Analyzing an existing wood truss (metal plate) 1

[AdamJSR]

Analyzing an existing wood truss (metal plate)


Hello All,

I am trying to determine the best method to analyze an existing metal connected wood truss to determine if there is any additional capacity left for rooftop installations. Are there any publications out there outline this analysis?

In most cases I won’t have any truss manufacturer specs or drawings.

I was considering just modeling the truss, applying DL’s and anticipated live loads (as per local code), determining reactions, and member forces. Once I have that I can determine stress in the material, but I know that there are other factors to consider other than just member stress,

Any ideas would be greatly appreciated!

Thanks

 

[asixth]

That's how I do it, calculate the tension in the chord and make sure it's less than the capacity. Compression of the chords have a bit more capacity as long as they are restrained. Calculate the deflections and probably add 10% to allow for the nail slip. Check the shear transfer thru the metal side plates as well.

 

[theshearstud]

I recently spoke with one of the author of the NDS Specification about this same topic. His response was that since these products are proprietary (the metal plates) , there is no way of knowing its capacity. He said the only way to properly evaluate this type of structure is to go through the manufacturer. I would check out the Truss Plate Institute and see if they have any recommendations. Goodluck!

 

[asixth]

Good luck with that. Any design drawing will likely be unrecoverable. And most tech's working for Truss manufacturer's will not be engineers despite the fact that the are generating design documents and workshop drawings without understanding the engineering principles that go with them.

In Australia, there are truss manufacturers like Mitek and Pryda that have published capacities for metal side plates so I would have a look at those.

This sounds like a high liability risk application that you are working on. As soon as you rate it for a higher load than the original intended use than you are essentially taking on the responsible for the truss design.

 

[AdamJSR]

I am trying to determine if we can add about 4-6 psf to the roof. It is def getting frustrating trying to find info. I have purchased a few publications with no luck.

 

[theshearstud]

I don't know how large this truss is or how many of them there are, but you can possibly design another connection (spanning over the current connection) to take just the increase in load. Should be a small connection if the increase is only 6 psf. Seems like a solution that is probably unnecessary, but I can't think of any other approach.

 

[AdamJSR]

The number and size can vary a bit. Currently I am trying to analyze a Fink truss, 2x4 members, 2ft OC, 24ft span.

When you say to design another connection, what exactly are you thinking by connection?

 

[theshearstud]

Just an idea…You may not need something this excessive.

 

[jayrod12]

aggie7's detail works.

here we generally just use a square(ish) plywood gusset both sides with clinched nails. Especially for the ~300 lbs total that the truss is going to see, that's like sweet bugger all. 1/2" plywood with a couple nails into each member would cover that.

If it is only a small number of trusses that need reinforcing then this method is what we found to be easiest and cheapest.

 

[SteelPE]

I work with some people who do wood truss analysis/reinforcing and the joints are always the problem (as others have said). The method you describe for analysis is pretty what they do as well. They have managed to secure different metal plate capacities from somewhere. How many "teeth" are engaged and the depth of engagement is critical to the metal plates.

One other source you may want to look at is the IEBC. I believe they allow an increase of 5% in stress before you need to take dramatic members. This may help you eliminate the need to reinforce all of the members/joints in the truss.

 

[woodman88]

Be careful of mixed fastener connections.

My 2001 NDS (I do not have the later ones here at home)section 10.1.4 states "Methods of analysis and test data for establishing design values for connections made with more than one type of fastener have not been developed. Design values for mixed fastener connections shall be based on tests or other analysis (see 1.1.1.3)"

Usually the critical slip of one type on connector is greater or less than the other so you have the existing or the new connection taking a larger amount of the load than calculated.

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.

 

[charliealphabravo]

I've been there. I feel your pain. I try to do one or more of the following.

Plan A: See if the original Truss Design Drawings are available from the owner or building department.
Plan B: Take photos of the plates, sketch the truss, and see if the Mitek or Alpine engineer's will run an analysis for you. They will recognize their own plates and have been helpful in the past in similar situations.
Plan C: Establish a relationship with a local truss manufacturer and truss technician who can run an analysis for you based on your directions (plate sizes, loads, grade, etc). This will not be an engineered analysis since it will not go back to Mitek HQ for seals but it is still better than running it yourself. Insist on paying them a flat amount or hourly for the analysis time so they don't get tired of you. Join the local WTCA/SBC/TPI chapter and make some friends at the local industry meetings if you will be doing this regularly.
Plan D: Do a 2D analysis of the truss for member forces. Then do a shear check on the plates. This can be reliable for single span trusses. The member forces for multispan trusses are highly dependent on stiffness matrices and loading patterns. I'm not as confident in the results for those conditions.
Plan E: Share the load between nearby trusses with strongbacks or structural curbs until you are satisfied that the stress from the new superimposed load is below a reasonable threshold (5% is pretty conservative).
Plan F: Reinforce the truss with gussets, sistering, etc.

Good Luck

 

[TDIengineer]

In regard to mixed fasteners. I would assume he is designing the new connection for all of the load. Therefore if the existing (already stressed) connector begins to fail, the new connector carries all the load.

 

[AdamJSR]

Besides manually analyzing these, are there any good, yet cheap, programs out there? I have flirted with buying Engersoll (spelling?) 2D Truss analysis, but is there anything else out there?

I would have to ahve to sit and analyze this stuff by hand.

Thanks again for all of the good info.

 

[charliealphabravo]

It looks like Truss2D is the most economical thing out there. I am looking at the demo now.

http://www.finesoftware.eu/download/file/demo/truss4/30?lang=en&x=18&y=10

 

[charliealphabravo]

Sorry. That is the link for TRUSS4. I'm not sure what that costs but it looks like it includes a connection plate analysis of some kind.

I am looking at the demo for Truss2D (Engisssol) and it is about $100 for the full program. It is a simple frame/truss analysis app.

 

[AdamJSR]

Been looking and reading about how to design new connections or analyze existing plates (almost impossible) with little luck.

Any direction on how to go about designing the plate thickness and fastner arrangement?

Right now I am trying to analyze a metal connected truss, but I also come across wood gusset truss connections.

Thanks all

 

[TXStructural]

It is relatively easy to analyze a determinate truss using a spreadsheet. We re-engineered an entire building-worth of floor and roof trusses with two or three spreadsheets (and lots of data entry, since they were often skewed and varied in length.)

If you have a good contractor, as we did, you can support the truss, peel old plates and install new ones using a field press. On floor trusses, this was an excellent solution to undersized plates. On larger attic trusses, we used square/rectangular plywood gussets (3/4") and bolts. This was required because our experience told us that we would have no luck getting enough nails into the right places without splitting and splintering the members.

 

[AdamJSR]

Where I am stuck is the methodology behind the design of the gussett. Usually small residential applications.

Need to determine the wood thickness, then number of nails, and then pattern of nailing (or screwing)

Any Ideas how you designed the plates and nailing pattern (that you didn't use)

 

[charliealphabravo]

Here is a good reference for plywood gussets...