This is the first component of a what I hope to become a completely unattended yarn dyeing system.

The problem

Yarn is available in many materials and colors, but somehow it’s still troublesome to find something that’s just right for a project. Hand-dyed yarn gives beautiful and finely controllable results, but is not easily repeatable or controllable. I want to build a machine that takes a cone of yarn and produces dyed, washed, and dried balls of beautiful yarn, repeatable indefinitely.


The yarn path defines the gross structure of the machine:

  • The machine pulls yarn off the cone and feeds it into a dye bath. The heated dye and fixative mix is absorbed into the yarn.
  • (possibly the yarn should be heated additionally at this point)
  • The yarn continues into a water bath, where it is rinsed from unfixed dye.
  • The yarn passes through multiple jets of hot air, which dry it.
  • The yarn is then pulled onto a ball winder or cone winder.

In addition, a color mixing system is needed, that can continuously feed precisely controlled volumes of each dye. The passing yarn slowly depletes the dye pool, allowing newly added dye to create smooth color transitions. The dye must be either heated before being deposited in the dye pool. or heated inside the dye pool itself.

The pump

The central part of the dye mixer is a set of pumps. These need to fulfill several requirements: -They must be positive displacement pumps. That is, the pump needs to move known quantities of dye rather than just creating a pressure differential. -They must be easy to clean. Dye is messy. Contamination is unpleasant. -They must be acid-resistant. The most useful dye type has rather low pH. -They must be liquid-proof.

Peristaltic pumps, used in the medical industry and food processing, would fulfill all of these requirements. That’s what I decided to try first. A peristaltic pump has a number of rollers that make contact with the outside of a tube, compressing said tube against a wall. As the rollers move, the compressed section of tubing is pushed along the tube, sealing against itself and transporting a fluid forward. The major disadvantage of this is that the output pulses, having lower pressure during the open sections and higher pressure during the compressed sections. To reduce the effects of said pulsing, I’m using five separate rollers to make it pulse at five times the frequency.

Design iteration 1

The pump consists of a rotor and a body. The rotor itself has a top and bottom part, and 624ZZ bearings fit between the top and bottom and are held in with M4 bolts. The tubing is held between the motor body and the bearings on the rotor. The rotor has a nut trap for an M3 nut. An M3x10 bolt attaches the rotor to the 5mm motor shaft. The body is built around a NEMA14 motor. It has 4 mounting holes for the motor bolts and a hole for the motor shaft, as well as mounting holes for attachment to structural components of the machine.


-The tubing is not compressed sufficiently by the bearings - increase the distance from bearing to rotor. -The tubing gets pulled tight between the bearings - increase the number of bearings from 3 to 5 to reduce the chord length. -The NEMA14 motor does not have enough torque - even a NEMA17 barely does. Tricky. A gear would be in order here.

blog comments powered by Disqus


13 July 2012