metal plate connectors in existing truss

huangyhg

metal plate connectors in existing truss
i’m looking at some existing 2x4 wood floor trusses. i need to come up with an allowable load for a client that wants a change of use (which will require a larger load.) now i already know the answer: the trusses aren’t gong to have the extra capacity (125% of what the original should have been) because designing exactly to what’s needed is how the truss manufacturer makes money. but just for the sake of due diligence, i’d like to come up with a number for the guy. obviously getting stresses in members isn’t tough. however, i’m not sure how to evaluate the connections. short of tracking down the manufacturer, is there a good reference for the evaluation of metal plate connectors?
ansi/tpi-2002  national design standard for metal plate connected wood truss construction.  
i may be mistaken, but i think it's a time consuming and complicated task to calc the plates.  i might suggest having a truss designer run a truss per as-built, then run the same truss with the new loading, same web config.  you should be able to see the effect in the larger plates and higher grade lumber needed.  
you are correct that the key limit state (weak link in the chain) is the connector plates - not necessarily because they are truly the first to fail, but because you have no clue what capacity they provide.  many are proprietary plate systems with unique punched teeth.  unless you can find the original truss manufacturer to get you the actual tested, approved data on the plates, you are just guessing as an engineer.
now usually, a good engineered product would have its connections be stronger than the members...you could assume that the weak link is a member - analyze the truss and find that member and thus have a truss capacity.  but as an engineer, you really can't certify a truss based on this assumption.
one thought would be to suggest that the plate connectors would provide 75% of the   
metal plates usually have the manufacturer's name stamped on them, or some other identifying mark.  if the truss isn't too old, tracking down the plate manufacturer shouldn't be too difficult, since there are only a few in the u.s.  if you find out, you can contact the manufacturer and request their specifications for their plates.  (mitek, robbins, and alpine are three of the big plate manufacturers).  often, the controlling failure mechanism of the joints is plate pull out.  so, the lateral resistance values (gripping) of the plates is important to know.  this of course depends on the lumber used and the plate type (more specifically, the geometry of the plate teeth, which depends on the manufacturer).      
thanks for the insight, guys.
i did manage to track down the truss fabricator and we determined that the connections are going to govern the allowable load and there is, as i expected, no extra. it makes sense that the metal plates are almost always going to be the governing component. after all, the wood is generally limited to one or two sizes (i.e. single or double 2x4) and maybe 3 available species/grades, whereas there are a lot of possible sizes for the connectors, and a lot of connections in a single truss.
again, thanks for the direction.