Not exactly on subject, but problems designing the IBM 604 Electronic Calculating Punch due to the use of existing vacuum tube designs is discussed in section 2.4 of:
Charles J. Bashe, Lyle R. Johnson, John H. Palmer, and Emerson W. Pugh IBM’s Early Computers The MIT Press, 1986 The book says Ralph Palmer set up a vacuum tube laboratory with the ability to manufacture small quantities of tubes. This helped them get credibility when they went to a vacuum tube manufacturer with a request for a design change to achieve the needed reliability for digital applications.
