Bottom Chord Failure of Wood Roof Truss

[halemaneP]

We have 46 feet span roof trusses for an auditorium. One of the roof truss had bottom chord failure.We have temporary jack supports until the permanent fix is made. The truss failed at the splice. Splice was not at the node point, but almost at the center.

Since it is a clay tiled roof, the roof dead loads are heavy, about 27 lbs per sq.ft My preliminary calculations indicate about 13,000 lbs of tension. splice which failed was done with ordinary steel press connectors, and not bolted. bottom chord is 2, 2X6 commercial grade southern pine.

I plan to use steel plates with turnbuckle to pretension the truss, with both sides supported at nodal points while turnbuckle is turned.

The question is how to make sure that we pre tension to the exact load ?. How we can measure the tension load on the turnbuckle?

 

[dik]

With TPI (truss plate institute) type of connections, the splice can be almost anywhere on the bottom chord as long as it doesn't interfere with one of the panel point plates. You have to determine if it failed from overload or faulty installation, and if there are other trusses that are similarly affected. Faulty intallation would include if the truss plate was lifted away from the dimensioned lumber by a little bit. The truss may have been damaged during installation.

See if you can locate shop drawings for the trusses and see what the design loads are. Repair would likely be by jacking it into positon, not by applying a tensile load to represent what was in place. Once the jacks are removed, then the new splice would be tensioned. If you are not an engineer, you should check with a good local structural guy...

Rather than think climate change and the corona virus as science, think of it as the wrath of God. Feel any better?

-Dik

 

[halemaneP]

The truss separated with a wide gap. it is the failure of the joint. Bottom chord was completely de loaded up to next nodal point.
it was not supported with jacks fora week, and when it was noticed that adjacent trusses are being impacted, we have jacked the adjacent trusses also, until a fix is found.

There is false ceiling, which is to be removed to inspect all the trusses.

 

[dik]

Did the pressed steel plate pull away from the one side? You have to be careful about jacking to make sure damage isn't being done to the truss members. Were you able to obtain drawings to determine what the design load was? Did the structural drawings show what the construction was? Could all the trusses have been designed for the wrong load? Lots of questions before you engage a 'fix'. Was the truss locally overloaded in recent history.' If installed correctly my experience with truss plates is that they are pretty reliable unless damaged during erection. I've encountered where truss plates have worked away from the lumber due to thermal recycling and moisture.

Rather than think climate change and the corona virus as science, think of it as the wrath of God. Feel any better?

-Dik

 

[dik]

any photos?

Rather than think climate change and the corona virus as science, think of it as the wrath of God. Feel any better?

-Dik

 

[halemaneP]

We had the complete break in the chord one side by pulling away from the splice, and noticed by sagging under splice.
when we jacked we jacked only at next nodal points on either side gradually without distorting the truss anyway

The structure is about 8 years old. It is in a condominium association in central Florida.

We noticed first when we saw sagging false ceiling. I am a retired structural engineer, residing in that and only helping the association. Condo association has hired a consultant, we have not seen their report yet.

Truss is not twisted, it is bent in the vertical plane.

My primary question still is, if we use the turnbuckle, for a correcting the situation is there any way i can find the the force in the Turnbuckle ?

 

[dik]

I would not be using a turnbuckle, it's too difficult to estimate the force. I think it's best to jack it into position (maybe a tad higher), make the repairs, and slowly release the jacking load. You really have to determine what the design loads were.

Rather than think climate change and the corona virus as science, think of it as the wrath of God. Feel any better?

-Dik

 

[dik]

I'm not into wood... but occasionally dabble in it; there are many others on this site that will be able to offer better information.

Rather than think climate change and the corona virus as science, think of it as the wrath of God. Feel any better?

-Dik

 

[Compositepro]

I wonder why you would want to measure force rather than just closing the gap?

If you use an Enerpac hydraulic jack to close the gap the pressure gauge will measure the force being applied.

 

[halemaneP]

just closing the gap with steel plates, will not introduce Tension, without pulling two sides together to close the gap.

Can Enerpac be used to pull two horizontal members together ? if it is possible then we will try that. thanks

 

[Compositepro]

I meant closing the gap by pulling the ends together, not filling or covering the gap with something.

 

[dik]

The tension will be introduced once the jacking forces are removed; the truss has to be jacked into proper position and repaired, then the jacks removed. Once the repaired truss deflects, tension will be introduced into the bottom chord.

Rather than think climate change and the corona virus as science, think of it as the wrath of God. Feel any better?

-Dik

 

[halemaneP]

dik and Compositepro

Thanks for all your suggestions. it was really helpfulH

 

[SlideRuleEra]

My preliminary calculations indicate about 13,000 lbs of tension.

If static tension was 13 kips, it will take a lot more than that to get the break back into position. The equipment typically used for "pulling" is a lever hoist, better known as a "come along". Note: The link is just for information, not a recommendation on brand or load rating.

In general terms, I would use a pair of come alongs to prevent eccentric pulling. Connection points would be at the ends of the truss, so that the truss is really being horizontally "compressed" back together rather than "pulled". At the same time, a jack is vertically lifting the center of the truss. To rig this correctly then coordinate equipment operation is not a job for amateurs.

Turnbuckles are notoriously weak, providing enough torque to accomplish this "pulling" is probably neither practical nor safe.

 

[dik]

Thanks SRE... forgot to add the comealong to pull them together.

Rather than think climate change and the corona virus as science, think of it as the wrath of God. Feel any better?

-Dik

 

[SJBombero]

The splice repair could be detailed to pull the two ends together with a threaded rod on each side of the truss that tension the truss concentrically until it closes.

13k is a lot to ask of the NDS dowel bearing equations. I would look at shear plate connectors.

 

[XR250]

13K tension seems like a lot for a dbl. 2x6 chord. Have you checked the truss to see if it was designed for the current roof loading?

 

[halemaneP]

We will double check the bottom chord tension 13k load. The truss is not a deep truss, only 2'9" deep,46 feet span, and sloped tiled roof, truss spaced at 2ft centers, so the dead load is high. we will also double check the designed loadings.

Also come along is a good idea to compress the truss between the ends, until it closes, keeping it in the same plane. we will investigate it, to compress the two sides of chord together, along with raising the jacked portions near center which is level now, slightly higher. We are still awaiting Consultant Report.

I am really thankful for all the good suggestions.

 

[SlideRuleEra]

halemaneP - During compression of the bottom chord the broken ends may not line up by themselves. Should have four smaller come alongs, two on each broken end (pulling perpendicular to the trusses). Then each broken ends can be kept aligned as the truss is compressed. Adjacent intact trusses can be temporarily braced together to anchor the reaction end of each of these four smaller come alongs.

 

[halemaneP]

Thanks to SlideRuleEra, XR250, pvchabot, dik and Compositepro for their very helpful suggestions

Now I have clear ideas how to approach the problem.

I also recalculated. Dead load is about 30 lbs/sqft. With roof inclination about 22.5 degrees, spacing of truss at 2 feet, with 47 feet span, 68 Lbs loading per running ft of truss, as a beam, Maximum bending moment comes about 18777 lb.ft. with effective height of the truss being 2.875 ft, bottom chord tensile load is about 6600 Lbs, nearly 1/2 of what i stated earlier.


Thanks.
halemaneP