Yeah, that doesn't seem unreasonable at all, especially since producing a page on (made-up example, obviously) maintaining and replacing the starboard arglebarg

Yeah, that doesn't seem unreasonable at all, especially since producing a page on (made-up example, obviously) maintaining and replacing the starboard arglebargle frammistat is gonna require that the technical writer spend a fair amount of time talking with the engineers who designed the frammistat, as well as the techs who will perform the actual maintenance/replacement work. You've gotta know how to tell when the frammistat can be safely reused and when it has to be replaced, how much torque can be applied to it without messing it up, what lubricants and sealants can and cannot be used, maximum temperature ranges, all kinds of stuff, because if you mess up the docs, the plane falls out of the sky and kills the President of the United States, plus potentially many others on the ground. That would leave a significant resume stain. Then, of course, the final document needs to be reviewed and signed off on by all those people, likely with multiple revisions. 8.4 hours per page seems like it'd be on the low side, if anything.