What most fail to realize is that when a compression flange on a wood i-joist is repaired, we are not somehow magically taking the load out of the flange, diverting it around the joist, and then magically putting it back into the flange at whatever location we see fit. If that were the case, there would be no tension or compression couple within the flange (we took that load out and put it into the repair), and thus no fiber elongation, and therefore, no deflection at this location. To simplify, there is significant difficulty getting the load out of the flange and into the repair member, due to the significant stiffness difference and connection methodology of the repair member compared to the original joist.
What compression flange repairs typically provide is a reduction in the joist failure mode, which is typically euler buckling at the reduced section of flange (combined with the induced eccentricity from the notch). The repair helps to provide additional buckling restraint for the reduced flange section, but it does not typically return the joist to full strength. There are ways to achieve full strength, or near full strength, but they require significant composite action development of the repair, as well as resolution of the eccentricity to prevent failure prior to achieving allowable design loads.
