Truss chord connections 1

[Ipetu]

Truss top and bottom chords are WT sections. Truss web members connect to the WT webs. Is there any code restriction of connecting the truss web members directly to the WT web without using gusset plates? In this case the web members are double angles. If it is okay to connect directly to the WT webs are there any specific checks that are required.

A scan of the truss drawing is attached.

Thanks very much.

 

[nutte]

Yes, there are many specific checks that need to be done.

 

[ishvaaag]

Maybe the biggest problem is that most are expecting there gusset plates, a solution that has evolved from past histories of success and for which some decanted literature have filtered to the codes.

So you need to check for all the issues that are dealt with gusset plate connections, some may be alleviated, but you might to also consider specifics of the situation not present tor the alternative gusset plates, even when this may seem unlikely; and not being within the expectancies of codes and texts, you may end having to imagine the correct answer to your derived questions.

 

[jike]

We have often used deeper WT chords so as to eliminate all gusset plates.

 

[slickdeals]

Look at the following design example in the AISC 13th Edition CD
Example II.C-4 Truss Support Connection

 

[Stillerz]

You absolutely do not have to have gussets.
If the geometry of the connections b/t the web members and the top and bot chord allows for enough room for all members and clearances for welds and enough material to satisfy the connection desing than why add gussets?

 

[BAretired]

I would have expected the web members to be shop welded directly to each side of the WT webs. The use of a W530 instead of a WT 265 seems particularly uneconomical and cumbersome.

BA

 

[Stillerz]

ah yes....i too was speaking as if the truss was all welded.

 

[Stillerz]

i guess that would one hell of a long truss to have all shop welded....hard to ship.

 

[BAretired]

Stillerz,

I don't think transportation is too difficult, but there is no reason why they cannot use a bolted splice at midspan if they wish to reduce the length for trucking.

BA

 

[Stillerz]

it looks to be ...what..about 125' long? That would take some serious permitting here in the states.
i just had some VERY simlilar trusses built last summer that were 80' long and those took special permits.

I agree...shop weld and splice the two halves...63' is just over a normal 55' truck.

 

[Ipetu]

BAretired, Stillerz

There will be a couple of bolted splices. The fabricator wants to use bolted connections only. He doesn't want any welding. He would rather cut a W530 into a WT265 instead of welding on gusset plates etc.

For the bolted splice connections can one use 2 continuous plates instead of one continuous L to bridge over the butting truss chords?

Ipetu

 

[Mixtli]

Ipetu,

You can do a number of solutions, as long as you comply with the specs, i.e.min edge distance and spacing for bolds, ultimate limit states for connecting elements and bolts, ultimate limit states for connecting and supporting members, etc. I would suggest to follow the recommendations of the AISC manual 13 (the one that includes LRFD and ASD) in the chapter regarding connections.

Greetings

 

[BAretired]

Anything is possible, I suppose, but why does the fabricator not want to use welding? Doesn't he have any welders in his shop?

The type of splice I have used in the past for large trusses is end plates welded to the chords and bolted together in the field. This puts the bolts in the tension chord into tension. For the compression chord, unless you have a stress reversal, the bolts are not stressed at all.

BA

 

[Ipetu]

Thanks to all. Your help is really appreciated.

Ipetu

 

[ChipB]

If it were me, I would use two (2) splices. I would make the two end sections 35' long each, and the middle section, 55' long.

I would stay far, far, far away from splicing in the middle where the tension/compression forces are the greatest.

Just my 2 cents

 

[BAretired]

I prefer one splice at midspan, but I would probably use a different arrangement of web members, probably two diagonals per bay instead of three. This would reduce the number of connections, resulting in cost savings. As stated earlier, all shop connections would be welded. Field connections would be bolted.

As for staying far, far away from midspan, the benefit is illusory. Assuming 35', 55', 35' truss sections, the moment at splice points is 81% of the midspan moment. Considering the reduction in truss depth, the tension in the bottom chord at splice points is 86% of the midspan tension. Compression in the top chord is of no importance as it does not have to be carried by bolts. Thus, two splices requires 72% more bolts than one splice.

BA

 

[Stillerz]

BA-
Can you really just rearrange the diagonals independent of where the truss is being loaded?
I mean, i can only assume that the diags and verticals are there because that is where the roof purlins land.

 

[BAretired]

Stillerz,

You are quite right. Panel points of the truss should correspond with purlin spacing. However, it would be preferable to have the center two diagonals meeting at the top chord instead of the bottom chord to avoid a vertical member at the splice.

BA

 

[BAretired]

Looking again at the truss elevation, there is a seat above the top chord at every gridline, i.e. at 6.25m o/c. This suggests that purlins or secondary trusses are spaced at 6.25m o/c. If that is the case, then two diagonals between each gridline would still be the more economical choice.

BA