Bench Talk

Robots, both in the form of companion “humanoids” or as articulated production and assembly arms, are getting a lot of attention these days, and rightfully so. Precision motor control, enhanced vision and image processing, increased central processing unit (CPU) power, artificial intelligence (AI)-driven decision-making, and advanced grip technologies are making discussions about robots even more common in speculative-journalist stories as well as in the hard-working reality of everyday operations in various industries. This trend makes sense, considering these highly effective robotic models work well, work hard, don’t complain, don’t need breaks, and accomplish things that humans have difficulty doing.

You’ve seen the pictures and videos ranging from auto-assembly lines to hamburger-flipping units, and it’s impressive, for sure. Most photos of these robots show them to be sleek, streamlined, and clean, and the technical discussions about their design and construction focus on algorithms, performance, and capabilities.

However, for engineers and designers, what is just as impressive in larger-capacity robot design is the major shift away from the use of hydraulic power (with all its headaches) to electric-motor prime movers, especially over the last few years. Electric drives are cleaner, easier to control, and pose no risk of oil spills if a hose cracks. They are also easier to test and maintain, even though hydraulic units offer higher power density/volume. (To be fair, the hydraulic systems are getting smarter, too, with all sorts of electronics to control and oversee their actions—though the prime mover is still a hydraulic cylinder and piston.)

But there’s one aspect of an electric-powered robotic implementation that’s rarely mentioned: This aspect is the essential cables that travel from the control unit and power sources to the motors and sensors of the arms, joints, and grips. While these cables are usually tucked inside the frame of humanoid-type robots and are thus invisible, that’s generally not the case for the much-more widely used assembly/production units. Let’s face it: Exterior cable bundles are not the exciting topic, and visually, they are bulky and somewhat unappealing to most folks. (Did you ever notice that there are no alternating current (AC)-line cords shown in those glossy “house decorating” photos and only the table lamps are shown?) 

In the real-world, of course, a robot’s power and control cables are essential and usually attached along the outside of its arm, as there is no room inside the arms for these fairly bulky cables. However, this very practical practice leads to another problem: How to “dress” the cables so they stay where they should, allow the necessary range of movement, and don’t tangle.

The obvious solution is the simple one: Just use cable ties, special tape, or something similar to attach them to the articulating arms. That should do it, but it often doesn’t do it well. The fact is that in advanced robotic arms with many degrees of freedom, the cabling can become twisted, flexed, strained, or abused in excess, leading to intermittent performance and outright failure.

I hadn’t given this subject much thought until I saw an article in Tech Briefs, entitled “The Less is More Approach to Robotic Cable Management,” which discussed the issues and weaknesses of “obvious” solutions. Although it was written by a vendor of cable-protection raceways and enclosures and (as such) is perhaps biased, many of the points in the article are valid. The article even gives guidelines on where you should divide the overall cabling path into segments that can then be individually protected.

Long story short: If you are doing any work on highly articulated, multi-axis robotic arms, be sure to spend sufficient time and energy at the “forefront” of the project to properly address the mundane and unglamorous topic of routing cables. They are as essential as any other element for achieving and maintaining a reliable performance level.

Have you ever tried to track down an intermittent problem that is due to a broken cable standard? It’s frustrating, messy, and aggravating—and drains the energy out of the design team, debug specialists, and field engineers. However, it’s also an opportunity to leverage the expertise of others. For example, the robot cable must flex over and over, so why not see what the elevator folks have done for their hoistway cabling. Even though they don’t have the robotic-arm problems related to multiple degrees of freedom, they do know about building a cable that is flexible, carries its own weight plus significant power, and has a rugged housing.