Something similar happened with the Hubble telescope's infamous blurry mirror The manufacturer tried using two instruments to measure the shape of the mirror.

Something similar happened with the Hubble telescope's infamous blurry mirror The manufacturer tried using two instruments to measure the shape of the mirror. One instrument said that the mirror had unacceptable error. The other said it was perfect. The latter instrument was designed to be more precise, so they went with that and ignored the measurements from the other null corrector which disagreed. The problem was, the more precise null corrector was the same one that was used to help construct the mirror originally. All it could verify was that the mirror was built to the same specs as the null corrector. No one realized that the "better" null corrector itself was off until it was too late.