I think the point about accepting failure here probably gives an incorrect impression. He had the fundamental idea correct, and the many prototypes (I'm scepti

I think the point about accepting failure here probably gives an incorrect impression. He had the fundamental idea correct, and the many prototypes (I'm sceptical on the 5000 figure) were working on improving the design, not trying to make it work the stuff we expect a vacuum to pick up is not homogenous, but rather consists of a huge variety of materials and particle sizes. He found that one set of parameters would work well for certain particles, but not so well on others. Most of the prototypes consisted of tweaking dimensions and recording the resulting effectiveness.