Rafter Joint

[Zambo]

We are constructing a steel framed building with rafter spans of 24m. The steel has been fabricated in the factory by full strength butt welding of 2 x 12m long beams. This fabricated beam has then been delivered to site.

Our Client now claims the beams should not be welded in the middle. I am aware that for RC construction joints should be at 1/3 span or an area of low stress. However for steel construction I am not aware of any such requirement in fact it would seem unusual to me to contemplate cutting beams to introduce additional joints at a location other than midspan.

Am I wrong, is there any code requirement to locate joints away from midspan? Alternatively is there any statement anywhere that this is ok...

 

[UcfSE]

Did you ask the client why, specifically?

 

[valleyboy]

As far as I am aware, if a full penetration butt weld is used in a welded splice, then you can assume that the weld achieves the full moment capacity of the section.

Incidentally, what type of construction is this? Is the rafter simply supported, in which case the maximum bending stress will be at mid span as you state, or a portal or moment frame, when the moment at mid-span will be less under vertical loads?

VB

 

[youngstructural]

I would think that this is a case of the client needing his consultant's specialist training. IE: You need to (VERY politely, and QUITE gently) point out that there is no reason for needing these joints, and explain the actual mechanics of the system.

I am certainly not the most experienced poster here, and I will be quite curious to see what some of these guys have to say, however I cannot see a reason for a joint. Even if you calculate the thrust due to thermal expansion and use it as a load case on your wall (never a bad idea when someone brings something like this up), I would hazard a guess that it's going to be minimal and, therefore, inconsequential.

Keep us up to date!
Regards,

YS

B.Eng (Carleton)
Working in New Zealand, thinking of my snow covered home...

 

[COEngineeer]

young.. i think they joint it because they only have shorter members. 24m (79ft) is pretty long and I am amazed it is shop welded not bolted at the field.

If you beam is like this

______________________________
| | |

Then yeah usually you want to splice the beam at the inflection point (where moment is 0) so all you have to worry about is the shear at the connection. If you splice it at anywhere else (in your case looks like it is only 1 span simply supported), then you just have to make sure the connection is strong enough to the moment I guess. I dont have that much experience either with big beams like this.

 

[RARWOOD]

I believe since you are designing the building that COEngineeer may be right. The contractor may prefer to handle two shorter members instead of one long member. A 12m beam is a lot easier to place than one 24m long. So all the contractor may be asking for is two simple span beams.

If you weren't designing the building the issue might be that the center reaction for a beam continuous over two spans is 10/8wl compared to wl for two simple span beams.

My fab shop is famous for wanting to use one long rafter instead of two. They usually fail to account for the fact that they are now putting more load on the center support beam .

 

[COEngineeer]

I wonder how they transfer 24 m beam (shop welded) to the site? That is pretty darn long. Are you sure its not field bolted? Average flat bed truck in the US is 48 ft long. 24 m is way longer than that.

 

[csd72]

This is obviously not a US thread as it is given in metric (oh how I miss the simplicity of metric!)anyhow, in Australia, for example, you cannot get stock lengths of steel over about 18 to 21m so most portal frames require a splice.

The splice is normally done as a bolted connection for transport and to avoid site welding. It is even better structurally if it is fully welded (butt weld in the flanges, fillet weld in the web).

As the structural engineer, you need to point out to the client that this is a perfectly acceptable way of joining two pieces of steel together. Point out that some welded beams are actually formed by welding plates together into an I shape.

Regards
csd

 

[haynewp]

You can get a permit for longer flat beds.

 

[Zambo]

thank you all for your responses. I am in Thailand and a 24m long trailer can go on the road with some assistance from the police.

The beam is fabricated from 2 x 12m beams. They are welded with a full penetration butt weld at flanges and web. An additional splice plate is added with fillet welds to each side of the web splice (I don't think this is needed).

The rafters (beams) are bolted to the columns (which to clarify a couple of points mentioned in the responses are 24m apart). These beams are the roof rafters so the loads are deadload, service load plus a small allowance for people working on the roof.

The Client thinks that the beams should not be welded at midspan and that they should be welded at a point of lower moment. My point is that as the joints are equivalent to the strength of the parent material it would be a bit ridiculous to cut the 12m means to weld them together again to make the 24m beams from 3 pieces.

 

[xxpegasus11]

I see no reason why a splice that develops the full capacity of the section could not be made as you describe. Yes one tries to avoid them at midspan but they are often made for practical reasons. I imagine the Clients concern would be more on quality control of welding rather than the splice location but presumably one could demonstrate that its properly done.

 

[apsix]

Another point I don't think has been raised yet; for such a long span rafter it is probable that the size is controlled by serviceability issues (deflection), the full section moment capacity is unlikely to be utilised. If that is the case then even IF the butt weld was less than 100% efficient strength would still be adequate.

 

[RARWOOD]

Zambo

With full penetration welds done in the shop you should have no trouble.

Just as a side note, in the US, with special permits the company I work for has shipped glulam beams 105' long. Regulations very from state to state. Here in Minnesota some one was able to ship a 250' long industrial dryer.

The big reason for limiting length in most cases is cost. When the load reaches a certain point it requires police escort and there are restrictions on when the load can be moved. It might cost a contractor more to move a 100' beam 20 miles than to move a 75" beam 500 miles.

 

[csd72]

Zambo,

Just tell them 'may pen rai'(it doesnt matter)!

sawasdee (goodbye)

csd

 

[Zambo]

csd

khawp khun khrap. phom ja bok kow....

 

[csd72]

I dont speak that much thai.

I have been there a couple of times, and picked up a few terms.