Acceptable Steel Deviation

[Kalibre46]

Hi,

We've just had the first part of a large steel structure installed for our house extension. The key feature of this room/house is that internally there will be a two storey ceiling with one corner being a combination of bi-folding doors and fixed glazing above. The cool part is that there is no corner post to support the building, giving an illusion of a floating corner.

A picture paints a thousand words, so hopefully you can see this pic. You can see the cantilevered floating portion quite clearly.

The issue is that between the centre post (the bit the ladder is leaning against) and the edge of the floating corner (the bit the timber is rested against) the steel bends downwards by 40mm. The steel is installed straight however due to the 'plate welding and galvanisation process' it deviates at that end by this amount. Given that the distance from the post to the corner is 3250mm this effectively equates to a 1:81 gradient.

The steel company states that it is within their tolerances and it is not an issue, however the glazing company say that it goes beyond their tolerances for what can be 'packed' effectively.

Before having further awkward conversations and not having access to something like BS EN 1090-2 I thought I would ask you kind souls on here.

Acceptable tolerance or not?

My structural engineer has no concerns from a strength point of view and is therefore not commenting on what is acceptable or not.

Many thanks,

Doug

(P.S There will of course be eventual loading on this area so the deviation may even increase slightly.)

 

[^woody^]

Part of the calculations are for deflection, which is how much the beam bends and impacts on internal finishes and frames and such like. It is not about strength, and different to a bending calculation.

Ask your engineer if the defection is within permissible limits. Has he calculated the permissible limits?

 

[freddiemercurystwin]

Have you checked the sizes of the steels?

 

[Kalibre46]

Hi, the steels are all the correct size and I don't rally think it's a case of deflection as there is no load on them as yet, it's actually that the centre steel was produced 'bent' for want of a better word at one end. The steel company say that this degree of bending (40mm) is within their tolerances which I find hard to believe.

 

[bernardgreen]

If it is single length of steel and that central joint to the verticals is a bolted joint and not a welded joint then I would say there is no reason for than beam to have that much bend in it.

Welding if not done properly can distort an RSJ but there doen't apear to be any welding at the joint.

Galvanising should not cause a beam to distort.

I cannot recall the exact amount but when we built a timber framed house the 3 metre beams of natural timber did not have anything like that much distortion.

 

[theprinceofdarkness]

If it is single length of steel and that central joint to the verticals is a bolted joint and not a welded joint then I would say there is no reason for than beam to have that much bend in it.

Welding if not done properly can distort an RSJ but there doen't apear to be any welding at the joint.

Galvanising should not cause a beam to distort.

I cannot recall the exact amount but when we built a timber framed house the 3 metre beams of natural timber did not have anything like that much distortion.

 

[theprinceofdarkness]

I wonder if the 1/2 RSJ nearest the camera is the problem as it seems to be resting on the corner in question. The flanges seem to have been cut back to the left of the pillar.
Frank

 

[jeds]

40mm is way too much. The maximum allowable inclination of a horizontal beam under the current British Standard is (10 + l/500)mm. (l being the span) For a 3250mm beam that would be 16.5mm. Loading the beam will increase the deviation even more.

 

[tony1851]

What I cannot understand here is; is the beam continuous (ie running from the far post, fixed to the nearer post, and then cantilevered past),
or is it a shorter beam just bolted to the nearer post?
If it's the latter, that would be a poor design because of the moment at the connection.

Regardless of that, if it's just a 3.25m beam and has gone into an arc-shape, that would be poor welding. As above, a 40 deviation would be excessive.
We've specd. plates to be welded to UC sections longer than that, and none have deformed anywhere near that.

 

[Kalibre46]

Thanks for the replies.

What I cannot understand here is; is the beam continuous (ie running from the far post, fixed to the nearer post, and then cantilevered past),
or is it a shorter beam just bolted to the nearer post?
If it's the latter, that would be a poor design because of the moment at the connection.

Regardless of that, if it's just a 3.25m beam and has gone into an arc-shape, that would be poor welding. As above, a 40 deviation would be excessive.
We've specd. plates to be welded to UC sections longer than that, and none have deformed anywhere near that.

It's the former, i.e. a long post (6m) fixed at one end and then cantilevered in the middle. The end post and the middle post heights are mm perfect so the 'bend' is just in the last 3.25m.

40mm is way too much. The maximum allowable inclination of a horizontal beam under the current British Standard is (10 + l/500)mm. (l being the span) For a 3250mm beam that would be 16.5mm. Loading the beam will increase the deviation even more.

Thanks for the info. I've passed this along, however given the cost involved to the contractors/steel guys I don't think anyone is going to be replacing it anytime soon. I've had discussions about adding a steel plate in placed to 'square up' the openings to come within the tolerances for the glazing, so I think that's the plan.

Thanks for all of your help again.

 

[theprinceofdarkness]

So is the big beam lifting the side beam? or is it the side beam is lifting the front beam? You say the front beam is drooping but you can see the gap between the support for the side beam. Don't quite see what your problem is. Have you bubbled all the horizontal sections of the beams to see what is pulling on what?
Frank