steel connection (beam to beam and beam to column)

[gmd255]

Greetings.

Im designing connections of steel structure and I have some questions for more experienced guys on connections. Structure is anchored to an existing building.

I have designed primary beam to column connection as pinned (connected with 4 bolts). I think thats pretty simple and straightforward.

What I'm not so sure about is:

1) connecting secondary beams to primary beams. Primary beam is at 2 degree angle. Since I am dealing with 'open' sections I have to connect secondary beams (IPE 120) to a web of primary beams (HEA 180) so torsion effect is almost zero). Since angle of primary beam is small, there are little effects of biaxial bending of secondary beams. I know that open sections like IPE 120 are sensible to bending around the weak axis but I have checked this and its OK. I have designed beam to beam connections pinned. I'd like to make connection as shown (2 bolts connecting steel plate welded at the end of IPE 120 to a web of primary beam that is approx. 10 mm thick). Do you see any problem here? How does welded steel plate thickness affect connection? I was thinking about making it 5 mm thick.

2) horizontal stability is achieved by bracing (diagonals). They are placed right bellow secondary beams. I was thinking to fix them to lower flange of primary beams. can that be a problem? since tension forces will be acting at lower part of primary beam?

 

[gmd255]

Any advice/opinion please?

 

[dik]

Pretty conventional framing... sometimes the purlins sit on top of the rigid frames/girders. Check your loading, you may be able to reduce the number of bolts from 4 to 2.

Dik

 

[BAretired]

[li]I don't know if it's a problem, but 5mm seems thin for an end plate, even if it works in bearing. It will tend to warp after welding.[/li]


BA

 

[dik]

Thanks BART... missed that, must have been in the body of the text... typically use 1/2" 12mm minimum... I only looked at the pictures... my oversight.

Dik

 

[gmd255]

I appreciate answers. thank you.

I have been thinking... could I design columns supports as pinned instead of fixed, since I have bracing in roof plane. So any horizontal force will be transfered to existing building through diagonals and fixed supports are really not nedeed. Do you agree?

 

[KootK]

I do agree. And it's probably a better way to go.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[BAretired]

I agree with designing the column as hinged as well, but if you are relying on the existing building to provide horizontal restraint, why do you need diagonal bracing in the new roof? If diagonal bracing is deemed necessary, why cross bracing? It might be better to remove half of the bracing members and connect to the underside of the secondary beams as required to reduce effective length of the brace, thereby allowing each brace to act in tension or compression.

BA

 

[gmd255]

Well its pretty standard in my country that we use X bracing. In calculations we only consider diagonals that are in tension, the one in compression buckles so we do not consider them. Of course I could use only 2 diagonals for whole bracing but in that case a single diagonal should be able to take tension AND also compression forces(no buckling under compression).

Primary beam web thickness is only 6 mm, but since the web will be loaded mostly in shear (torsion as mentioned shouldnt be a problem because of type of connection)I think that shouldnt be an issue...

I've posted this thread mostly to seek some confirmation on my design and thought process.

 

[KootK]

Further to BA's point, you could eliminate two of the braces and still have a system that would be symmetrical and only require tension only bracing. Might simplify your detailing, save a few bucks, and perhaps be a bit more elegant. On the other hand, maybe you're going for a "look".

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[BAretired]

Personally, I would not use any diagonal braces even in the absence of diaphragm action from the roof deck. The existing columns are taking all of the lateral load with or without braces.

BA

 

[KootK]

I don't follow BA. Th existing columns do take the lateral load but you still need something to keep the roof diaphragm from racking in plane. Of course, another approach could be to omit the bracing and call the thing a whole bunch of moment frames. Given the scale, that would surely work.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[BAretired]

Sorry KootK, you are correct. The new structure could rack in plane parallel to the 7.30m dimension. I was thinking about the other direction. So if moment frames are not considered, diagonal bracing would be required to prevent racking and your sketch with the red strikeouts would apply. In that case, a N-S wind load is added to the existing building.



BA

 

[n3jc]

I have to ask? isnt there an issue regarding web of primary beam - I'm talking about the outer primary beams:

Isn't there a danger that a web can bend since its pretty thin? Or can we say that this is not an issue since secondary beams will not rotate enough to cause this? Vertical stiffener would eliminate this issue if there is an issue at all? The connection is pinned after all...

 

[hokie66]

n3jc,
That concern is unfounded, for the reason you gave.

gmd255,
What is the pitch of your bolts? They look very close, with those little sections. I would use two bolts rather than four, as dik suggested. I wouldn't use less than 10 plate for the end plates.

 

[gmd255]

You guys are right - support reaction at the end of secondary beams is around 6 kN. Bolts are loaded only in shear so
I will use 2 bolts (I had 4 M12 (8.8) bolts before), so 2 x M12 (8.8) should do it.

I chose 10 mm end plate welded to secondary beams.