I Beam Reinforcing Question

[MER3]

We built a cart to handle a 35kip heat exchanger. Due to an interference, the construction crew want to modify the cart as shown below. The beam is a W10x30 and they need to remove 6" of the beam for a clearance issue. Obviously solution 1 is to just not do it, but I am seeing if it is feasible at all. Due to the geometry of the cart, the loads in this particular beam are not great, the max. moment is only 72 in-k. My question is about reinforcing the underside of this beam after we cut out most of it. I can take my reinforced section with the plates we intend to add and show that the section modulus of the reinforced section still greatly exceeds that required for the loads and I can take the eccentricity into account for the small axial loads, but my question relates more to how in general to be sure that my new beam section is fully developed and is sharing the load. Is there any guidance for something like this?

Thanks for the help

 

[mike20793]

Where in the span are they trying to do this? At first glance, that fix doesn't look like it would cut it. I wouldn't let them cut my flange like that, even at the end of the beam near supports. You could check the remaining section as a WT and strengthen based off the extra strength but I would certainly extend it laterally to encompass the cutout more than currently shown.

 

[JAE]

Is the W10 only about 3 ft. long? Or is the drawing more schematic and not to scale?


Check out Eng-Tips Forum's Policies here:
faq731-376

 

[MER3]

The W10 is actually about 14ft long, but it is supporting no direct vertical loads. This beam only sees an applied moment due to the loads from pulling the cart.

This cut is actually two cuts located at each end of the span where the moments are the highest. I don't have it detailed there, but my reinforced section will extend past the cut on each side by ~6" before the taper begins back to the original bottom flange.

Here is the section view I should have included at first

 

[MER3]

Section View of the W10

 

[WARose]

Normally (for web openings) you have stiffeners on all sides......but in this case: the flange is missing. (Even 'Design of Steel and Composite Beams with Web Openings' (AISC Design Guide #2) doesn't address this to the best of my recollection.)

If you can make it work, I'd hang another W10 on the bottom flange (along with the stiffeners around the opening). You'd probably have to bolt it because a field weld would probably bring the whole thing down.

 

[MER3]

No it will be shop welded with no load so no concerns with collapse.

 

[DenverStruct]

I would add vertical plates on both sides and weld them to top flange of existing member and bottom flange of the new member. I am not sure I like the way you reinforce yours.

 

[MER3]

DenverStruct are you suggesting something like a welded gusset on each side?

 

[Once20036]

Any time you have a moment, you have a compression force and a tension force.
I don't see any load path for your compression force.

Away from the cut, you have compression in your top flange and tension in your bottom flange.
At your cut, you have compression in your original bottom flange and tension in your reinforcing.
Adjacent to the cut, the tension can "flow" into the reinforcing, but how does the compression "flow" into the bottom original flange? Through the web? That's taking the large area of steel in the original top flange and significantly reducing it / concentrating it through the web.

Why not add a second pair of diagonals, mirroring the geometry of the reinforcing to create this load path. Essentially it would be a kinked beam.
I`m not sure I'd love this approach for heavy loads or where deflection is critical, but for your structure it might be appropriate.

Keep in mind with this "kinked beam" approach, any axial force will create p-delta moments, possibly increasing your 72inkip significantly (or maybe not)

 

[MER3]

I considered my scab plates welded to the web as providing strong-axis resistance to localized flexural forces and making sure the portion being added to the underside of the beam is fully engaged.

 

[Mjkkb2]

There are a few misteries in this problem:

* you said the moment is the highest at ends where the cuts would be located - is the beam a fixed end beam? How is it supported??
* if the above is true them you do have a moment at the cut section and you need some continuity of the (tensile) flange that what you propose to add does not provide.
* if the beam is in fact simply supported then you don't have much moment at ends and your concern should be shear which is carried mostly by the web.

I think some more information is needed to see the whole picture of the problem.
A plan view would be helpful also..

 

[MER3]

The beam is attached to a vertical hydraulic jack at each end that serves as the leg of the cart. The load is carried directly over these hydraulic jacks therefore no applied vertical load to this beam. The only load this 14' beam sees is when the cart is rolled. Friction from the caster wheels acting opposite of the direction of travel makes the hydraulic jack legs want to fold under the cart. This moment couples out over the end of the W10 which I analyze as a simple beam with an applied moment at each end.

