changing the workline after reinforcing a truss

huangyhg

changing the workline after reinforcing a truss
good afternoon,
i recently had to do a large truss analysis / rehab project for a local factory.  there were approx. 20 carrying trusses, all spanning 60' and approx 15 jack trusses spanning 80'.  after i determined the amount of additional steel that a particular member would need, i'd recalculate the members centriod --> then the members i --> then s --> and down the line to determine the new capacity of the section.  i wired a spread sheet that (depending on the reinforcing type, whether it was a chord member or a diagonal and either compression or tension) would re-calc all those things above using only the dimensions of the additional steel reinforcing plates as input.  
in a nut shell i would try to 'tweak' my reinforcing as to not change the overall centriod of the member too much as that would create an ecentricity (because the axial load would be now multiplied by the distance between the old and new centroids).  was this too much of an analysis, would you of bothered worrying about that additional moment from the changing centroid.  the code implies that we don't need to worry about eccentricty of connections, but i couldn't find anything stating the same about members in this type of circumstance.  
also, if anyone does do reinforcing this way, how much eccentricity is acceptable?   
make all added steel symmetric about the centroid, i.e. use two plates, one on each side or four plates with one each on top, bottom, and each side.  this eliminates eccentricity and having to calculate section modus of top and bottom.
i think you did the design in a proper manner.  i would worry about introducing eccentricity into the connections.
both civilperson and rarmbj have good posts...
there is a well-known collapse of a roof to a library in the states which was due to eccentricity at connections in timber trusses.  the "unknown quantity" as i have heard it put which permits owsj to have non-alligned panel points does not, in my view, constitute good engineering practice, precisely because it is both unknown and unnecessary.
you did exactly the right thing, and should do it again when faced with similar situations.
cheers,
ys
b.eng (carleton)
working in new zealand, thinking of my snow covered home...