Truss Question 2

[bpstruct]

I have a project where an architect is wanting to use roof framing as shown in the pics attached. I can't see that these web members are doing much of anything....especially those in compression. The compression members don't seem stable since that bottom chord is unbraced. The top chords appear to be welded at the ridge and the bottom cable I'm thinking is doing nothing more than keeping the supports from spreading. I'm certain that, in an uplift situation, nothing is doing any work except the top chords in bending. Has anyone done anything like this before? I'm not too worried about the idea, as this is just a roof for an outdoor amenity center. I just want to make sure I'm not overlooking anything. I'm actually thinking that I would prefer to not have any web members. Under deflection, I'm thinking that any solid web member might have a tendency to push the cables out of alignment with the top chord above. I'll post a couple more pics in responses to this post.

 

[bpstruct]

Pic 2

 

[bpstruct]

Pic 3

 

[dik]

In pictures 2 and 3, the horizontal member is keeping the ends from 'springing' apart... in picture 2, the vertical element is keeping the horizontal member from sagging... I can't make any sense out of picture 1.

Dik

 

[SteelPE]

I have see trusses done like the first picture you gave. One building that I know about has 4-1 inch diameter bars that support the bottoms of columns that were removed during a renovation. Under gravity loading, since the bottom chord is in tension rods are possible your compression struts will take load if sized appropriately.

I agree with the fact that there is noting to prevent the bottom chord from coming out of alignment. I remember a post on this website from a number of years ago where people discussed using K=1.0 for the end diagonals of steel joists as there was nothing to prevent buckling of the bottom chord/compression diagonals out of plain. The thread got to technical for me so I stopped reading after a while.

 

[msquared48]

The second and third are just collar tied situations - very normal really.

The first is a modification of the collar tie, but unstable laterally at the bottom harped connections unless bradd to the side. Perhaps the metal diagonal at the end has some fixity to account for the lateral int=stability, but the connection to the top chord would have to be designed for it too. This is a similar arrangement to the post-tentisioning solution for glulab beam upgrades too.

Mike McCann
MMC Engineering

http://mmcengineering.tripod.com

 

[JoshPlumSE]

I agree with Msquared. The 2nd and 3rd are just collar ties. Though perhaps modified a bit to be more architecturally interesting.

The first one is odd. Kinda like a queen's post truss, but inverted so that the compressing members are inclined. If those vertical member connect to the top chorch properly then I think it could work. Though, this truss certainly does not look like it was designed for an uplift scenario.

 

[msquared48]

Man. I just have to proof read my comments! That was bad!

Mike McCann
MMC Engineering

http://mmcengineering.tripod.com

 

[woodman88]

As I see it. The compression web members hold the bottom horizontal cable at typical height as the tension webs are tighten. Without the compression webs, as you tightened the tension webs the bottom cable would lift.
Why it was done? I would say that the owner/architect wanted more than just a vertical member (or three verticals) in the truss to control sagging of the bottom cable.
As for out of plane buckling, under gravity loads the tension in the bottom cable should control it. Under uplift loads, there will be no forces in the webs or bottom cable to cause buckling.


Garth Dreger PE - AZ Phoenix area
As EOR's we should take the responsibility to design our structures to support the components we allow in our design per that industry standards.

 

[msquared48]

Also notice that the bottom tie is harped at the end to get a vertical component that will go to the end of the beam to bend it down.

I think this was for the look, more than to control any sagging here, but that's really a moot point.

I would have to disgree though on the lateral instability issue. If the end diagonal gets out of plumb for any reason, the combination of the lateral misalignment with the compressive force in the diagonal will create a P-delta effect that could cause it to fail. That's why we have rat runs.

Mike McCann
MMC Engineering

http://mmcengineering.tripod.com

 

[dik]

Not that bad Mike... Josh understood you...

Dik

 

[dhengr]

Mike:
I just thought you were working on a new version of the English Language for Structural Engineers..., with fat fingers, mostly thumbs, and small keyboards, who can’t type. And, now you gotta spalin to us what a rat run is, or why you don’t just tell those rats to quit running there so they don’t set up any lateral, out of plane vibrations.

Won’t any compression in the strut just tend to align the harped tension members, and keep them in plane, unless the strut actually buckles? I wouldn’t hook up a trapeze set-up to the harped tension members. Then, any roof uplift just tends to make these lower members go slack (or at least have less stress) and causes the rafter member to act alone in bending upward, as suggested by Woodman. As JoshP mentioned, these rafters are just two inclined queen post trusses tied together, in a harped fashion, to prevent them from spreading to much at their bearings. And, they look light and sexy too.

 

[BAretired]

The truss in Picture No. 1 is similar in configuration to the wood truss shown in the link below except that the latter has a straight bottom chord whereas the former is slightly harped.

It would be good practice to provide a lateral brace at each bottom chord panel point. However, I do not agree that there is nothing to provide lateral bracing. While perhaps not ideal, tension in the bottom chord limits lateral movement. In order to move sideways, the bottom chord must elongate.

BA

 

[SteveGregory]

For the truss in Picture No. 1, you could add cable ties to the bottom chord panel point at the web compression member. The cable ties could be attached to the top cord panel point of each adjacent truss at the other end of the compression web member. You would then have two rows of x-braced cables for your trusses.

 

[BAretired]

You could do that, Steve but I'm guessing the Architect would not like it. The cleanest solution is to use a tied arch as in Picture #2 or #3.

The structure in Picture #1 is already built, so the question is, do you insist on additional bracing or leave it as built?

BA

 

[SteveGregory]

BA, I agree with you. It just depends on what look the Architect wants. The construction as shown in #2 and #3 would not need any changes in concept, but #1 would require the extra bracing.

I assume BP is just giving us examples of the styles that are being considered for a new structure.

 

[BAretired]

What if you took the roof in Picture #2 or #3 and added web members as in Picture #1. Would you require extra bracing?

BA

 

[BAretired]

In other words, if we start out with an acceptable structure, namely a tied arch or, as others have called it, a collar tie, why should it suddenly become unacceptable when we add a couple of web member on each side? Think about it.

BA

 

[RFreund]

I think I agree with BA, I believe the truss is stable as shown in figure 1. Assuming the webs are stable under compression (pin-pin condition)
My first thought was that that the bottom chord would want to move laterally and the truss would buckle but in order for that to happen the bottom chord must elongate as BA had mentioned. Makes sense now.

EIT

http://www.HowToEngineer.com

 

[hokie66]

New to this thread, but I don't agree that Truss 1 is like Trusses 2 and 3. 2 and 3 are tied arches, and are stable under gravity loading.

Truss 1 has the bottom chord in tension, and a component of the tension must be resisted by the webs. Without bracing, or the webs themselves being able to cantilever, the panel points can move laterally. Yes, tension in the rods will tend to equalize this movement, but we don't model it that way.

None of these trusses works for net uplift unless the continuous gable chord is sufficient in bending. There are a lot of old buildings standing with this type roof, particularly churches, but presumably they have the protection of a higher authority.