Integrated Buildings & Structures

The trusses are designed theoretically by assuming that joints are pin jointed. But practically the joints in trusses are either welded or bolted. How is it justified??

------------------------------
Binol Varghese Aff.M.ASCE
Assistant Professor
Muthoot Institute of Technology & Science
Ernakulam
------------------------------

The key is the arrangement of members in a triangular pattern in trusses and not how the joints are actually created. Since a triangle in two dimensions is a rigid geometric pattern, there is very little bending in members unless a load is directly applied to a member. You can easily verify this using a computer program. Model a truss using beam/frame elements and then using truss elements. You'll see very small bending moments in the frame model.

------------------------------
M Asghar Bhatti P.E., F.ASCE
Professor
UNIVERSITY OF IOWA
Iowa City IA
------------------------------

If the projected centerlines of the web members and their connections coincide with the centerline of the chords, there is no eccentricity. Thus, they will behave as pinned. This assumes the chords are loaded at the panel points/joints, not continuously, or that they are evenly loaded on either side of the joint. 

Also, the further apart you place the chords of the truss (the longer the web members), the more the moment in the connections reduce, even if there are eccentricities. At some point moments become negligible (act as pinned). 

There are second order effects when the truss deflects that can cause the connection to deviate from purely pinned. Again less pronounced the longer the web members are. 

This justifies using pinned connections for the majority of trusses that will not have large deflections or large, unevenly placed loads. 

------------------------------
Jeffrey Walkley P.E., S.E., M.ASCE
Vice President
Michael J Walkley, PA
Towson MD
------------------------------

Justified because truss behavior is stiffer than bending by an order of magnitude. Test this by modeling a hypothetical truss first pinning all the joints, then fixing them.

Loads must only be applied to the joints. No redundant supports, meaning pin one end and roller support opposite end.

Behavior should be almost identical in both the pinned and fixed scenarios.

Jonathan S Price, PE


Original Message:
Sent: 8/23/2023 11:20:00 AM
From: Jeffrey Walkley
Subject: RE: Moment Transfer in Trusses

If the projected centerlines of the web members and their connections coincide with the centerline of the chords, there is no eccentricity. Thus, they will behave as pinned. This assumes the chords are loaded at the panel points/joints, not continuously, or that they are evenly loaded on either side of the joint. 

Also, the further apart you place the chords of the truss (the longer the web members), the more the moment in the connections reduce, even if there are eccentricities. At some point moments become negligible (act as pinned). 

There are second order effects when the truss deflects that can cause the connection to deviate from purely pinned. Again less pronounced the longer the web members are. 

This justifies using pinned connections for the majority of trusses that will not have large deflections or large, unevenly placed loads. 

------------------------------
Jeffrey Walkley P.E., S.E., M.ASCE
Vice President
Michael J Walkley, PA
Towson MD
------------------------------

I analyze and design Transmission Towers which are analyzed as trusses. There is a ASCE Standard, ASCE 10, which governs our Industry. It is one of the very few industries that does full scale testing of a new (or old) tower design in order to validate the truss assumptions. The single angle compression equations were the result of testing full scale towers over many years. If a T-Line tower is detailed properly, the axial truss loads from the lacing to the corner legs will induce very little moment to the legs. We do apply distributed wind loads on the members but the software assumes the loads are only at the joints. There is also ASCE 74 that describes the way wind loads are calculated.  

------------------------------
George Watson P.E., M.ASCE
CenterPoint Energy
Houston TX
------------------------------

Just to add my two bits in

IOver the years I have checked the validity of the pin jointed truss compared to having continuous chords or fixed members.  I found that I obtained similar results using pin jointed trusses.  Pin jointed truss analysis provided slightly  more conservative design but not unreasonable.  Alot of what I am going to say has been already said in different ways.

The things I have found is that detailing is critical

Lining up the centroid of the sections at work points  is really important.  There have been failure of trusses caused by detailers making trusses constructable while missing the additional moment caused by the eccentricity at the joint.  Selection of web members and chords profiles may be governed by aligning of the centroid lines.  You may wish to consider a truss as a single member when checking the stability of a structure. 

------------------------------
David Thompson P.E., M.ASCE
Principal
KTA Structural Engineers Ltd.
Calgary AB
------------------------------