How to analyze a truss with a panel missing?

[stuctural101]

How would you analyze (by hand) a truss with a panel/diagonal missing? I don't think the method of joints would work.

 

[frv]

If I understand correctly, you are ending up with a "square" bay somewhere within your truss.

If this is correct, the truss would be unstable under the assumption that the ends are pinned.

I guess you could weld everything and analyze as a frame?

 

[stuctural101]

That's right. For my line of work (conveyor systems), we usually assume simple beams & pinned connections (moment ignored) for member sizing, which is conservative. I really want to leave analyzing it as a frame as a last resort. Any other suggestions?

Here's the background:
I'm analyzing wind load for the top chord as a 2-D warren truss w/ verticals (16 original bays w/ 2 bays at the end removed).

 

[jike]

This is not easy by hand.

I would model it as three panels, the vierendeel panel in the center and the two adjacent panels. The bending continuity of both the top and bottom chords (3 span continuous beam) will help carry the shear across this panel. Don't forget to also add in the axial forces of the chords.

You might also load the truss with uniform load from one end to the center since this will give you the maximum shear at the center panel.

 

[hokie66]

Your last post indicates that you want to leave out the diagonals at the end bays, which is of course where the shear is greatest and the diagonals are required most. There is no option but to design these bays as rigid frames.

 

[JLNJ]

A few simple approximations will get you close.

At which point you might find out that leaving a diagonal out is not really viable without major upgrades of the chords.

 

[rb1957]

if you don't have a diagonal member to take the transverse load, then the chords have to be beams.

your two choices (as i see it, assuming you can't add the diagonal) are ...
1) make the chords pinned ended beams, actually probably beam columns as i imagine that one of them is carrying a compression load, or

2) if this makes the chords too large, then build-in the ends, making the chords cantilevered beams (only slightly more difficult to analyze, a different page in roark). this will reduce the peak moment in the chords (by about 1/2), but you'll have to provide the structure to react the end moments (gussets, bolt group) and carry the moments into the adjacent bays.

good luck, analytically it doesn't look that difficult.

 

[csd72]

This is easy, analayse it by method of sections taking into account the equilibrium of bothe shear and bending.

 

[jike]

stuctural101:

I posted my response at the same time that you posted more info. As others have stated, the end panels will have the highest shear which will most likely require modifications to the chords and verticals. The chords to take the bending from the shear (I am assuming that this is a parallel chord truss) and the verticals to keep the chords from rotating too much. You should be able to do an appoximate analysis just like a vierendeel truss (since these are vierendeel panels).

Your first approximation would be to assume zero moment at the center of the chords and verticals (just like a portal frame laid on its side).

 

[RARWOOD]

I would not even attempt to analyze it by hand. With the wide availability of computer programs to do this type of analysis, I think it would be not be prudent to design something like this by hand.