Openprinter Wiki

What if the print head was actually....

what if actually there was a bar...

8.5 inches wide....

and it would print one row at a time...

lets say 300 DPI...

it would need to have 2550 places for ink to eject onto the page...

8.5 x 300

yeah, that's an interesting concept, back in the day it was called a line printer(bar/chain/drum, several variations)

the line printer would print one line of type at a time, very fast, big and noisy.

there was also a dot matrix variation of this that was used for check writing that would offset the printing bar left-to-right during the printing process to increase the resolution of the printer by overlapping the dot impact location.


unfortunately your idea of having 2550 micro holes spanning 8.5 inches means that the size of the ink ejection hole has to be small enough to accomodate the next non-hole space, followed by the next hole... for example, if you were going for a 300dpi printer, the holes might need to be 1/600th of an inch in size, and the material between ink ejection holes would also be 1/600th of an inch as well.

i think the problem with this theory is that it's expensive to manufacture a print head mechanism that has to span the entire width of the printing meduim.

an ink head that travels across the media width only requires a small number of ink holes, and by varying the speed of travel of the print head and the linear motion of the print media, you can adjust the precieved resolution of your printer.
i.e. fast print head spead + fast media speed = greater spaces between dot/low resolution slow print head speed + slow media speed = less space bettween dots or overlapping dos/high resolution


how about using a very small (gauge 33 or something like that) Hypodermic needle for the printerhead

A hypodermic needle... Really, are you serious?
How do you propose to align multiple 33GA needles parallel to each other, with all of the needle orifices perpendicular to the printing media?


For the un-initiated. A print head isn't just a bunch of tiny holes lined up next to each other. For a thermal based inkjet print head, there is the hole but also a heating element next to each hole and also a hole bunch of control circuitry that is there to power the heater and a small piece of piezo electric material that pushes out the individual drops of ink. So, unless we can come up with some type of home photolithography setup, then the project will have to use a comercially available print head. See picture below for an example, the black dots along the right hand side are the nozzles.

Picture from

i agree with your last entry, absolutely, the only way to produce this type of a thermal print head would be to have a home photolithography set up... not very practical or realistic.... however, there might be a possibility to produce a micro-fluidic device using shrinky-dink material and an inkjet printer. see this article:

also take a look at the researcher's paper here:


I like the concept but it seems that they are using plastics only. This would mean no piezo material for pushing out the ink and also no heat. I personally think that, unless we can find a source of comercially available printheads, we will have to go with a laser printer.

why not take an existing printhead/ink resevoir and encorporate that into an open printer design?

realize that designing all components from the ground up is unrealistic, this process might need to take a staged approach, using commonly available components which are too costly to manufacture, and replace them as the build progresses...

this is not uncommon in other sectors of manufacturing where OEM manufactured parts are used until manufacture of those parts can be internalized(think motors, engines, transmissions, etc). irregardless of whether the first printer is to be an inkjet or laser based printer, it's not reasonable to design and build the first gen device to be 100% home grown.

i'd like to suggest basing a first gen device to use an hp designjet cartridge(readily available thru aftermarket inkjet resellers), and resolve the printhead portion of the build to a 2nd or 3rd gen device. a quick search on alibaba( will yield plenty of manufacturers that can provide inkjet printhead components in bulk.


What if you used ten heads, so they'd all need to move only .85 inches

The printer should be designed in a such a way that it can use any print head or type of print head. In other words, make the printer -> print head interface generic enough that it can be replaced with a laser system, ink jet, dot matrix style print, etc. That would be a much better idea than directly using a commercial printhead, which would make the process less "open".

This could consist probably of an intelligent print head/head connection module with some minor circuity (simple avr/pic microcontroller perhaps) to place the burden of software decoding on the printhead maker vs. the one who makes the rest of the printer mechanics. The printer's communication with the print head should be simple enough that interfacing can be done with analog circuitry if desired, but structured well enough to make communication with modern microcontrollers possible.

Printer <--> Print Head OR t Printer <--> Printhead interface module <--> Print Head

There may be a somewhat simpler way of implementing this idea, but it doesn't come too mind right now.