OTT Structural Engineering?

[Radian]

Trying to be diplomatic here. Our architect has drawn up our living room to have floor to ceiling glazing on two walls meeting at a corner. This is to make the most of a nice view. We understood that there would have to be a vertical steel of some kind to support the floor above (a room-in-roof constructed from timber attic trusses with slate cover @ 50' pitch) but the SE drawings have come back with a massive 200 x 200 I beam rammed right up to the glazing frames. That's before boxing around and finishing in the impossibly tight gap to the glass. On asking if this could be switched for a slimmer alternative, we're simply being told 'no'. Similar constructions I've seen elsewhere utilise 100mm square posts although these may have been single storey. Anyway, a wide variety of sections and wall thicknesses are available out there but the SE won't budge. He's been paid already and TBH we've only just picked this up while studying the drawings.

 

I can't see a way out of this as hiring a different SE we may be back in the same situation only £££'s worse off. It doesn't seem to be any real fault of the architect although they might have warned us. We haven't got the engineering knowledge to call out the SE on the decision and the plans already got building approval for the giant steel. Any thoughts on how to handle this?

 

I mentioned diplomacy at the start of this post because IME these kind of professional can get a bit  cagey when their decisions are brought into question. If not done right, all flexibility can go out of the window.

[Declan52]

I used 100 mm heavy wall box section for mine.  It carries a room above it as well.  Never went with a structural engineer though.  

[ProDave]

Cantilever beans above the windows to support some of the weight and require less of a corner post.

 

Might cost £££ more to enable a thinner corner post.

[scottishjohn]

sounds as though it would simpler to just build a brick pillar or an RSJ and have 2 windows rather than one corner window 

[Construction Channel]

I'm no engineer but I would have thought you could get away with a smaller profile box section like declan. But in all fairness assuming the glazing profile will be approx 75-100mm  and you offset them a bit towards the inside there will Will be minimal interruption o. The internal face. 

[PeterW]

What is the window span..? That will be affecting the steel sizing as the pillar will be taking a lot of the load. A 200x200 UC isn’t that big tbh and you will need that for the top plate anyway just for the end bearing of window lintels or steels. 

 

Do you have any drawings ..?

[Radian]

1 hour ago, PeterW said:

What is the window span..? That will be affecting the steel sizing as the pillar will be taking a lot of the load. A 200x200 UC isn’t that big tbh and you will need that for the top plate anyway just for the end bearing of window lintels or steels. 

 

Do you have any drawings ..?

 

4m x 6m window spans with steels above. The 200x100 UB supports one end of the roof trusses running laterally across this highly simplified diagram: 

Masonry walls (not shown) support the roof trusses at the left of the diagram. That's the essence of the plan.

I don't have the exact dimensions of the window frames but the architect seems to think the 200x200 can be boxed-in and decorated with next to no clearance to the glazing. Also, whatever its finish, it's right on display from the outside.

 

2 hours ago, Construction Channel said:

I'm no engineer but I would have thought you could get away with a smaller profile box section like declan. But in all fairness assuming the glazing profile will be approx 75-100mm  and you offset them a bit towards the inside there will Will be minimal interruption o. The internal face. 

 

Offsetting the steel from the glazing would make more sense to allow for finishing but might not suit the necessary location of the steelwork above. Also, it would still leave us with an ugly column in pole position. I'm sure this sort of application has been engineered before. Only the other day I saw a corner window without a prominent post on the TV show "Ugly House to Lovely House with George Clarke".

[TerryE]

3 hours ago, Radian said:

Our architect has drawn up our living room to have floor to ceiling glazing on two walls meeting at a corner.

 

This is the sort of frill that architects love, but that you will find a total PITA when you actually come to live in the house.  It will add a lot to the cost of the build, the complexity of the structural engineering and the thermal management of the lived-in environment.  Think of a number in the region of £30K+ and ask yourself: is it really worth spending that sort of cost to be able to look out of the corner of the room  -- including whilst lying on the floor or standing on a step ladder.   

Edited January 25, 2019 by TerryE

[PeterW]

@Radian where in the UK are you ..??

 

Those steel sizes are very odd - they don’t match anything in the blue book and the closest sizes I can find I am struggling to really see how they can support the weight of the roof and the wall on the corner. 

 

Assuming that’s a 178x102x19kg UB and a 203x102x23kg then the quick fag packet I’ve just done would mean the 178 lightly loaded over 6m would exceed the l/360 limit for deflection and the 203 would fail deflection due to the roof loads. Neither of those are big steels especially for a 6m span. 

 

I would ask for a copy of the engineers calculations... and his PI insurance ..!

 

Also, what’s the construction method here as you only show a single skin ..??

[the_r_sole]

.

