Perhaps you could ask him?
I've never built a printer, but some of my gut feelings include:
-- A "dot matrix" printer wouldn't be that hard. The printhead is
effectively a bunch of tiny solenoids... and if you look back far enough,
some of them weren't exactly "tiny!" 9 and 24 pin dot matrix printers were
quite the norm for a number of years, and I believe really old ones used as
few as 7 pins.
-- Daisy wheels are even simpler (just one "hammer")... if you can still
find the wheels.
-- To build an inkjet with a reasonably small dot size, you need access to
an IC fab process -- although it can be a *really* old one (older inkjet
heads have feature sizes on the order of tens of microns).
-- The right person could put together some Frankenstein-esque laser
printer, but if you don't have access to an optics lab and a good machine
shop, I wouldn't even try.
Part of the problem you're having here is that you haven't told us what
*kind* of printer you're trying to build -- hence the people asking for
books on building a space shuttle -- and whether you're trying to just do
some science fair experiment, perform a "proof of concept" demonstration for
venture capitalists, build and commercialize a product out of your garage,
etc. Something like a 40+ PPM office laser printer or a mid-level consumer
photo printer has so many "pieces" that there's almost no individual who
could build one himself. On the other hand, if you're trying to build a
customized version of, e.g., the $29 printers at Wal*Mart, you'll have a
shot if you get the printhead from the likes of HP or Canon. (I seem to
recall HP once had a program whereby they'd provide inkjet printheads and
ink cartridges, and then you built the rest of the printer around that.) Of
course you'll spend far more than $29 to do any of this (where I went to
college, there was a group that had designed a braile printer which was
eventually turned into a commercial product... due to the low production
volumes, the cost per unit was still thousands of dollars... the low-end
printer marketing model is effectively to give away the printer at cost and
make money on the ink...).
For some "diversity" requirement in college I took an industrial engineering
"electronics manufacturing" course where they have an exploded view of the
original IBM ProPrinter compared to that of the ProPrinter II... they
reduced the number of discrete mechanical parts from something like 100+ to
a couple dozen. It's a cool drawing... (and the ProPrinter II was a very
good dot matrix printer -- I bet there are still some in service
somewhere...)
That's about as close to a book on printers as I can get you.
As I say, you really need to narrow down just *what* you'd like information
on: Printheads? Carriages? Control electronics? ...etc...
---Joel