moment connection 1

[LOKSTR]

Any suggestion how to provide a moment connection between two steel beams ( one steel beam sitting on top of other beam )so that it can be analized as grid/grillage structure in computer model.

Looking for both bolted as well as welded options.

will providing a diagonal top and bottom plate( to sandwitch both beams) with 2 holes and connected with threaded bolt work as moment connection or not?

 

[KootK]

Here's one option: Link.

I don't understand the scheme proposed in your last paragraph. Why would the plates be diagonal? If you're going to clamp the beams in some fashion, you'll probably want stiffeners in both beams between the clamp plates.

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

 

[LOKSTR]

The top beam web is perpendicular to bottom beam web.
There are total 2 beams at bottom and 2 beams sit on top perpendicular to bottom beams. Beams are spaced about 5 ft apart. A large column base plate sits on the top beams.
This is a part of steel grillage foundation which will NOT be encased in concrete, just soil.
For grillage analysis, the intersection of top and bottom beam is taken as rigid joint, but in reality, how to make this arragement to act as
a rigid joint?

 

[KootK]

Now that is a different animal altogether. I disagree, grillage analysis does not require a rigid connection between crossing beams. It only requires continuity of each beam through the joint and it sounds as though you have that. You'll probably want stiffeners in both beams at the crossings and perhaps enough torsional restraint that each beam can be considered braced at the crossover. Stiffeners an a few bots should accomplish that.

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

 

[LOKSTR]

KootK,

Forgot to include in the previous post that the beam ends are simply supported( 18 ft span) on steel piles which are about 10-12" above the ground to support bottom 2 beams ends and about 24" above the ground to support top 2 beams.

To have the whole grillage system to act as a 2-way slab, the intersection points need to be rigid.

I think it is reasonable to model both the top and bottom beam as if they are at the same level connected with rigid joints.

 

[KootK]

Thanks for the extra info Lokstr. Always good to have an accurate picture of what's going on.

To have the whole grillage system to act as a 2-way slab, the intersection points need to be rigid.

You'll need to be more specific with regard to what you mean by rigid. Both beams need to be major axis flexurally continuous through the crossing joints. In that sense, yes, rigid. Or, more precisely, partially rigid. That requires no special connection detailing other than what we've discussed above, however.

Full rigidity in the joints also requires coupling between major axis bending in each beam and torsion in each crossing beam. This, you do not need. In fact, it's probably best not to consider it for two reasons:

1) It may lead to ugly, unnecessary torsion detailing in the joints, as you've suggested at the top and;
2) Particularly with wide flange beams, it shouldn't matter much. Wide flange beams don't possess much torsional stuffness.

I think it is reasonable to model both the top and bottom beam as if they are at the same level connected with rigid joints.

Absolutely. At the crossing joints, I would model the beams coming in as having all of their degrees of freedom fixed with one possible exception: torsion. The answer here is that you can model this system as a grillage and nothing exotic needs to be done regarding connections. Win, win!

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