Reinforcing steel beam

[ajk1]

To strengthen a steel beam that has ready access one side only, I am considering welding in a 35M weldable grade rebar top and bottom, one side of the web only, where the web meets the flange. However that makes it a Class 3 Section, because it is not doubly symmetrical. Can adding lateral braces at close spacing to the compression flange allow the beam to be checked as a Class 2 Section? I am using CSA Standard S16.1.

I know that there are other ways to strengthen the beam, such as welding a tee to the underside, but that requires breaking out the block wall below, spray fireproofing the tee and rebuilding the block wall below.

 

[KootK]

How does that get me to a "do nothing" state?

When your steel beam deflects there will be a couple of parallel waffle ribs that will go along for the ride and assume more or less the same curvature. And that means that they'll be sharing the load. Waffle slabs are usually fairly thick so I'm guessing that you might have a pair of 12"-ish continuous concrete beams sharing the load with your 16" simple span steel beam. That sounds like meaningful help to me.

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.

 

[FishTheStructure]

When I reinforce beams with bars I use round bars, not reinforcing bars - they fit better without the deformations, the end product looks somewhat cleaner, and there are no questions regarding their weldability.
Also, I have found that while smaller diameter bars are easy to weld to the beam, the welds become more difficult with larger diameter bars as you end up welding in a deep groove which is both difficult to complete as well as review/inspect.



Great spirits have always encountered violent opposition from mediocre minds - Albert Einstein

 

[KootK]

I have found that while smaller diameter bars are easy to weld to the beam, the welds become more difficult with larger diameter bars as you end up welding in a deep groove which is both difficult to complete as well as review/inspect.

I've been noodling on this a bit too. I was wondering if one might groove weld the rods to a small sized angle in the shop and then weld the angle legs to the beam in the field (clean fillet welds). You'd loose a little effective depth but gain some extra area so maybe that's a wash. Not sure how one would accomplish a splice if that's necessary.

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.

 

[structSU10]

Have you thought about adding an angle that toes into the underside of the flange and into the web? It makes a small box of the flange, which should help with potential torsion from asymmetry as well. That can provide a lot of steel, and the angle will be thinner so welding is less of a problem (I imagine with a big bar you have to preheat and that seems like a pain). If you have transverse beams into the one you are reinforcing, I can see it being a problem though...

 

[ajk1]

to Kootk - the slab is 5" deep rib plus 2½" slab = 5½" total. The beam seems to be centred under one rib. Your suggestion of welding the bars to an angle is interesting, but in the end must still weld the angle to the beam.

to Fishthestructure - very interesting advice about using round bars. I'll keep that in mind.

to structSU10 - what you suggest of making a box of the angles was my very first idea (although I did not mention it here) but our chief engineer was not enthusiastic about it because he said it was not an efficient way to do it (since the centre of gravity of the angles was closer to the beam mid-depth).

New Information:
I was on site again this morning and found that the block wall is not directly under the beam but about 2" clear of it. This suggests to me that the best thing would be to weld a tee section to the soffit of the beam. It is I suppose overhead welding, but is that really so difficult/expensive? The welding would have to be done from one side only, but if a continuous backer bar were welded to one side of the web of the tee before it was lifted into place, then I would think that a full penetration butt weld could be made from one side only. What do you think?

 

[ajk1]

The attached is what I have sketched up for the added tee on the bottom, for concept only. I have not yet done the calculation of its resisting moment. Have to read up on how the plastic resisting moment for this configuration should be calculated.

 

[RPMG]

Search "Newman beam strengthening". There is a webinar print out document that is very helpful for beam reinforcement.

 

[ajk1]

To RPMG -- thank you, I will read that. Sounds like just what I am looking for. Much appreciated.

 

[ajk1]

Can the existing 2.5" of dry pack grout between the top of the steel beam and the soffit of the waffle slab ribs (that run at 90 degrees to the steel beam span, at 2 foot centres) be considered as lateral support for the steel beam? I don't think so, but someone suggested to me that it might.

