the tension zone of the beam, using a reduced "d" for shear and moment calculations, and proper detailing. But I would advise reading over these papers because I am skimming here.

However, for very small openings (relative to the beam size), I normally don't get too concerned as long as proper detailing is done.

 

[Agent666]

https://www.crcpress.com/Concrete-B...is-and-Design/Mansur-Tan/p/book/9780849374357

This text is by the same authors as WARose listed. It's pretty much the go to reference.

For smaller openings strut and tie can also be your best friend in trying to justify you don't need to do anything particularly special in terms of reinforcement.

 

[Agent666]

Just make sure for smaller openings that you realise that the (b) type of failure below can occur and it may mean you get limited engagement of beam stirrups adjacent to the opening, forcing you to put some smaller stirrups above/below the opening. This unreinforced section seems to be something people always miss based on my experiences.

 

[KootK]

The little bit of special detailing that I do for small holes is to place a double stirrup set either side of the hole. That, in anticipation of the failure mode on the sketch to the right above. Even then the stirrup anchorage starts to get a bit sketchy which is, in part, why I like to be closer to 25% depth for these and, as always, steer them out of high shear regions. A 1/3 depth hole starts to feel "large" once it gets near to a support, even if it's not drifting into Vierendeel mechanics.

 

[abusementpark]

KootK/Lomarandil - Do you guys think a minimum of 3 diameter spacing would be enough? And do you think I even need to avoid penetrating those quarter end spans, if the Vu/(Phi*Vn) isn't close to 1.0?

 

[KootK]

Do you guys think a minimum of 3 diameter spacing would be enough?

I'll go 5D for a canned beam response. Folks seem to expect 3D because, I think, you see that a lot with slabs and walls. In comparison, there's a lot more load concentrated into your average beam and a lot less redundancy. If I decide to push the 5D, I'll at least examine something like the diagram that Agent666 posted and ensure that I can tell some kind of plausible strut and tie model story, even if I don't run the numbers.

And do you think I even need to avoid penetrating those quarter end spans, if the Vu/(Phi*Vn) isn't close to 1.0?

I like the 1/4 point rule with a pre-approved penetration. One tends to have a lot going on at the ends of beams with moment reversal, anchorage of positive bending steel, and your fanned shear struts aggregating into one at the support. I think that it's a tall order to say that's all okay by inspection. And, often, you'll wind up being about 25% into the span by the time that you're past all of that stuff anyhow. I might be willing to abandon %span in favor of a multiple of beam depths. Maybe 3-5 x beam height or something?

 

[abusementpark]

Thanks KootK.

 

[Lomarandil]

Sorry, I haven't done the Vierendeel analysis to have a solid frame of reference.. My gut says that in many cases the 3D would work (away from supports, point loads, etc)

I do think that St Venant's principle would back up Koot's idea about restricting holes closer to a support than the lesser of 25% of the span or 3x beam height, rather than the blanket 25% of span.

And like KootK (I'm inferring here), whether I'd tell a contractor 3D/3x or 5D/5x (or more) would depend on the level of trust/control I have in the situation. There are always caveats and exceptions in the real world.

----
just call me Lo.