Most industrial robots have six axes and are bolted down. Six axes let the arm reach a point from almost any angle, but the arm itself does not go anywhere. Its working envelope is a sphere around its base, and that sphere is fixed.
A seventh axis is a linear track that moves the whole robot. Along the floor, up a wall, or overhead on a gantry. The robot keeps its six axes of articulation and gains one of travel.
It sounds like a small addition. It changes what the cell can be.
Reach without buying a bigger robot
The obvious use is reach.
If your parts are spread over ten metres, the instinct is a robot with a longer arm. Longer arms cost more, weigh more, take more floor space, and lose payload capacity as they extend. A long-reach robot holding a heavy part at full extension is working near its limits, and accuracy suffers there.
A smaller robot on a track often reaches further, holds more at the far end, and costs less than the bigger arm would have. It also takes less room, because the track is a narrow strip rather than a wide swept circle.
One robot doing several jobs
The more interesting use is coverage.
A robot on a track can serve multiple stations along a line: load a machine, move down, tend the next one, move back. Where the rate allows it, one robot on a track replaces two or three fixed robots.
That is not only a capital saving. It is fewer controllers, fewer safety zones, fewer programs, fewer things to maintain, and less floor space. The payback on a robot doing double duty is substantially faster.
The limit is cycle time. Travel takes seconds, and those seconds are in your cycle. If your rate has no room for the robot to move between stations, you need the second robot. This is the calculation to run early, because it decides the whole layout.
Working in a cell that is not a circle
A fixed robot's envelope is a sphere. Real cells are rarely spherical. They are long, or L-shaped, or built around existing equipment that will not move.
A track lets the robot's envelope follow the shape of the work rather than forcing the work into the shape of the envelope. On cells built around equipment the customer already owns, that is frequently the only way the layout closes.
Mounting: floor, wall, or overhead
Floor-mounted is the default. Simplest to install and maintain, and it takes floor space.
Wall or elevated frees the floor and can reach over equipment, at the cost of a more involved structure and harder maintenance access.
Overhead gantry covers large rectangular areas and keeps the floor clear entirely. Most expensive, and the structural requirements are real.
There is a fourth option worth knowing: put the robot below floor level. On an aerospace material handling cell we ran the primary handler on a seven-axis track from a below-floor pit, specifically so maintenance vehicles and people kept their access to the cell during production, and so the customer could still intervene manually. That robot was a FANUC M-2000iA/2300 rated at 2,300 kg, and the pit was excavated to six feet and reinforced with rebar throughout to absorb the forces the machine generates.
What the track has to be
A seventh axis is not a rail the robot happens to sit on. It is a servo-driven axis under the robot controller, coordinated with the other six.
That means:
- Rigidity. The track carries the robot's mass plus the dynamic loads of acceleration and deceleration. Deflection at the track becomes position error at the tool.
- Repeatability along the full length. A robot that is accurate at one end and drifting at the other is worse than no track.
- Protection. Ways, drives and cable management have to survive the environment. In a weld cell that means spatter. In a machining cell, chips and coolant.
- Cable management. The dress pack now travels. It is a common failure point and worth over-specifying.
When a track is the wrong answer
- When cycle time has no room for travel. If the robot cannot afford to move, buy the second robot.
- When the stations are close together. If everything is within reach of a fixed robot, a track adds cost and a maintenance item for nothing.
- When the environment will destroy it. Some environments are hard on linear motion. Sometimes a fixed robot with a better-designed cell layout is the cleaner answer.
The question to ask
When someone proposes a bigger robot, ask what a smaller one on a track would do instead. And when someone proposes a second robot, ask what one on a track would do.
Neither is automatically right. But the comparison is cheap to run at design stage and expensive to discover afterwards.