Weighed grain and gold at the harbor scales for twenty years. The scale was wrong by five grains. It was always wrong by five grains.