Structural steel "T" cantilever beam design.

[DavidKingsmill]

I am attempting to design a structural steel "T" cantilever beam for a balcony. This is proving very difficult. I have managed to model it to find a deflection but moment capacity is proving impossible. Can anyone help?

Thanks.

 

[Stillerz]

What does 3.6.3 state?

 

[DavidKingsmill]

Here are a few extracts including 3.6.2 and 3.6.3 etc

 

[Paperandpencil]

Can you follow the method given in clause 3.6 and subsequent clauses? Alternatively, for a capacity check, limit the theoretical web ratio D/t to 8 x epsilon, or 15 x epsilon if you are feeling bolder, and calculate effective section properties, Z eff ? Clause 4.3.8.3 gives guidance for an angle section, and a T is like two back to back angles. The web of a 178 UB is pretty thin. I have used fabricated T sections before so that you get a thicker stem. Beware bolting this lot to a masonry wall.

 

[Stillerz]

read closely section 3.6.2.2 on Effective Area. It clearly states what effective widths should be used for class 4 web and flange elements in conjuction with figure 8.

Read closely 3.6.2.4.

Section modulus is simply I/C.
or Zx = Ix/Cx.
In your case Ix will be based on effective widths "b eff" and therefore will change or reduce Ix for the full section. I am assuming that Cx (distance to the cnetroid) will also be based on the REDUCED Ix.
Also, you will wind up with two different Z values as there will be two values for "C" as for a Tee shape the centroid will not be at mid depth of the section.

 

[DavidKingsmill]

Bolting it to a masonry wall is the plan.

Unfortunately I have got to go.

Thanks to everyone again.

I will have a look at this thread again tomorrow.

Cheers.

 

[Stillerz]

I should state clearly here that I have never used this code before and I am only trying to offer some insight based an VERY limited reading (10 minutes worth)

 

[DavidKingsmill]

Cheers stllerz.

Will read up again tomorrow.

 

[BAretired]

I suggest that you invert the tapered T beam and add a welded top plate. Then you have a symmetrical section which is a bit easier to analyze because lateral torsional buckling is not likely to be critical.

The supporting beam will rotate torsionally so that the deflection will be greater for interior members than exterior. This will be partially offset by the 80 x 60 HSS at the tips of the cantilevers.

Your connection to the wall is questionable. Are you expecting to transfer all of the moment at each end through the 120 x 80 HSS by means of a couple of through bolts? Not a good idea!

BA

 

[ooox]

The Steel Designer's Manual gives section properties for structural Ts cut from universal beams/columns.

I would also look closely at your wall connection. The last time I did one of these using Hilti anchors the recommended shear and tensile capacity for an M12 was around 3.0kN. See if its possible to do some pull out tests if your brickwork properties are unknown.

 

[Ussuri]

David

Can I suggest you read the SCI advisory desk note AD311 T sections in bending - stem in compression.

It provides two methods for designing the T using modifications to BS 5950.

You know how to calculate Z from I? Zeff is done the same way but instead of using the I value for the full section, you use the I value for the effective section as detailed in Figure 8.

 

[ishvaaag]

I did some worksheet around 2000. This is a printout of one, check if everything seems logical, and helps, hope so.

 

[BAretired]

The cost of adding a flange on the bottom of each tee is peanuts. I suggest you do that and quit worrying about a problem which is far beyond the average engineer's understanding.

Instead, focus on the issue of how to safely secure the steel structure to the wall.

BA

 

[DavidKingsmill]

Thanks BA.

I am trying to give the client and architect what he/she wants.

I have managed to do some analysis of moment and deflection.

Thanks all.

 

[hokie66]

Using tees for cantilevered beams is, for some reason, trendy with architects. It is difficult for them to get it through their artsy heads that the use of steel in that manner is not structurally honest.

I suggest that if you are forced to do this, you should look at using a welded tee instead of a tee split from a wide flange section. That way, you can use a thicker web, more resistant to buckling.

 

[SarkTower]

I have written a small VB application for calculating the relevent section properties for T sections. It's metric and member resistances are calculated in accordance with BS5950 and AD311. If anyone is interested in verfying the results, I will post it on this thread once I've tidied it up.

 

[patswfc]

Sark, that would be good thanks

 

[SarkTower]

I hope this is correct and of use to all interested. Please use with caution, I normally check my stuff using published examples, but guess what? - I can't find any for this scenario. Barring any silly mistakes, I think all the geometric properties are calculated OK and I have checked a few against published tables. It's my interpretation of the British Standard which is possibly subject to error, so anyone who knows enough about it's application to verify the results, would be greatly appreciated.
I have only uploaded the exe file as I think most PCs will have the supporting files necessary.

 

[BAretired]

Sark Tower,

Your program does not consider unbraced length of the bottom of the web. Or am I missing something?

BA

 

[SarkTower]

BA
No you're not missing anything. It's one of the things still to add, just ran out of time, I'm only considering local capacities at present (still to digest the appendices of the BS). I hope to get this done by the end of the week, and I'll post again.

 

[DavidKingsmill]

The lack of material available, my lack of experience/knowledge, the small fee and limited time for the job has proved difficult.

I am amazed there is not a similar example published somewhere.