In our practice, we quite often check existing buildings. The point is to establish whether they can carry new loads, or whether a photovoltaic installation can be mounted on them. We often come across calculations done 20–30 years ago, and not infrequently the structure turns out to be damaged already.
Many years ago, RM-Win was the main software used in Poland to calculate 2D structures. It is great software and we used it for years ourselves. It had, however, one default setting that caused a lot of trouble: the buckling length was equal to the real length of the member. This could easily be changed, yet many designers did not do so.
In the drawings we show a deliberately exaggerated example, so that the difference in results is clear: buckling length equal to the spacing of the braced purlins (the correct approach) versus buckling length equal to the distance between nodes. The utilisation came out at 36% against 215%!
Can this be repaired on site? Yes, but it will not be a cheap repair.
