×
INTELLIGENT WORK FORUMS
FOR ENGINEERING PROFESSIONALS

Log In

Come Join Us!

Are you an
Engineering professional?
Join Eng-Tips Forums!
  • Talk With Other Members
  • Be Notified Of Responses
    To Your Posts
  • Keyword Search
  • One-Click Access To Your
    Favorite Forums
  • Automated Signatures
    On Your Posts
  • Best Of All, It's Free!
  • Students Click Here

*Eng-Tips's functionality depends on members receiving e-mail. By joining you are opting in to receive e-mail.

Posting Guidelines

Promoting, selling, recruiting, coursework and thesis posting is forbidden.

Students Click Here

Jobs

Modeling Truss Panel Points

Modeling Truss Panel Points

Modeling Truss Panel Points

(OP)
Folks,

I am modeling a deep floor truss that will carry floor beams at each panel point and one in between each panel point. The panel points are at 16' on center and the beams are 8' on center.

I am modeling the truss with moment releases (minor and major) on the diagonal web members. I am unsure about the releases at the top and bottom chords.

At the panel point where the diagonal frames in, should I release the moments as well? Because the top chord of the truss will have studs and concrete will be placed on it, wont the top chord act as a continuous beam supported on panel points?

The bottom chord will act as a simple beam spanning between panel points.

Are these assumptions right? I would appreciate your suggestions.

RE: Modeling Truss Panel Points

If the truss top and bottom chords are continuous, unbroken members, then, where the bottom and top chords intersect the panel points, they need to be fixed to properly model the truss for member stresses and deflections, especially since you have intermediate loading between the panel points.  

As for any diagonal and vertical members, they are usually pinned at the ends.  

If the top and bottom chords have splice points, then these locations need to be treated as pinned joints too.  Just remember though that at any panel point, one joint must be fixed to avoid spiining in space, ultimately resulting in no solution.  The fact that one of the member ends is fixed though, with all the rest pinned at the same panel point, will still give a solution as if all the members are pinned at the joint.  Such is the typically the condition at the ridge of an open web truss.

Mike McCann
MMC Engineering

RE: Modeling Truss Panel Points

(OP)
Mike,
Thanks for that response.

RE: Modeling Truss Panel Points

You can model the truss with either continuous top and bottom chords or pinned top and bottom chords.  Either is statically admissable.  Utilizing the continuity and designing for the moments in the chords will generally result in lighter sections and certainly less deflections within your analysis.  Traditional true "truss" behaviour assumes pure axial behaviour (pinned top and bottom chords).  If you design your members with the results from such an analysis, the members will just have to form plastic hinges to achieve that (still stable) state of equilibrium.   
 

RE: Modeling Truss Panel Points

(OP)
WillisV,

I have a condition where I am using WF members (web horizontal). I am getting high moments due to intermediate panel point loads, even though I have a vertical there.

I am unsure whether to model one side of the top chord pinned (like Mike suggested) or to design for that moment.

Any ideas?

RE: Modeling Truss Panel Points

slickdeals,

The key to modeling structural systems in a program is to approach it with the goal of replicating the actual configuration and composition of the system as close to reality as possible.  

What Mike was discussing had to do with how typical structural programs deal with global stability in the matrix (i.e. a node must have one member not-pinned).

If you add a pin, then you must account for the presence of that pin in your design - by detailing the connection not to take moment.  

If you have a vertical in the truss at the point load, why are you getting "very high moments" there if the vertical is also connected to a diagonal?  It seems to me that a proper lay-out of truss would negate any significant chord moements.

 

RE: Modeling Truss Panel Points

(OP)
JAE,
I totally agree with you in that the idea of modeling is to replicate the exact situation.

Please see the attached ETABS file. It is a simple truss with panel point loads. I have a vertical for loads in between panel points.

In Elevation 2, the truss top chord and bottom chord are modeled continuous. I still get moments in the chord and it looks like the chord is spanning between the diagonals and not between the diagonal and intermediate vertical.

