Calculating the deflection of a wood truss 3

[AELLC]

I don't have truss or 2D frame software - if I assume the Moment Area of Inertia based on the cross section of the top and bottom chords, as if they were connected with solid web that had no deflection resistance by itself, i.e. solving it as a simple beam, is that ballpark or will give a deflection result too much, or too little?

 

[AELLC]

I am concerned about it because it is the girder truss with very heavy loading, 33' long, and only 28" deep. The truss software (from the truss mfr) shows a total load deflection almost to the allowable ratio.

 

[rb1957]

sure you can estimate the MoI as the effective two caps, this is slightly conservative.

Quando Omni Flunkus Moritati

 

[paddingtongreen]

That will under estimate the deflection. In most beams, shear deformation is small, but in trusses it is more significant, but I know of no general method of estimating it, short of performing the calculation.

Michael.
"Science adjusts its views based on what's observed. Faith is the denial of observation so that belief can be preserved." ~ Tim Minchin

 

[AELLC]

Thanks, Mike, that is what I needed.

When I did a hand calculation as described above, I got the exact same value (by coincidence) as their truss software.

Therefore I suspect their software is a bit unconservative, especially because this girder truss is so shallow (span to depth) and is very heavily loaded.

 

[dcarr82775]

I find it isn't a terrible estimate to use 80-85% of the I of the chords

 

[AELLC]

dcarr

Thanks.

 

[rb1957]

MoI of chords < MoI of the full section, yes?
deflection based on MoI of chords > deflection based on full MoI, no?

Quando Omni Flunkus Moritati

 

[JoshPlumSE]

If you look at the SJI or Vulcraft catalogs or such, it appears they tend to use a 15% increase in their deflection. Probably to account for shear deformation. This would agree nicely with what "Dcarr" posted.

 

[AELLC]

rb,

Yes

Yes

 

[paddingtongreen]

rb1957, You would be correct for a beam but wrong for a truss. The deflection caused by the strain in the interior members of a truss increases the deflection.

Michael.
"Science adjusts its views based on what's observed. Faith is the denial of observation so that belief can be preserved." ~ Tim Minchin

 

[XR250]

Heavily loaded or shallow, long span, wood trusses also suffer alot of creep due to nail plate slip. Something the truss companies will deny, but I see all the time.

 

[AELLC]

Excel,

Yes that is worrying me a lot about this girder repair that the truss mfr came up with - I recommended they make it deeper but they couldn't, claiming it was impossible because of existing electrical conduit - grrrr.

It is only 28" deep, has a span of 33', and a dead load of about 480 plf

Then to add another layer of anxiety, they stupidly did not originally check how the supported trusses load transfer to each of the 5-plies, so they stupidly had to (in a hurry) used 1/2" dia bolts that further reduced the MoI, and they stupidly made the statement, well...the truss software can't take that into consideration.

Of course, worrying about the huge torsion on that girder would be something they wouldn't be capable of either - they had to use top flange case Simpson THA222-3 hangers, which IMO exacerbate the torsion problem.

All this had to be resolved Friday because they had to fab the girder Friday night and ship it to the jobsite Saturday (yesterday)

 

[AELLC]

^^^THA222-2^^^

 

[XR250]

Can you run some straps from the bottom of the supported trusses, across the bottom of the girder and up the back side to resolve the torsion?

 

[Jerehmy]

just make a truss stiffness matrix...

 

[AELLC]

I think the 2x scab does help resist torsion because it effectively "clamps" the girder against rotation, but this is yet another example of something we can't figure a rational number.

I asked them how they came up with that number of nails but got no response - see attached

 

[theonlynamenottaken]

I clearly remember from college my structural analysis professor telling us that for parallel chord trusses you could get a rough initial estimate of the deflection as:

Twice the deflection calculated for a beam, with the beam moment of inertia calculated only from the chord areas (excluding the bh^3/12 of the individual chord elements)

as AELLC referred to in his original post.

See the attached reference from Andrew Orton's, "The Way We Build Now: Form, Scale and Technique" that seems to agree.

 

[AELLC]

I checked the truss software deflection, and it exactly matched the method of using only MoI of chord areas - to 3 significant digits.

However I am doing a study in my spare time, saying that is moderately inadequate.


Where did he get that? Sounds over-conservative.

 

[AELLC]

It doesn't say twice the delta, it just says delta is higher. See above posts by others, they are estimating 15-20% higher, not 100% higher.