Lower envelopes are fundamental structures in computational geometry that have many applications, such as computing general Voronoi diagrams and performing hidden surface removal in computer graphics. We present a generic, robust and efficient implementation for computing the envelopes of surfaces in 3. To the best of our knowledge, this is the first exact implementation that computes envelopes in three-dimensional space. Our implementation is based on CGAL and is designated as a CGAL package. The separation of topology and geometry in our solution allows the reuse of the algorithm with different families of surfaces, provided that a small set of geometric objects and operations on them is supplied. We used our algorithm to compute the lower and upper envelope for several types of surfaces. Our implementation follows the exact geometric computation paradigm. Since exact arithmetic is typically slower than floating-point arithmetic, especially when higher order surfaces are involved, we minimize the number of such operations, to gain better performance in practice. Our experiments show interesting phenomena in the behavior of the divide-and-conquer algorithm and the combinatorics of lower envelopes of random surfaces.