Any help with the model is appreciated.

Thanks

RE: Modeling Truss Panel Points

slickdeals - can't open your file - can you post a pdf of it?

RE: Modeling Truss Panel Points

I can´t see your model, but probably the vertical element is just acting as a tie that "shares" the point load between the top and bottom chords.  If there are no diagonals forming a triangle, it may happen. It may also have to do with the relative stiffness of the top and bottom chords.  Are they similar in size?
Can you post a pdf or jpg figure?

RE: Modeling Truss Panel Points

The beauty of a truss is that it's members take only axial load and concentrated loading is at the connections only.  Doing so ensures that you will always have a relative high ratio of axial load to bending.

Why would you want to add significant loads between panel points?  THis suggests to me that the panels need to be closer together or the loads moved out to the panels.....

Regards,
Qshake
pipe
Eng-Tips Forums:Real Solutions for Real Problems Really Quick.
 

RE: Modeling Truss Panel Points

(OP)
QShake:My truss is 8' deep and my panel points are spaced 16' apart for a 45 degree angle. That forces me to have a load in between panel points (beam spaced at 8' o.c.)

PanamaStr: I think you are correct. I have members significantly bigger than they need to be for the chords.

RE: Modeling Truss Panel Points

Slickdeals:

What I am hearing you say is that you have no interior vertical members, only diagonal members.  Not the typical Pratt or warren Trusses.  If you can add the verticals too, then the moments you mentioned will go away, or at least become extremely manageable.   

Mike McCann
MMC Engineering

RE: Modeling Truss Panel Points

If your truss is 8' deep and the diagonals are sloped at 45 degrees, your panel points are at 8' centres horizontally, not 16', but perhaps not at both top and bottom chord.  If you can't post a file we can read, maybe you can describe the truss generically.  Pratt, Howe, Warren, modified Warren.  In your third post, you said that you have verticals at the intermediate load.  Sounds like modified Warren, which would, as Mike said, deal with the moments.

RE: Modeling Truss Panel Points

I also think you should model the top and bottom chords as they would be in real life,--if continuous then model as continuous, if not, then don't. The continuity of the chord should affect the allowable axial buckling load of the chord (think of a column that is continuous between floors rather than individual separate column segments between floors).

RE: Modeling Truss Panel Points

Based on your description, I think you should consider a warren truss with panel points 8 ft oc in both chords. Good luck.

RE: Modeling Truss Panel Points

(OP)
I think I may have not explained it right. I have a 8' deep truss. The intersection of the diagonals are 16' apart at the top and bottom chords. I have a vertical member at 16' on center (or in between diagonals at the top chord).

I am trying to eliminate the bending in the top chord by using the vertical member, however, I am still getting some bending. Are there ways to minimize this?


My truss profile looks so, I am just being lazy to sketch and scan. It's a Sunday morning :)
-------------
\  |  /\  |
 \ | /  \ |     and so on....
  \|/    \|
   ----------

 

RE: Modeling Truss Panel Points

Have you considered sub-dividing the diagonals and lower chord member to provide additional panel points?  

As noted above, individual truss members shouldn't be subject to bending or you wind up with problems such as you're having....

Regards,
Qshake
pipe
Eng-Tips Forums:Real Solutions for Real Problems Really Quick.
 

RE: Modeling Truss Panel Points

I would first try to model the top chord as a composite (with concrete slab) continuous beam with intermittent diagonals and verticals pinned to it.  

RE: Modeling Truss Panel Points

slickdeals,

What you have drawn is a modified Warren truss.  There should be negligible moment in the chords if the loading is all at the panel points in the top chords, including over the verticals.  If you have substantial bending of the top chord, your model is incorrect.  Suggest you analyze it by hand using method of joints or sections, or graphically if you know how.  Compare results with your computer output, and then figure where the difference is.

RE: Modeling Truss Panel Points

slickdeals,

Forgot to say, first thing to look at is the force in the verticals.  Compression in each vertical should match load at that panel point.

