What degrees of freedom means
A degree of freedom is one independent way something can move. A drawer that slides in and out has one. A door on a hinge has one. A robot arm’s degrees of freedom, often shortened to DOF, are the number of independent joint motions it can make. In practice, for a typical arm built from simple joints, it usually equals the number of motors driving those joints.
The count matters because it sets what the arm can do. More robot arm joints mean more possible positions and orientations, but also more parts, more weight, more cost and more to control.
Joint types
Two kinds of joint account for most hobby arms:
- Revolute joints rotate about an axis, like an elbow or a hinge. They are by far the most common, since servos and geared motors provide rotation directly.
- Prismatic joints slide in a straight line, like a telescoping rod or a lead screw carriage. They appear in gantry-style and Cartesian arms.
Some joints rotate about the length of the arm (a twisting or “roll” motion), while others swing it up and down (a “pitch” motion) or side to side (a “yaw” motion). A shoulder is often built from a base rotation plus a pitch joint.
How many DOF are needed
To put a hand at any position in 3D space, you need at least three DOF. To also control which way it points, you need three more, which is why general-purpose industrial arms commonly have six. In 3D space, a rigid object has six degrees of freedom: three for position and three for orientation.
That does not mean every arm needs six. The right number depends on the task:
- 2 to 3 DOF: enough to pick up objects at a known height and place them, or to draw on a flat surface.
- 4 DOF: common for hobby desktop arms. A base rotation, two joints to set reach and height, and a wrist for the gripper’s angle can handle many pick-and-place jobs.
- 5 DOF: adds the ability to tilt and rotate the gripper for more varied approaches.
- 6 DOF: allows any position with any orientation within reach, at the cost of complexity.
A gripper’s open and close motion is usually counted separately as an end-effector function, not as one of the arm’s positioning axes, although conventions differ between sources.
An arm with more DOF than the task needs, sometimes called redundant, can reach the same point in several ways. That gives flexibility around obstacles but makes the control problem less clear-cut.
Reach and workspace
Reach comes from link lengths and joint ranges. The set of points the hand can touch is the workspace. Its shape depends on the geometry:
- A rotating base with two pitch joints sweeps out a dome-like region.
- A Cartesian (gantry) arm with three sliding axes covers a box, which is simple to understand and program.
- A SCARA-style arm with vertical pivots moves quickly across a flat plane.
Joint limits also cut away parts of the workspace, and the arm’s own body can block it. The edges of the workspace are the least accurate and weakest regions.
Torque grows with reach
The farther a weight is from a joint, the greater the turning force that joint must resist. A shoulder joint carries the arm’s own weight plus its payload, multiplied by how far out they sit. Rough sizing is a matter of arm weight and payload times distance, with a generous margin. For a hobby arm, that is why:
- Light links matter. Every gram at the end of the arm is leveraged against the shoulder.
- The heaviest motors belong near the base, sometimes with belts or linkages carrying the motion outward.
- Stated payload on cheap kits is usually for the arm in a favourable pose, with much less at full stretch.
- Holding a pose costs power. A motor holding an arm against gravity draws current continuously and can overheat.
Gearing is the usual answer to torque needs, but gearing adds backlash, which makes positions sloppy. The trade-offs are covered under gears, belts and torque. The motor families behave differently too; hobby servos are convenient but vary in precision and strength, as the motor type comparison explains.
Why extra axes complicate control
Controlling one joint is easy: command an angle. Controlling the hand’s position is harder, because it depends on all joint angles at once.
- Forward kinematics takes the joint angles and works out where the hand ends up. It is a straightforward calculation.
- Inverse kinematics goes the other way: given where you want the hand, find the joint angles. This is harder. There may be several solutions, such as elbow up or elbow down, or none at all if the target is out of reach.
- Singularities are arm poses where the arm loses the ability to move in some direction, for example when fully stretched out, and small hand movements demand very large joint movements.
- Coordinated motion needs all joints to start and finish together if the hand is to travel in a straight line, which requires planning velocities, not only end positions.
- Errors accumulate. A small angle error at the shoulder becomes a larger error at the hand, and each extra joint adds its own backlash and tolerance.
Each added axis multiplies these difficulties. This is why a well-built 3- or 4-DOF arm often does a more reliable job than an overambitious 6-DOF one.
Feedback and calibration
Hobby servos report no real position back to the controller, so the program assumes the arm went where it was told. That is open loop, and it works until the arm is bumped, overloaded or powered up in an unknown pose. Better arms add encoders or potentiometers on joints, and many use limit switches so the arm can find a known “home” position at start-up. After homing, the controller can trust its count.
Smooth motion also benefits from feedback control rather than jumping to targets, and PID control is the standard approach for holding joint positions accurately.
Building and testing a hobby arm
- Design the parts first. CAD modelling lets you check reach, clearances and joint ranges before printing anything.
- Choose materials sensibly. Printed arms depend on filament materials with enough stiffness, and on print tolerances at the joints so they turn freely without wobble.
- Mount carefully. Joints and brackets see leverage forces, so solid mounting hardware matters.
- Plan power. Several servos moving at once can pull a large combined current. Good hobby power planning, a supply rated for peak load, and a proper driver arrangement prevent resets and overheating.
- Try it in simulation. Checking movements virtually first reduces crashes.
- Keep hands clear. Arms can pinch, and a joint can swing unexpectedly when power is applied or removed. Plan an emergency stop and think about which way the arm will fall when power is cut.
Start slow, test with no payload, and extend the design one axis at a time. Understanding where the complexity comes from helps you choose just as many joints as the job needs.