Edited September 26, 2019 by the_r_sole

[Radian]

11 hours ago, PeterW said:

@Radian where in the UK are you ..??

 

Those steel sizes are very odd - they don’t match anything in the blue book and the closest sizes I can find I am struggling to really see how they can support the weight of the roof and the wall on the corner. 

 

Assuming that’s a 178x102x19kg UB and a 203x102x23kg then the quick fag packet I’ve just done would mean the 178 lightly loaded over 6m would exceed the l/360 limit for deflection and the 203 would fail deflection due to the roof loads. Neither of those are big steels especially for a 6m span. 

 

I would ask for a copy of the engineers calculations... and his PI insurance ..!

 

Also, what’s the construction method here as you only show a single skin ..??

 

Just to be clear, above this room there is only a 'Room In Roof' formed by prefabricated attic trusses. It's a Chalet bungalow. I hadn't had any concerns about the lateral beams (up until now) but setting those aside for the moment,  what do you make of the 203x203x60Kg? I only very roughly sketched the layout to show relevant parts. Cavity walls shown solid.

[PeterW]

3 minutes ago, Radian said:

 

Just to be clear, above this room there is only a 'Room In Roof' formed by prefabricated attic trusses. It's a Chalet bungalow. I hadn't had any concerns about the lateral beams (up until now) but setting those aside for the moment,  what do you make of the 203x203x60Kg? I only very roughly sketched the layout to show relevant parts. Cavity walls shown solid.

 

Ok so to be horribly frank that is even worse in terms of loading as you have the bottom chord of the truss carrying the roof, floor and wall loads onto the beam. 

 

To give you an idea of why I’m not sure the calcs are correct, to take a static roof load of 7.5m span across a 4.5m opening, we specified a 254x146x31 and that did not need to take the floor loads. A 178x102 at 6.2m will bend more than its deflection limit just being picked up in the middle so that does not sound right. Given what you have said about the design then there must be a gable wall above the 6m opening - a 178 with masonry loading only is limited to 4m by NHBC, I would be seriously considering asking for a second set of calculations. 

 

Also, if these are the inner walls, what is holding up the outer skin ..?? 

[Mr Punter]

An architect I know is keen on the corner window and he achieves this with a fairly slim circular steel post on the inside which I imagine supports some steel beams above.

 

I would expect the architect to understand how his design was going to work thermally, structurally, aesthetically etc and the engineer calculates loadings, steel sizes connections etc. and supplies you with a drawing and calcs for review.

[Declan52]

Would it not just be much cheaper and easier to divide the long panels up and build a block pier in between.  

So instead of a 6m run you build a single block pier in the middle.  This will give you two openings of 2550mm which is still a massive stretch of glass.  

For the  4m run you could use a 4.2m prestressed concrete lintel which will give you an opening of 3550mm. Still a big window.  

In terms of cost the difference will be huge.  6 prestressed concrete heads will be much much cheaper than all that steel. 

 

[Declan52]

Or if you really want a full 6m opening build the corner in block and use a heavy duty steel catnic or a rsj beam. 

[Radian]

1 hour ago, PeterW said:

 

Ok so to be horribly frank that is even worse in terms of loading as you have the bottom chord of the truss carrying the roof, floor and wall loads onto the beam. 

 

To give you an idea of why I’m not sure the calcs are correct, to take a static roof load of 7.5m span across a 4.5m opening, we specified a 254x146x31 and that did not need to take the floor loads. A 178x102 at 6.2m will bend more than its deflection limit just being picked up in the middle so that does not sound right. Given what you have said about the design then there must be a gable wall above the 6m opening - a 178 with masonry loading only is limited to 4m by NHBC, I would be seriously considering asking for a second set of calculations. 

 

Also, if these are the inner walls, what is holding up the outer skin ..?? 

 

Gable is entirely timber + cladding. Sorry again for the crappy drawing - the cavity wall stubs are drawn as a single block representing their footprint.

[Radian]

1 hour ago, Declan52 said:

Or if you really want a full 6m opening build the corner in block and use a heavy duty steel catnic or a rsj beam. 

 

Having a long hard think about this. Your suggestion of creating masonry piers makes a great deal of sense.

[PeterW]

Do you have a proper wall section as currently a few of us are guessing at build up and can help but only really with some proper detail to work with ... Even a timber gable has a weight to it to be considered, and when you consider a BRegs min wall thickness is 350mm these days, losing a 205mm UC becomes easy. 

[Russell griffiths]

Have you got an elevation drawing, there’s no reason to not have what you want

it just depends if you want to pay for it. 

