Digital BW, The Print

On 15 Apr 2009, at 8:00 AM, Walker Blackwell wrote:

> We need a system that is beautifully engineered,
> simple, manual, and of the highest quality. In other words, we need
> Hasselblad 500C printers.

I'd suggest "view camera printers" instead: no proprietary/fiddly  
technology at all, beyond the lens & shutter (in this case, the print  
head stands in for our lens).  Everything else could be made by hand  
-- either low-end (homebuilt) or high-end (like Sinars).

> A bare-bones print engine might be cost effective if we didn't have
> ubber sensor tech like auto nozzle electrified ink sensors, and auto
> paper align sensors, and auto load sensors, and on and on and on. If
> we had a mechanic way to align the paper and stop during a paper-jam,
> etc, it would be a start.

Part of the problem is that Epson et al. are designing for the mass- 
market consumer, even in their high-end systems.  So everything needs  
to just work, with as little interaction as possible.  I think that's  
one of their reasons for selling inks & papers that are designed to  
function well with their printers.  And it's also why they "dumb down"  
the system more and more over time.  The market wants more and more  
perfect prints, with no mess, jams, clogs, or color issues.  So that's  
what Epson engineers for.

As a printmaker, though, what I want is something totally different.   
After over a decade doing digital printmaking, I'd be very happy to  
get away from rolls of paper and go back to sheets.  So do away with  
roll feeding entirely, and give me a flatbed printer with:

	- A vacuum bed that would take a single sheet of paper of any  
thickness, texture, or size (depending on needs).  Allow that paper to  
be pin-registerable, either for re-printing on the same printer, or in  
cooperation with other printmaking techniques like letterpress or  
intaglio.  (Maybe a roll adaptor could allow for rolls to be used, but  
the prime "platform" would be a single sheet of paper.)

	- A 2D cross-arm plotter system that would accurately move the head  
assembly to any X,Y point on the paper.  It doesn't have to print from  
top to bottom like a regular roll-feed printer, but instead could take  
advantage of going backwards to do things like overprinting.

	- A head assembly that's mounted above the cross-arms.  It could take  
a number of heads: perhaps a simple 2- or 3-channel head for a simple  
B&W inkset, or multiple heads to do a complex B&W/color/GLOP/etc.  
inkset.

	- A flexible ink-supply system, either based on cartridges, bottles,  
or any other source.

	- A small computer that acts as a simple controller for the vacuum  
bed control, plotter, and heads.  This could be a basic single-board  
microcontroller setup, either an Arduino board or a small Linux  
system; it would have an Ethernet or WiFi controller aboard to  
communicate with the host software.

	- Software that runs on your desktop/laptop that communicates with  
the printer computer.  This could be simply a printer driver (ala QTR  
on the Mac), or it could be a standalone program that would read image  
files and then generate the commands to cause it to print.  The  
important part here is that all parameters of the printer are  
accessible and programmable: ink setup, cleaning/flushing techniques,  
dithering patterns, head channel, head position (X/Y on the plotter  
bed), and so on.

I believe the hardest part of this would probably be getting the heads  
themselves, along with the software knowledge on how to drive them.   
It may be possible to buy bare heads from a variety of manufacturers  
(see the links in the "Inkjet head design / Fixed head" section of http://en.wikipedia.org/wiki/Inkjet_printer) 
, but I don't know how easy that would be, especially in small  
quantities.

Another tricky bit would be the ink supply system, including the usual  
pumps & vacuums that get the ink flowing regularly enough to be used  
in the heads.

Anyone else interested in this kind of thing?

--John