 

[ajk1]

And a further question: Should we jack the slab up say 1/4" to take some of the load off the steel beam before welding the new steel reinforcement to the beam, or is that not necessary? Have to be careful not to do more harm than good (such as making more cracks in the block wall that sits on the floor here, although we would monitor the movement with dial gauge. I have jacked up a wood floor about 1/2" a few years ago and it came out quite well, but there were not much brittle materials involved).

 

[structSU10]

I picture it as a somewhat complicated problem. Your existing steel has stress/strain in it. If you attach the reinforcing to it, and assume the current stress/strain remains in the member, when it takes more load there is potential for the original member to begin to plasticize much sooner than the reinforcement. But if this happens, the greatly increasing strain in the existing member is restrained by the new reinforcing that may still be elastic - so perhaps it starts to dump extra load into the reinforcing due to the differing strain rate at the interface? Basically though, I think its generally OK to not jack the member due to the large ductility of steel. In the US at least we tend to hold L/600 or 0.3" (the more stringent) for something supporting masonry to help prevent cracking.

 

[ajk1]

To structSU10 - thanks.Much appreciated. Is the L/600 based on live load or total load?

 

[structSU10]

The L/600 applies to whatever load is applied post construction of masonry. Generally I use super-imposed dead load (include weight of masonry) and live load.

 

[KootK]

I'd lean towards not jacking the beam to avoid damage to the existing block wall. And I'd prefer a more positive lateral restraint for the top of the beam. Maybe you can tie back to a few of the perpendicular ribs.

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.

 

[ajk1]

To StructSU10 - ok, thanks. I think I have used that limit as well (I think that may be a published number somewhere). Thanks again. much apprciated.

to Kootk: I agree with what you say. It will apply to a number of beams and may happen at other floors up the building. Anyway, if that is what has to be done, that is what will have to be done, but we will have to check the situation on the other floors. Maybe the beams are not as heavily loaded. It would be interesting if anyone has ever done a load test for this condition to see if the beams buckle laterally.

 

[ajk1]

How to determine the length of longitudinal weld near the end of a steel T section welded to the bottom of a wide flange steel beam to reinforce it (the stem of the Tee would be welded to the beam bottom flange)? For a steel reinforcing plate, this longitudinal weld length can be taken as 2 times the width of the plate. How would the required length of longitudinal weld be determined for a Tee?

 

[KootK]

America does a good job of this. According to the AISC manual:

1) The attachment beyond the theoretical cutoff point should be adequate to develop the tee's portion of the flexural strength in the beam at the theoretical cutoff point.

2) If the tee will be stressed beyond fy, it's tricky and you can approximate this as the the axial yield strength of the tee.

3) If the tee will not be stressed beyond fy, you can use the formula M/QI.

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.

 

[ajk1]

Kootk:

Thanks. I have some follow up questions:

1. I am checking the reinforced section by limit states design which assumes that the entire portion of the reinforced section each side of the plastic neutral axis (PNA) has yielded. How do I determine if the tee will be stressed beyond fy?

2. For a plate used as reinforcement, there is, in addition to the requirement for the weld to develop fy times the plate area, a requirement that the length be not less than 2 times the plate width (if there is no transverse end weld). For the Tee section, is there only the requirement to develop As x fy?

 

[KootK]

1) Do M/S at the theoretical cutoff point.

2) Rules like that are basically addressing shear lag I think. As such, I'd apply the shear lag provisions for tension members to the tee considering the "connection" to be the portion of the tee extending beyond theoretical cutoff.

What percentage of the span do you plan to reinforce?

For what it's worth, I usually specify stabilizer plates at the ends of the tee. My understanding is that they are required sometimes. Unfortunately, I haven't yet figured out how to assess that need so I've been specifying them all the time.

When using plastic moment capacity, I'll do an extra check of my own. I'll ensure that the reinforcing can develop it's yield capacity on either side of the peak moment with a 2x safety margin. It's not been an issue yet.

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.

 

[ajk1]

ok, thanks. Much appreciated.
I have not yet determined what percentage of the span to reinforce. That's my next step.
If the TEE is welded intermittently along its length (say 2" every 12"), and at each end to develop tee area x fy, can I take it that for service load deflection calculation the inertia to use is the composite section of beam and tee, or is it something less than that due to the intermittent welding?