Deflection of 25mm from SE, problem for sliding doors

[Heather85uk]

 

Apologies to add to the theme of questions about deflection of rsjs over sliding doors..

slight panic.. as we are in construction.. (a two storey rear extension and loft conversion with gable window. )

we have a goal post design from our SE ( who we have not had a good experience with, another story)  

she has calculated to use a horizontal UB of 254 x 254 x 89kg 

we also want to spot weld a steel plate all the way across. 

however, express bifold have stated they require a deflection of no greater than 3mm 

And she has said that the deflection is 25mm and that it’s impossible to reduce this.

the length of the UB is 6.7m span over sliding doors 6m. They are bottom hung. 
 

After hours of reading I am gathering that increasing the height of the beam may help?

356 or 406 using a shallow UC..( taller and lighter ?) 

 

Will pre-cambering help? any ideas on cost increase? 
 

is it possible to add a layer around the frame using a compress-able material that will absorb some movement ?  As well as specialist silicone. 
 

Live load and dead load shown on calcs below.. 

 

Any suggestions on how to not have

dis functioning sliding doors and cracked brickwork? Losing sleep over this one ! Any help gratefully received! 

 

Thanks in advance, 

Heather 

 

 

 

 

 

 

 

 

[markc]

A UB would be much better than a UC. And increasing the depth would make a massive difference.

weight wise a deeper beam would be lighter than a UC.

25mm deflection is a ridiculous amount and would create many more problems.

pre cambering  would definitely help,

 

 

[nod]

As above Deflection of 25 mil is ridiculous 

We got over this with a welded hit and miss plate on all three large openings 

and intend to do the same again on our next build 

 

 

[MikeSharp01]

I think the expected deflection is 22.233mm but is still daft across this span. Were they trying to get the depth of the beam the same as the floor if not then cannot see why it can't be deeper? Also although pre cambering will help with the dead load portion  it won't deal with the live / imposed load portion. IE it can be straight under the dead load but will then deflect, as calculated, under the worst case imposed load. The imposed load here is, simplistically, 2/3rd of the deflection so I am wondering how it is derived.  The imposed point loading @ 22Kn seems quite high, its like having two ton bags of ballast on a point along them beam, we cannot see the rest of the design but so it could be something like another beam, supporting a floor possibly, coming in from the side, but it feels high. Also remember, but perhaps don't try and work with it - IE by saying we are never going to do it, that the imposed load could be based on / derived from you having a crowded dance party in the room above / adjacent to the beam. 

[Gus Potter]

8 hours ago, Heather85uk said:

 

Apologies to add to the theme of questions about deflection of rsjs over sliding doors..

 

 

Hello Heather85uk.

 

No need. This subject crops up on BH a fair bit.

 

The calcs raise a number of questions and you mention that the steel section is part of a goal post. There are a few permutations of "goalposts" and each permutation can influence the deflection and the choice of beam depth / width and so on. I think you need to take a few steps back first to be able to understand more of the principles, then you can go back to your SE with a bit more knowledge and ask why you have what you have. For example I can see that the beam is designed as simply supported, but why? The deflection is shown as overall deflection but is not split out into live and dead load defection and so on.

 

If you wish can you provide a bit more info. It would be helpful know what is forming the posts each side and what they rest on at the bottom. Maybe a cross section that shows the wall thickness, buildup and materials. Also are your founds cast and how far have you got with say building the substructure? Are there height restrictions on the beam depth over the doors?

 

Although this topic has been explored to some extent on BH if I can I'll write a general post over the weekend that explains more about the structural behaviour of the different types of goal post arrangement for those who are less familiar with the subject, how the different permutations influence the deflections, twisting effects and so on. I'll try and tailor it to your situation if possible and hopefully this will help you find the right solution for your build stage.

 

You may have already read this thread on BH about door deflections.

 

 

 

 

 

 

 

 

[saveasteading]

On 02/09/2021 at 22:56, Heather85uk said:

we also want to spot weld a steel plate all the way across. 

 

And she has said that the deflection is 25mm and that it’s impossible to reduce this.

6m is a long way for the beam to span, so a significant deflection is likely and 3mm is going to be tricky.

It is very much easier for the door supplier to 'require' it than the SE to design for it, so please be collaborative.

 

'Impossible to reduce' is not correct. Either there is a misunderstanding, or there is something we don't know, such as a limited space being available for this beam, hence the column section that has been selected. 

 

Pre-cambering will not reduce deflection, but control it to some extent. When closing the doors, the runners will be going uphill, requiring effort, and the beam will begin to deflect down to the required level. I think a lot  can go wrong with this.

 

A plate welded on the side will stiffen the beam enormously, and even better if done both sides.

That then requires a more complex calculation, and I suspect that the programme used will not be able to do it. 

Presumably this SE or someone else in the office will be able to calculate from first principles. 

This will take some time and skill so expect a fee for the reworking.

 

The beam is designed as simply supported. (On the calculations see 'rotationally free') If it can be built hard into the walls at either side, or become integral with the posts then this will stiffen it a lot.

Costs will apply in steel, opening size and foundations,  but this is what happens with very large door openings.

 

Possible solution then, subject to site particulars,  using one or more of these:

use deeper beam if space allows.

or  add plates to thicken the flanges

add side plates

design as an integrated 'goalpost' portal frame.

[Dave Jones]

what a load of tosh!

 

used a 127 x 76mm x 13kg/m beam with a 300mm welded plate to hold up a gable and its not deflected more than 3mm.

 

Will have zero problems with bifolds.

[George]

The deflection in the calculation is dead + live.

 

The dead load deflection will not effect the bi-fold doors.

 

You need the SE to look at only the live load deflection and to keep that controlled. It may also be useful to consider what the likely loading pattern will be as you've got 3 sets of live load and it is unlikely to be fully loaded simultaneously, although without knowing exactly what these are, it's hard to say for sure.

 

Pre-cambering is OTT for this application.

 

End conditions should be taken into account, especially in regards to live load if you do need a more realistic deflection calculation.

 

Steel beams with masonry on top don't actually deflect as much as is calculated due to the effect of friction forming a composite beam (steel + brick). However, while this was researched in the 50s, there is no recognised guidance on actually accounting for this effect (mainly because if it does crack, the effect is lost). (As an aside, I did take advantage of this in my own house by attaching frame ties to the top flanges in the perps of the blockwork, then reinforcing the blockwork with bed joint reinforcement. No idea what the actual effect was, though! And the beam was within tolerance anyway.)

 

 

Edited September 8, 2021 by George