Edited January 26, 2019 by Russell griffiths

[Radian]

8 hours ago, PeterW said:

Do you have a proper wall section as currently a few of us are guessing at build up and can help but only really with some proper detail to work with ... Even a timber gable has a weight to it to be considered, and when you consider a BRegs min wall thickness is 350mm these days, losing a 205mm UC becomes easy. 

 

Honestly, the only wall sections are on the architects drawings and they have no detail showing how the gable is supposed to be constructed - but this is probably because the attic trusses are yet to be designed by the manufacturer. I only have large format print copies of the drawings so they're not easy to digitise. Here's my best shot at digitising the SE plan that I attempted to boil down into the simplified sketch above:

 

[Radian]

23 hours ago, TerryE said:

 

This is the sort of frill that architects love, but that you will find a total PITA when you actually come to live in the house.  It will add a lot to the cost of the build, the complexity of the structural engineering and the thermal management of the lived-in environment.  Think of a number in the region of £30K+ and ask yourself: is it really worth spending that sort of cost to be able to look out of the corner of the room  -- including whilst lying on the floor or standing on a step ladder.   

 

Hi @TerryE I get what you're saying here. However there's a great deal to be said for the quality of natural  light that results from floor-to-ceiling glazing. This area is to be used as a day-room that will get extensive use all year round. Far better I think than a typical bolt-on conservatory.

BTW, I know you from your generous contributions to NodeMCU/LUA. I've lost count of the number of ESP8266's I've embedded with that firmware into various projects over the last few years.

[mvincentd]

Maybe work backwards from your choice of glazing.  If you want and are prepared to pay for a frameless joint of glass to glass you are going to have to accept a ‘pole’ in that corner, maybe approx 200mm inbound from the glass ( I’ll take a wild guess at a 168 x 6.3 CHS).

If the glass is part of say a sliding door scheme (can’t imagine 6m of fixed panes) what is the manufactures profile.  Mine are  175mm so cladding a 200mm H beam out to say 300mm wouldn’t make a world of difference to ones perception of the view out especially when diluted by the extent of glass you have flanking it.

[PeterW]

44 minutes ago, Radian said:

 

Honestly, the only wall sections are on the architects drawings and they have no detail showing how the gable is supposed to be constructed - but this is probably because the attic trusses are yet to be designed by the manufacturer. I only have large format print copies of the drawings so they're not easy to digitise. Here's my best shot at digitising the SE plan that I attempted to boil down into the simplified sketch above:

 

 

Ok got to say I’m even more confused now ..! This is a single room, and there is a room above, and a standard roof structure on that..? And all of that sits on a single 4m long 203x102 beam..? 

 

What is more bizarre about that design is there are very detailed SHS post detail but nothing about the intersection of the corner post and it’s two intersecting beams ..?? Is there a detail section for the corner shown with the dotted edge box ..? 

 

I would expect that there is a welded plate to the top of the column and end detail for each of the beams and how they are bolted to the column. There is partial end restraint detail on the beams (for example the 203 has a spreader but it doesn’t say what or how it’s attached to the wall) whereas the SHS is very detailed. 

 

Using 2 different profiles for the beams makes it significantly more complex to join at the corner post. What would make more sense is to use a matched pair - one will be oversized but the additional cost steel will be lost in simplification of the fabrication. 

 

If it was me, I would redesign this to use a goalpost setup that is entirely freestanding, and take the roof loads down to the foundations using UC steels. This would mean that there would be a 4m heavy section goalpost with I would expect something like a 254x146x31 UB with a welded plate on the end face drilled for a number of M20 bolts. This would bolt to the end of another 254x146 or 254x102 running across the 6m opening and then have a 146 or 102 UC at the far end to create a fully bolted and secured corner structure. 

 

All of that needs proper design, and tbh I don’t think that the current design is safe as those steels are significantly under sized and there are vital design elements missing. 

 

[Radian]

11 hours ago, PeterW said:

Ok got to say I’m even more confused now ..! This is a single room, and there is a room above, and a standard roof structure on that..? And all of that sits on a single 4m long 203x102 beam..? 

 

What is more bizarre about that design is there are very detailed SHS post detail but nothing about the intersection of the corner post and it’s two intersecting beams ..?? Is there a detail section for the corner shown with the dotted edge box ..? 

 

I apologise for all the confusion! I'm having difficulty digitising the drawings in my present situation but here's the sectional views accompanying the previous plan view I posted:

 

 

[PeterW]

I take it he normally builds car parks .....?? Never seen a 2 CuM pad for a single post.... 

 

There is detail for what is above the 203 - bolted wallplate from the picture - but nothing on the 178..?? I think he’s missed a detail here and as I’ve said before, I can’t see how that will not deflect more than 16mm in the centre which would be l/360. The engineer should have provided this as part of his calculations.