I do indeed have a moment at the cut section, which depending on the direction of travel of the cart could be compression or tensile. My rational currently for my design is that the scab plates I use to frame the opening will be bent about their strong axes as the section tries to open up. Their stiffness will help carry the load into the reinforcing provided on the underside of the beam.

I do not have much concern for shear, I am overlapping my spliced section with the W10 for a sufficient length to take the small shear loads. We can take the "steel is cheap" approach for sure, but I would like to try to put numbers to whatever I end up doing.

 

[dhengr]

Hardworker87:
If you don’t fix your sketch..., you don’t deserve another second of our time or our guesses at what you really have and are doing. Either the W10 is about 3' long, as JAE suggests, and your sketch proportions show; or it’s 14' long as you suggest and your sketch is all messed up, and you can’t have it both ways. You would be surprised at what meaningful first impressions an experienced engineer can draw from a well proportioned sketch with a few dimensions, sizes, thicknesses and some loads and their directions, etc. etc. We have no idea how you got your 72 "k moment, how and where it’s applied, how the beam is supported, etc. We can’t see it from here. You want our help, and you are perfectly happy to have us waste our time guessing what you have, all becuase you are to damn lazy to provide a good/meaningful sketch and explanation of your problem. I’m really tired of this kinda crap here on E-Tips. E-Tips is supposed to be about engineers talking to engineers in a meaningful way, not 20 questions for people who can’t define an engineering/technical problem.

 

[MER3]

I do not believe I have failed to define anything. If you look at my sketch, the length of the cut out portion is 12", the only thing I did not draw is the other 12' of beam off to either side of the cut. I was doing it for diagrammatic purposes only. My sketch is perfectly proportioned with regards to the reinforcing scheme, other than the extra length of unaffected beam not shown. I did not provide reinforcing plate thicknesses because I did not expect someone else to run an analysis, only point me towards how it might be done so that I do it on my own.

As far as the length, I have answered that question that it is indeed 14' long.

As for the loading, I developed that on my own and did not ask for a second check regarding the loading.

As for the support, I stated above that due to the configuration of the cart this beam is simply supported with an applied moment of 72k-in at each end. They need to cut the beam in two locations and want to reinforce as I have shown.

 

[Mjkkb2]

I have to agree a little bit with the above by dhengr. For many like myself that are in building structure design we envision some sort of a member that's part of the floor system. Obviously this is not. So without some additional sketch or drawing I myself have a hard time figuring out what we are talking about.
Regardless,if there is a flexural moment at the reduced/cut out section then you need to provide some flange continuity. Reinforcing the web is not enough in my opinion.
I know that you have everything else figured out when you're only asking about the reinforcing of the cut out part but if you provided us more detail of your whole set up you would've gotten a much better response. Good luck.

 

[MER3]

As an attempt at reconciliation, here is an updated sketch showing the to-scale lengths of what I want to do. I have provided a BMD in the sketch showing the moment distribution. Like I said, the moment diagram is subject to some change (up to and including a complete reversal) depending on which direction the cart is rolled, but the max moment is never greater than 72 k-in. I added more stiffener plates as well.

 

[BAretired]

From your latest sketch, you have a variable moment with maximum value of about 0.8Mo or about 58"k and a constant shear of 2Mo/L or about 430# across each short span. As I see it, you are proposing to use the bottom flange of the W10x30 to serve as the top flange of the notched portion. I can't see much of a problem with that, considering the low moment. If the moment was much higher, you might want to add stiffeners in the beam above each of the sloping flanges.

A simpler idea is to offset the 14'-0" long W10x30 and add 90 degree jogs at each end as required to meet the supports, but this may introduce more clearance problems of which I am not aware.

BA

 

[BAretired]

Actually, the W10x30 is really much too big for the job. You could consider a smaller member, offset as above. A 5x5 HSS might be a good choice.

BA

 

[MER3]

The cart was built a long time ago and they overlooked this one clearance issue when they built it, but otherwise yes a W10x30 has a FOS of over 10 in this application. Even when I calculate the Section Modulus of my new built up section the FOS is still around 6. And those FOS's are to AISC allowable stress, not yield.