Machine Design

A straight path to clearer images

Artifact lines can ruin digital scans. Linear actuators that don't run true are one likely culprit.

The correct diagnosis may depend on a clear X-ray image.

Composite bearings are mechanically simpler than conventional built-up bearings. For example, a typical 32-in. stroke built-up bearing requires about 300 individual components. The composite Dolphin guide has just two.

Large-format digital image scanners and high-end inkjet printers produce clear architectural drawings, maps, medical x-rays, posters, billboards, signs, and other graphics. In all cases, scanned outputs should be free of artifact lines caused by the scanning process itself. Key to clearer scanned images are linear actuators capable of consistent motion.

Linear actuators have two basic parts, a screw-drive mechanism and a linear guide bearing. Both parts contribute to overall motion integrity. For instance, consistent motion requires that torque be evenly applied to and by the drive mechanism. Torque can vary with rail surface finish, lubricant type, drive-mechanism type (leadscrew, ball screw, etc.), drive-mechanism support-bearing type (ball, angular contact, etc.), cage creep in rolling elements, drive-mechanism-to-carriage connection, motor type and motor coupling, and other factors.

Assuming the drive mechanism is up to the task, next consider the linear guide bearing. Here, accuracy and repeatability quantify motion consistency. Positional accuracy is the difference between a slide's desired and actual position, in any plane or direction, not just along the line of action. Repeatability equals the variation between multiple moves to the same position. These deviations can be measured from end to end (pitch), side to side (roll), or twisting moments (yaw), in either one direction or both. Such errors typically trace back to the manufacture of individual components. Each part is made to some tolerance range, but assembling them into a product can contribute to tolerance stack-up and larger error margins. In the case of imaging and scanning, excessive and unpredictable movement of slide carriages produces artifact lines on outputs.

Built-up rolling ball slides can be preloaded to remove excess play. But preloading typically distorts the assembly which tends to worsen motion variance and shorten the life of rolling elements and raceways. The upside to rolling-ball slides is low friction. Such designs typically maintain a coefficient of friction (COF) of about 0.05 when properly serviced.

However, low COF does not necessarily translate into consistent motion. This is because most linear bearings contain 40 or more hardened (loose) steel balls that continuously move in and out of the load zone while counterrotating and banging against each other, the raceways, shafting, and the bearing ends. Each ball movement has the potential for stick-slip, galling, and vibrations that feed back through the assembly. Metalto-metal contact of balls with shafting magnifies vibrations. Add to that potential damage to individual balls from shock loads (intentional or inadvertent), corrosive wear, or contaminants. Flat spots, skidding, and galling also can compromise consistent, smooth motion.

One alternative to built-up rolling ball types are two-piece composite linear guides, such as the Teflon-based FrelonGold Dolphin Guide from Pacific Bearing Co., Rockford, Ill. The composite moleculary bonds to the carriage, eliminating internal moving parts, metal-to-metal contact, and the associated vibration. Because Teflon is relatively soft, it lets contaminants embed into the material surface rather than damage the bearing or raceway on which it rides. And because they need no preload, composite bearings tend to produce more consistent motion. Composite bearings have a higher coefficient of friction than rolling ball designs (0.125), though, like rolling-ball types, friction remains constant over the life of the bearing.

Composite linear guides in this case run truer than rolling-ball guides. A theoretical zero was set at rail ends and measurements were taken of the 32-in.long guides at 2-in. intervals. Five different assembled bearings of each type were measured. Results are averaged values.

Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.