RE: Modeling Truss Panel Points

(OP)
"If you design your members with the results from such an analysis (using all pinned connections), the members will just have to form plastic hinges to achieve that (still stable) state of equilibrium."

I have a question regarding this. Assuming that I model the truss chords continuous, then there will be some moment induced at the panel point if there is a difference in T/C forces at the joint (there will be a vertical component to the T/C force).

However, due to the continuity, there will be some reduction in the axial force of the chord. However, will the P-M interaction cause a higher stress ratio in the member as opposed to one with a slightly higher axial force but no moment?

Your input is welcome.

Thanks

RE: Modeling Truss Panel Points

(OP)
"Secondary Stresses in Trusses" - By Shankar Nair published by AISC gives me insight into what I was looking for.

The flexural stresses induced from continuous chords may be neglected if the axial forces from analysis were consistent with a pinned joint assumption.

RE: Modeling Truss Panel Points

Yes, but even AASHTO has a limit at which secondary stresses must be considered and that limit is low, 4000 psi.

Regards,
Qshake
pipe
Eng-Tips Forums:Real Solutions for Real Problems Really Quick.
 

RE: Modeling Truss Panel Points

(OP)
Folks,
Please see attached a crude sketch of my gusset plate. I wanted to know if the bolt group attaching the gusset to the chord will need to be checked for eccentric shear in addition to the axial force due to difference in chord forces?

Is my eccentricity drawn correct? The moment in my gusset plate is equal to the horizontal shear force * distance from WP to last line of bolts. Isn't this the same moment that will be induced in the bolt group due to eccentric shear?

I am breaking my head with this and I would appreciate your inputs on this. I don't want to miss any checks on the gusset plate.

Any references or design check examples will also help.

Thanks

RE: Modeling Truss Panel Points

The gusset plate you show in your sketch has no eccentricity because all members meet at a point.  The compression diagonal on the left and the tension diagonal on the right intersect at the centroid of the top chord.  

If all bolts are centered on the c.g.of each member meeting at the joint, the bolts carry no eccentric moment.

 

RE: Modeling Truss Panel Points

(OP)
@BAretired,

I agree that there won't be any moments in the members, but is it also true of the gusset plate?

Maybe I am complicating it way too much.

The way I see it is that all the force from the braces have been transferred into the gusset at the last line of the bolts. At the last line of bolts I have a tension and compression. When resolved into components I have a horizontal and vertical shear.

The horizontal shear is at a distance of "e" from the centroid of the chords. The shear has to be transferred into the chords as an axial force via shears/moments/axial force in the gusset.

Similarly the bolt group in the chord is subject to an eccentric shear (or axial force) and not a direct axial force.

Are my fundamentals completely wrong? Can someone please explain it to me with a sketch of how the loads are being transferred from the brace to the gusset and then to the chord?

Thanks

RE: Modeling Truss Panel Points

There are five forces acting on the gusset plate, two from the top chord, two from the diagonals and one from the applied load which you have not shown.  For equilibrium, the sum of these vectors must be zero.  If they all intersect at a point, there can be no eccentricity on the gusset plate as a whole.  If there was, the plate would not be in equilibrium.

In the case of the left diagonal (compression), the clockwise moment caused by the horizontal component of force is precisely balanced by the counterclockwise moment of the vertical component.  A similar argument applies to the tension diagonal on the right.  Moment from horizontal component is balanced by an opposite moment from the vertical component.  This is true only if the resultant force is directed through the common point, your work point.

The actual state of stress at any point within the gusset plate is not so easy to predict however.  You would need to perform a finite element analysis on the plate to find how the stresses vary.

The gusset plates (one each side of the joint) must be adequate to transfer the horizontal shear from the two diagonal members to the chord member, but the moment is zero.  

Hope that helps in your understanding.

RE: Modeling Truss Panel Points

I think I got my clockwise and counterclockwise reversed in the above.  Sorry about that.

