truss connection 2

[structSU10]

the steel fabricator wants to use the attached detail for the gusset plate connections of a large truss. The web member loads range in magnitude from 60K to 550K, and the chord forces max out at 1700K. Something about this set up has me worried and wanted to see others thoughts on this. The chord flanges are 1.75" thick as well, and they are coming up with 1/2" gussets to full pen weld to them. Assuming they design it properly, taking into account offsets of the connection from the work point, are there other concerns for this connection? My gut says no to this, but maybe I am over-reacting.

 

[msquared48]

Bothers me too. I would run the gusset over the outside of the top chord member to within 1/2" of the edge of the flange and weld the gusset all around. This would be much better for any torsional or moment forces to the connection. I would also consider adding some additional bolts to the top chord for gusset plate support and architectural appearance. Why Architectural you may ask? Well, I do have some taste... Not much, just some.

Mike McCann, PE, SE (WA)

 

[rb1957]

looks pretty wimpy to me too.

as with mike, overlap the gusset and the chord flange, outer face is more obvious, but the inner works if they want a flush outside; use packers if needed. some more bolts would be a nice touch.

Quando Omni Flunkus Moritati

 

[BAretired]

It is going to be difficult to fit the pieces together in the field. Is the vertical member a column with top chord continuous over it?

A sketch of the truss and its supports might help.

BA

 

[KootK]

It's unusual but I don't see anything wrong with it from a capacity standpoint. I'm surprised that the fabricator wants to do it this way for two reasons:

1) Member fit up as BA mentioned.

2) I'd think that it would be difficult to align the gusset plates to be nearly vertical.

The greatest trick that bond stress ever pulled was convincing the world it didn't exist.

 

[JAE]

I would suggest you send this to the AISC Steel Solutions Center.
From the AISC website:

For people who need technical assistance, innovative solutions, or tools to make structural steel design even easier, the Steel Solutions Center is your #1 source of information for structural steel.
Contact the Steel Solutions Center at 866.ASK.AISC
or solutions@aisc.org


Check out Eng-Tips Forum's Policies here:
faq731-376

 

[hokie66]

My main concern would be fitup. The web members are presumably of different depths, so tight control is required, and would not work if different webs occur at the same joint. The lapped gusset with fillet weld solution doesn't solve the fitup problem either..

 

[structSU10]

Thanks for the replies. I will try the solutions center. The thing I was concerned with is that you may have to consider a moment being dumped into the flange of the chord in some cases. They plan to shop fabricate 3 sections of the truss, and field splice them, - and they plan to leave a 1/8" gap ( I think my snapshot got cut off somehow).

 

[SteelPE]

Now, I understand that there would be moments required in that connection, but how would moments end up in the chord if the truss is concentric? I am confused.

Also, how did you expect them to fabricate the truss/connection? I would have expected the connection to have plates that covered the flanges of the chord, similar to those indicated by msquared. Is that what is shown on your drawings?

 

[jike]

The more typical detail of this connection is to use gussets lapped with the chord flanges. Fillers should be used to provide allowance for fitup. I usually require the fabricator to assemble truss in the shop and disassemble before shipping to avoid fitup problems in the field.

Caution: Avoid welding on jumbo sections in tension to avoid lamellar tearing. I cannot remember the (Jumbo sections) Group Number, since I am retired and gave away all my books.

 

[WinelandV]

If filler plates are used, be sure to adjust the bolt shear capacity per Section J5 (13th Ed.), assuming the truss was designed using AISC.

 

[KootK]

@ StructSU10: I wouldn't sweat moment transfer into the truss chord. All of the connection strategies discussed so far would be more or less equivalent in that regard. As long as your members are concentric, the secondary moments should be of little concern.

@ Jike: Can you elaborate upon your jumbo section lamellar tearing comment? I've not heard of that before. Beyond a certain thickness of material, does cross grain tearing become an issue in all directions?

The greatest trick that bond stress ever pulled was convincing the world it didn't exist.

 

[structSU10]

Yes, We show the standard connection as gussets bolted to the flange of the chord.

The moment would be generated from the free body diagram of the gusset - at least in a 2 web member configuration, at the interface of the gusset to the flange there is a moment since that plane is offset from the work point. At least that's what I see when I look at it kinda like the uniform force method. I am not well versed in that, so maybe I am looking at it wrong.

jike - is there a reference you have for the lamellar tearing?

 

[KootK]

Ah... now I see where you're going with the moment concern structSU10. In this instance, I think that the proposed connection's ability to convincingly transfer moment from the gusset plates into the chord flanges is actually beneficial. Firstly, the moment should be easy to deal with, both in the plate and in the connecting weld. Secondly, that moment applied to the chord flange / plate edge is necessary to keep there from being moment in the chord beyond the joint.

The greatest trick that bond stress ever pulled was convincing the world it didn't exist.

 

[JAE]

Lamellar tearing wouldn't be a problem with the detail you have.
The issue with Lamellar tearing is with the weld detail and if you create constraint in the weld.

The AISC manual has some examples of bad details (see section on welds - Part 8 in the Thirteenth Edition).
Your detail is a plate end to plate end detail so this shouldn't be an issue.
Also - it is usually only an issue with plate thicknesses larger than perhaps 1 1/4".



Check out Eng-Tips Forum's Policies here:
faq731-376

 

[BAretired]

If this is a splice point in the truss, you could splice the top chord with a couple of end plates at the center of your detail, relying primarily on bearing to transfer the 1700^K compression, using end plates only to hold the top chord members in line and resist shear.

The two diagonals could be shop welded instead of field bolted. If it is a parallel chord truss, the vertical member carries no load and could either be omitted or attached with nominal bolting.

The bottom chord splice must be designed for 1700^K tension and is best located to miss web members.

BA

 

[nutte]

Looks pretty clean to me. There will be eccentricity on the gusset at the CJP weld, so calculate it and size the gusset accordingly. (This is no different than if the truss chord were turned web-vertical, with the gusset welding to the flange of the chord. That weld would also have moment on it, and would be sized for such.) Another benefit to this configuration is that if there are any members framing into the flange of the truss chord at this panel point, they can connect directly to the flange. If they were to connect to the gusset, those forces would have to be resolved in the attachment (bolts or welds) of the gusset to the truss chord flange.

I don’t know what size the top chord is, but the flanges look pretty thick. Punching or drilling holes through it might be difficult.

And as has been mentioned, the fit-up challenges are the same if the gusset is welded as shown or if it’s connected across the flanges of the top chord.