Un-building an ink jet printer
In the process of building my 1"x42" strip sander and the pantorouter trade show display, I used up most of the 8 mm shafting I had kicking around. HP ink jet printers like this all-in-one are a good source, so I took this one apart. I found it on the curb a year ago, figuring I might still use it if I found a power adapter to go with it.
Every time I take apart an old printer, I'm always amazed at the mechanical ingenuity inside, so I figured it was worth writing about.
The scanner part sits on top of the printer. Popping it apart, I'm holding the entire guts of the scanner. It's a multi coloured LED light strip, a sensor strip, and a small DC motor with some gearing on it to drive it along a rack gear on the bottom of the scanner enclosure. Scanners used to be much more complicated affairs, with stepper motors, fancy optics with mirrors, and polished shafts for gliding along. It's no wonder that scanners have become so cheap. Last time I wanted to buy a printer I bought a printer/scanner, because that was the cheapest option!
The plastic enclosure forms a tub around the printer mechanism. The electronics is just a small board in the corner. No visible power transistors or power ICs. I think the small DC motors this printer uses are more efficient than the stepper motors printers used to use, so less need for power electronics.
Here you can see the polished 8 mm shaft that the print head slides along, the timing belt that drives it (my finger is pushing on it), and a transparent plastic ribbon with very fine black bars on it for position sensing.
A close up of the plastic strip with the fine black bars, about six per millimeter. The strip is stretched taut across the travel of the print head. An optical sensor on the print head reads the strip. Using quadrature encoding, the printer knows exactly how far and which direction the print head has moved.
This allows the printer to establish the print head position with very high repeatability, more precisely than using a stepper motor and timing belt. This is necessary for high resolution bidirectional printing.
The motor driving the print head is just a small DC motor. DC motors are faster and more efficient, but positional feedback is needed to achieve precise control. But position feedback is already necessary for the print head, so might as well use it to control the DC motor too.
On the left side of the printer are two rubber pads that move up when the print head pushes against a pin. These pads cap the printer cartridges when the print head is parked to keep them from drying out.
The left-most white gear in the photo has a transparent disk with a very fine set of lines around it. An optical encoder (on the brown circuit board below) detects precise movement of the paper advance, similar to how the print head position is detected.
I had been puzzling over what activates the squeegee mechanism on the right. I discovered how it worked after I broke it. There is a small gear (A) that activates the squeegee, and this is pulled up by a piece of plastic (B) which I broke. This plastic was part of the assembly at the end of the feed roller. When the print head moves all the way to the right, it turns this mechanism, which causes part B to pull up gear A to engage it with the gear on the feed roller. So the paper advance motor powers the squeegees. This means the squeegee can only be used between sheets of paper.
The paper feed mechanism has a lot of gears, and these, I think, engaged by running the paper feed motor backwards.
Here are the pieces I kept. The main paper feed is also on an 8 mm shaft (bonus), while the other rollers are on 6 mm shafts. I used up most of my 6 mm shafting when I made lots of followers for my pantorouter, so it was nice to replenish that stock too.
What's inside a Sonicare toothbrush (video only)
Strip sander (using
8 mm shafts)
Making pantorouter followers (6 mm shafts)
Reviewing the cheapest USB scope I could find online (video only)
Scavenging wood from the curb, timelapse
To my Woodworking website.