RE: Modeling Truss Panel Points

(OP)
@BA,
I agree that there are no moments at the workpoint. But how does the force get into the chords from the gusset after the force is transferred into the gusset from the diagonal. I mean the force from the diagonal is not transferred into the gusset at the workpoint, but at some distance away from it. The eccentric force can't all be transferred in shear.

Incidentally, I found this during my web searches.

http://ntsb.gov/Recs/letters/2008/H08_1_Design_Adequacy_Report.pdf

Folks, thanks a bunch for helping me get to the bottom of this. I may not be the smartest, but I don't want to leave things to chance. Your help is appreciated.

RE: Modeling Truss Panel Points

I have not read your reference, but at a glance it appears that the diagonals do not intersect at the work point.  Sometimes it is necessary to spread members out a bit in order to make room for the connection.  When this happens, there is an eccentricity applied to the joint.  

If the two diagonals in your sketch had intersected 'e' above the centroid of the top chord instead of at the work point, then a moment would be applied to the joint.  That moment would be the sum of the horizontal components of the diagonals times 'e'.  It would be distributed among the four intersecting members according to their stiffness and the applied moment to the gusset plate would still be zero.  The bolts connecting each diagonal would be carrying axial load plus a normal force to resist the moment tributary to that member.  

RE: Modeling Truss Panel Points

I started looking through the Interim Report you referenced earlier.  What is being done in the report is not entirely clear to me at the moment.  I think I would like to reconsider my former response but will need a little time to do it.

RE: Modeling Truss Panel Points

You shouldn't have to worry about moment in the gusset for your case.  If you were coming into the flange of a chord and the workpoint was at the centroid of the WF then I would say you do, but the difference between where your gusset meets the chord and the workpoint is only tw/2.  To be correct, you should make sure the weld from the gusset to the chord can take that moment out.

RE: Modeling Truss Panel Points

(OP)
StrEIT:
I agree with you in that I would need to check for moments if I were coming to the flange.

However, I still feel that the axial load has been transferred into the gusset at the last line of bolts from the diagonal to the gusset. The vertical and horizontal components of this axial load will have to get transferred to the centroid of the chords. Are you saying that this will happen with only shears and no moments?

RE: Modeling Truss Panel Points

Joint U10 of the Interim Report, contrary to my earlier post seems to have all members meeting at a point.  The writer calculates for Section A-A that the shear, V in the gusset is 2,723k.  This is the sum of the horizontal components of the two diagonals. It is also the algebraic sum of the forces from the top chord.  That horizontal shear is constant between the work point and the end of the diagonals.

The moment at the intersection of the two diagonals is 0, but Section A-A (the upper end of the diagonals) is 14" below the work point which gives a moment of 38,118k" at that section.  That is an internal moment within the plate.

Returning to your sketch, you show a shear force in the gusset which would be the sum of the horizontal components of the diagonal members.  

You show a curved arrow which suggests to me that the plate has a moment acting on it.  It does not.  It has a horizontal force acting at the work point.  At the end of the diagonals, that horizontal force has a moment of H*e where e is the vertical distance from the work point to the last line of bolts in the diagonals.

The force from a diagonal to the plate is an axial load.  The four bolts connecting a diagonal to the plate are acting in shear parallel to the direction of the diagonal.

The bolts connecting the top chord to the gusset plate are also transferring an axial load so they will not be subjected to eccentric shear.

RE: Modeling Truss Panel Points

(OP)
@BA,
Yes, you are right. The plate has an internal moment (which it should be checked for) that it should be able to transmit, in addition to the axial force and shear.

Thanks

Red Flag This Post

Please let us know here why this post is inappropriate. Reasons such as off-topic, duplicates, flames, illegal, vulgar, or students posting their homework.

Red Flag Submitted

Thank you for helping keep Eng-Tips Forums free from inappropriate posts.
The Eng-Tips staff will check this out and take appropriate action.

Reply To This Thread

Posting in the Eng-Tips forums is a member-only feature.

Click Here to join Eng-Tips and talk with other members!


Resources