Truss Failure

[smcochran]

I recently looked at a truss failure where the vertical truss member from the bottom chord to the ridge had pulled out vertically from the metal press plate at the ridge. There is a high humidity condition within the attic and there was record snowfalls in the area where it occured. Any thoughts on causation of the failure? I think that the high humidity condition allowed the wood to swell leading the plates to slightly separate from the member. Your thoughts are appreciated.

 

[msquared48]

Is this a cold attic - insulated at the ceiling level instead of at the roof structure level?

If so, and there was a roof leak in the area of the failure wetting the connection, it is possible that swelling due to freezing could have contributed to the plate loosening grip too... Heck, this could have happened too since there was a lot of moisture in the wood as you describe - just a different scenario I suppose.

Then again, maybe this was the critical connection to fail under the high load condition. Was there drifting at the time of failure? Possibly an unbalanced snow condition?

Mike McCann
MMC Engineering

 

[MiketheEngineer]

That is usually not a highly loaded member and is usually only in compression - depending on design. Was this a scissors truss - both bottom and top chords pitched??

See if you can contact the mfg or the truss designer who might be able to shed some light on what is going on...

Truss plate might have been too small or placed incorrectly.

 

[swsengineer]

I tend to disagree with you Mike, the centerline vertical member from bottom chord to ridge is usually under tension loads and sometimes pretty high. But you are right, it depends on the design.

 

[MiketheEngineer]

We just need a whole lot more info to hazard an educated guess. I have seen kingposts separate from chords - esp on scissors trusses - but not too common on standard trusses unless the plates were too small or misplaced or truss was way overloaded.

Design, grade and size of lumber, pitch, span, spacing, loading, etc... all need to be reviewed

 

[smcochran]

Thank you for all of the feedback. I will look into all of the possibilities that were presented here.

 

[CJJS]

Agree with swsengineer. The kingpost is usually in tension.

 

[MiketheEngineer]

I mis-spoke - usually in tension - but a lot more info required to hazard a guess

 

[ticat900]

question on roof truss,es in general.Take a 40 foot span and design a truss thats able to stand this span with a 60 lb load factor.Then lets just say after I have bought the truss,es I decide to add a bearing wall down the middle.
Will this bearing wall added after the fact have any bearing what so ever regarding future truss failure of this building???

 

[MiketheEngineer]

Yes - it actaully could - because now some of the web members may (will) be reverse loaded. But bracing them might solve the problem. Also check chords.

Also the internal bearing might be overloaded?? - maybe

 

[ticat900]

Thanks Mike
I am not actually doing this.I suggested this theory to a builder friend of mine whom says I am 100% wrong and that adding a bearing wall down the middle to cut the free span in 1/2 would in NO WAY add to less possible truss failure do to overloading.I said to me that defies logic?
He says the truss always fails in the outter perimeter
In other words towards the outter bearing walls so therefore putting a bearing wall down the middle would in no way enhance the overall structural integrity of the roof truss in this situation??whos right?
Thanks

 

[MiketheEngineer]

Well - again - I am worried about reverse loading in the webs and chords. While he is basically right about trusses generally failing at the ends (perimeter) - a longer span truss could fail due to reverse loading because no one looked at that possibilty. A long tension web or chord is all of the sudden now in compression - failure might loom.

The internal bearing will now carry about 5/8's of the total load. This could cause a problem with perpendicualr to grain loading (allowable about 500 psi depending on grade) or might over load the bearing wall or beam below??

Simply - all loading conditons need to be checked.

 

[Tmoose]

Adding a "bearing" wall at the middle of >>some<< of these trusses would likely break something

http://www.troutcreektruss.com/Assets/Images/truss_types.gif

 

[BAretired]

A few years ago, I inspected a failure similar to the one mentioned in this thread. The king post pulled away from the top chord for all but the end trusses in the building. The building had been a pig barn, so the humidity was very high and the snow was pretty heavy that year.

Underneath all of the rubble, we found a feeding trough which had been suspended from the bottom of the trusses. The trusses had not been designed to support a suspended load and it is probable that this is what caused the collapse.

Best regards,

BA

 

[MiketheEngineer]

BAretired --

Funny you should mention that because it reminds of a failure I ran into many years ago.

Seems like a farmer bought some trusses off the yard - pre-made 24' trusses for residential use.

He put them like 8' oc and hung chicken cages from them.

All went well until the chickens fattened up. You can guess the "rest of the story"

A lot of dead chickens and one mad farmer..