Beam Splice Model

[dgkhan]

Please see attach sketch. One splice in one span works fine. But when I placed two splices, left most span, both sap and staad show instability. But it is a real life situation and a evry senior engineer is doing it manually. I just wanted to cross check and learn using s/w. Any tips ?

 

[JAE]

The first pin on the far left (just to the right of the far left support) should not be there. You can't have a pin like that and remain stable.

Also, it appears on the far right spans that you have double pins (two small circles side by side). A joint must have at least one non-pinned member framing into it.

 

[hokie66]

JAE is correct. I hope it is not "a real life situation".

 

[BAretired]

But it is a real life situation and a evry senior engineer is doing it manually. I just wanted to cross check and learn using s/w. Any tips ?

What is s/w? If senior engineers are doing this, they ought not to be senior engineers.

BA

 

[jberg]

I believe that s/w is software. At least that is what a few other engineers at my office call it.

Joel Berg

 

[BAretired]

I assumed as much. Is it too much trouble to write the word software, particularly when you are expecting members of the forum to provide thoughtful answers to your question?

If all supports are truly hinged, then the structure is stable, but I seriously doubt that every senior engineer is "doing it".

My tip to you is "don't do it".



BA

 

[Lion06]

It's not stable. The span on the left is the issue. It has a mechanism in it. If the support farthest to the left was fixed instead of pinned then it would be stable, but as it stands, it isn't stable. It shouldn't take software to see that.

If you draw a FBD of each individual member you will see that the two left-most members are like a simple span beam with a hinge along the span - again, it's a mechanism.

 

[Tomfh]

Spans 2 and 4 are also problematic, not just span 1.

 

[hokie66]

The thing is...with a four span cantilever and drop-in system, you can only have three hinges and remain stable. The sketch shows six, although in two cases they are sort of double hinges, which makes no sense. Perhaps the OP can advise what he really means.

 

[msquared48]

I agree with Hokie here. This setup makes no sense in the real world. It is inherently unstable three times over.

Mike McCann
MMC Engineering

 

[csd72]

dgkhan,

I agree with what the previous posts have said.

Only one of those thre spans is truly stable.

In real life, a system of members with higes at each end is called a 'chain' great for tension but not much good as a beam. Three of your spans contain a chain link.

It is a worry that neither you or your senior engineer picked this up.

 

[racookpe1978]

Could his "splices" be coded wrong?

Assume his beam was a W12.

Each approved splice (what we see as a single "pair" of bolt holes in the software) could be actually a pair of 10x6x3/8 thick plates with 6 bolts through each web, making a sandwhich joining the two beams.

 

[BAretired]

If all supports are truly hinged as shown, the member can have any number of hinges throughout its length and exhibit a form of "stability" similar to that of the flat arch in the recent thread "A Tricky Question" by the same OP. The members would be primarily tension members, not bending members. They would need to stretch and deflect enough to enable them to carry the tension.

If all but one of the supports is changed to a roller, more of a real world situation, then I agree that the structure is unstable.


BA

 

[msquared48]

True, but in the real world, what would be the point of setting up this configuration for a tension only situation? This scenario does not seem likely, let alone feasible to me.

Mike McCann
MMC Engineering

 

[dgkhan]

I have attached a more clear sketch. My question is how to model splice in S/W (software).
Question 1. I think both beams meeting at splice release the moment therefore I showed two circles of moment release at splice location?
Question 2. When there is two splices in one span than S/w always shows instability? any way to overcome it

 

[BAretired]

1. It is just a diagram. We were not clear about what you meant with the two circles close together.

2. There can be two splices in one span provided the adjacent spans continue through to the next support. That is called a drop-in span.

You cannot have two splices in an end span, so remove the splice near the leftmost support.

If you wish, you can move the next splice to the right of the second support. You will have two splices in the second span and the structure will be stable.

BA

 

[oneintheeye]

can he have a moment splice and then not model it?

 

[racookpe1978]

Well, that was the intent of my question: He HAS a moment splice (or several moment splices), and HAS modeled it (them) incorrectly.


So, back to the orignal question. Draw - not in your modelling siftware, but on a piece of paper - then scan it - exactly HOW the splice is constructed. Not how it is modelled. How it is built.

 

[msquared48]

Is the support at the left a wall of another building with, perhaps, the left most beam actually continuing further to the left to another support? If so, then this span would be stable.

Mike McCann
MMC Engineering

 

[dgkhan]

I do not know the splice as-built details. It is existing structure. Ignoring splice, beams start to fail in moment if treated as simple supported. Obviously, it should not be the case. So by looking closely at vague drawings, I figured out splices. In my understanding, purpose of this mechanism is to reduce moment and make this beams work. Left most span is not the last one but it continues. Probably I will increase my model and see the effect. I can do it manually but will loose money on it. It is a large frame.400 x 400 feet. So wanted to use s/w(soft ware) to run down time. Thanks for all the help.