pinned connections in trusses

[lvkb]

Its been a while since I've had to do structural detailing and from my dim memory of truss design in college I seem to remember that if the centrelines of all the members at a joint intersect then the joint has no moment capacity - i.e is a pinned joint. Is this right?

 

[JAE]

The orientation of the intersecting members has nothing to do with pinned vs. moment connection characteristics. The orientation simply determines whether eccentric axial loads will develop at the joint.

The type of connection at the joint determines whether they are pinned or fixed (moment).

 

[Bagman2524]

I think a truss analysis assumes pinned connections.

 

[rb1957]

truss analysis assumes that the members carry axial load only, that each strut is a simple two force member. this theoretical state is only achieved if the members join with a single pin (well greased?) and (if you're getting picky) on the neutral axis of each of the members.

in reality the connection is a multiple pin arrangement (therefore capable of reacting moment) and offset from the neutral axis (of at least some members) therefore inducing bending in the strut (the load axis isn't aligned to the strut). in reality these two effects are usually small enough to be neglected.

to answer the OP, the strut axes may meet at a point and the struts can, or cannot, react moment depending upon the attachment between the struts (whether it is a single pin or a multiple pin).

 

[kxa]

If you are using a computer program to analyze a truss, you woud have to make sure that you release the end moments at the right plce to create the pin connection. For example, your top chord may be continuous but have many diagonal members attaching to it. In this case, you should release the end moments for each diagonal member only.

 

[aggman]

It is typical in structural engineering to neglect the rotational capacity of the web members in design. In certain cases, long span, or welded connections, at some times to include the moment effects of the web members. While theoretically to include web member rotational restraint typical practice neglects the moment capacity of the connection in design even if it has multiple bolts in the connections. Almost all, I would say 90% or more, of civil structural trusses utilize multiple bolted connections and truss action is assumed in analysis. This has been successfully practiced long before any of us were engineers. If you have eccentricity at the connections (centerlines not lining up at the panel points) then you need to include this effect in your analysis. If you choose to analyze the truss utilizing rotational resistance of the web members, which is valid in some cases, then you have a whole different set of detailing and analysis items you have to deal with.

 

[kxa]

If your members are not deep, should be able to neglect eccentricity.

 

[aggman]

Dag I need to read my post before I send them! Above should say, (Sentence 3) "Analysis including web member rotational restraint is theoretically correct typical..."

Sorry about that.

 

[mrMikee]

If a "typical" truss were analyzed on a computer as a frame it's my guess that the bending stresses would be not be significant, assuming of course that centerlines and work points intersect. When member sizes become relatively large then secondary stresses would need to be considered, but I don't remember any rules of thumb that would be helpful.

Regards,
-Mike

 

[mrMikee]

kxa,

In your post I would say instead that "if your members are not deep" you should be able to neglect secondary stresses due to moments. I would however be very careful with any eccentricities caused by centerlines not intersecting at a point.

-Mike

 

[kxa]

It all depends on the way the truss is configured and how it is loaded. You could have significant negative moment over the joints for your continuous members.If every single member is pinned at the joints, then the bending moment depends on the load each member sees.

 

[kxa]

mrMikee, I agree. I was concerened with joint/connection design. I guess you can always model a dummy rigid member to carry the centerline to the surface of the member at the connection/load transfer point.

 

[SlideRuleEra]

lvkb - I have a worked out, pinned truss problem (zipped PowerPoint file), in the Miscellaneous portion of my web site, link below. The title is "Truss Analysis With Point Load". This may help to refresh your memory, since the problem seems a lot simpler than it is.

http://www.SlideRuleEra.net

 

[prex]

What someone above seems to miss is that the moments developed in the elements of a truss (with intersecting neutral axes) need not be negligible (and often they aren't) to be neglected in the analysis.
In fact those moments give rise to secondary stresses (as they are called in pressure vessel analysis). Such stresses are not load controlled quantities, instead they are deformation controlled quantities: the distortions they cause will tend to decrease them, and even if they reach the yield condition, they will stay there and no more increase (whilst a primary stress, if the external loads increase, will increase till the structure collapses).
Our grandfathers didn't know the theory behind that, but that's why the truss method of analysis works so well.

prex

http://www.xcalcs.com

Online tools for structural design

 

[lvkb]

Thanks for all your posts! They are all helpful.

 

[IFRs]

Having designed many trusses and observed their behavior, I think that anayzing trusses with pinned ends reasonably approxiamtes real life UNLESS there are significant deflections or the truss is made of very slender members. The definition of significant is of course relative to each truss and the connection designs. When deflection under load become significant then the members begin to "feel" the end restraint. It can also be said that for certain trusses, the member end fixity determines the resistance to deflections. So, it's not simple unless your truss is designed, loaded and behaves in a certain way. Fortunatly, this tends to be the most common scenario.

 

[valleyboy]

I agree with the sentiments expressed in the above posts if a conventional truss arrangement (like a pratt or warren type truss) is used.

However if a Vierendeel type arrangement is used then the members should be designed to resists axial load, moments, and shears.

VB