SCARA robots cover a 3–20 kg payload band, a 400–1,200 mm reach band, and ±0.01–0.02 mm repeatability, with standard pick-and-place cycles in the 0.3–0.5 s window per recent OEM guidance [S1][S2][S3]. Sizing them is straightforward on paper and punishing in practice: underspec the payload by 10–15% and you eat joint wear inside 12–18 months [S1].
Engineers who already own a SCARA robot cell run them at 70–80% of rated cycle speed, measure the largest pick-to-place distance, and add a reach margin of 50–80 mm to keep the arm out of its slow outer band [S1][S2]. The rest of this piece walks through the sizing math, the variant decision tree, the comparison lines against cobots and 6-axis arms, and the failure modes that show up when those rules are skipped.
Payload Bands: 3 kg, 6 kg, 10 kg, 20 kg
Payload is the first filter and the easiest to underspec. The rule is part weight plus gripper weight, and then add a 10–15% safety pad on top: under-speccing the rating drives premature joint wear and repeatability drift inside 12–18 months in field data [S1]. Production models split into four common bands: 3 kg, 6 kg, 10 kg, and 20 kg, with 3 kg fitting 3C and PCB handling, 6 kg covering connector insertion and screwdriving, 10 kg handling mid-weight medical and tray transfer, and 20 kg taking on automotive sub-assembly and heavier packaging [S1][S2].
The same 2–20 kg window is confirmed by independent integrators, who add that 2 kg mini-SCARAs exist for ultra-light electronic component placement, while anything above 20 kg or beyond 1,200 mm reach is no longer SCARA territory and shifts to 6-axis selection logic [S5]. For buyers cross-shopping a SCARA robot against a 6-axis arm, payload is usually the first spec where the decision forks: if the part plus gripper plus fixture mass lands under 20 kg and the work stays roughly planar, stay in SCARA and capture the cycle-time and price advantage.
Reach Options: 400 mm to 1,200 mm
Reach is the maximum horizontal distance from the robot base center to the tool center point, and industry-wide it runs from roughly 400 mm to 1,200 mm [S1]. The right way to size it: measure the largest horizontal distance between pick point and place point in the cell, then add 50–80 mm of margin, because an arm operating at the edge of its reach trades cycle time for distance and also runs the motors harder, which shortens service intervals [S1][S2]. Production SKUs cluster at 400 mm, 600 mm, 700 mm, and 1,000 mm, with a 1,200 mm ceiling in the long-reach models [S1][S2].
Reach ties directly to cell footprint. A 400 mm arm fits an inline bench cell under 600 x 600 mm; a 600 mm arm covers a small conveyor or two-station bench; 700–800 mm covers typical pick-and-place trays; 1,000 mm and above opens multi-lane conveyor or pallet-style cells [S1]. If the planned Z stroke exceeds roughly 200 mm or parts arrive at varying heights, reach is not the binding constraint, and 6-axis kinematics usually make more sense even within the SCARA payload band [S1][S5].
SCARA vs Cobot vs 6-Axis: Decision Map
The selection call is not really SCARA-versus-cobot, it is geometry-versus-cycle-versus-cost. The four questions integrators ask are: is the motion planar, what cycle time must be hit, does the end-effector need to approach at an angle, and what is the payload-and-reach envelope [S5]. If the answer to question one is yes, two demands a sub-0.5 s cycle, three is no, and four is under 20 kg plus under 1,200 mm, the cell is a SCARA cell every time, with cost and speed advantages over 6-axis on the same work [S1][S5].
For comparison, the three architectures line up roughly as follows on the criteria that matter at selection: SCARAs cover 2–20 kg payload, 400–1,200 mm reach, ±0.01–0.02 mm repeatability, and 0.3–0.5 s cycle; cobots add 6-axis orientation freedom and human-safe operation but trade away cycle rate; 6-axis industrial arms cover the same orientation freedom at higher speed and higher cost but still lose to a SCARA on a purely planar move because they carry more linked mass through more joints [S2][S5].
Cycle Time, Repeatability, and the 70–80% Rule
Standard pick-and-place cycle on a SCARA sits in the 0.3–0.5 s window for production models, with a top-end index rate above 40 cycles per minute on a known planar move [S2][S5]. Vendors do not recommend running the published cycle rating flat out: integrators routinely size and program at 70–80% of the rated cycle to leave headroom for acceleration profiles, part variation, and vision-correction steps that get added later [S2].
Repeatability is one of the SCARA's sharpest specs. Production units hit ±0.01 to ±0.02 mm at the tool center point, which is the reason the architecture dominates insertion and screwdriving at 3C and connector plants, and the reason it pairs naturally with vision-guided correction when upstream feeder tolerance loosens [S1][S2]. The mechanical compliance in the X-Y plane is what allows that: small lateral misalignments during insertion are absorbed by the kinematics, which lowers fixture and feeder precision requirements, and in turn lowers tooling cost and changeover time compared with systems that need zero lateral error before insertion [S2].
Servo Integration and Drive Architecture
Modern SCARAs are driven by integrated servo modules: a servo motor, drive, encoder, and control electronics collapsed into a single housing, with one module per joint, typically the base rotation axis, the shoulder, the elbow, and the vertical Z-axis [S4]. Putting the drive on the motor shortens the signal path, reduces electrical noise, and tightens the closed-loop response, which is exactly what fast acceleration, high positional accuracy, and compact joint design all demand on this robot class [S4].
The practical selection point is not whether to specify integrated servos, because every current production SCARA ships with them, but what torque, encoder resolution, and communication protocol the joint modules expose to the system integrator [S4]. For buyers who also specify a linear guide for the Z column or a crossed-roller guide at the wrist, the servo integration story is the same: it is the joint module that decides cycle, repeatability, and service interval more than the mechanical structure it sits on.
Who Should NOT Pick a SCARA
The clear failure cases are tall vertical lifts, parts that must be approached at a compound angle, and work envelopes past 1,200 mm reach or 20 kg payload [S1][S5]. If the cell needs orientation freedom mid-flight, for example, re-orienting a part between bowl feeder and assembly, a SCARA would need a re-orient fixture and a second pick, which a 6-axis wrist does in one move, and the integrator's rule is that the 6-axis earns its keep there on motion geometry alone [S5].
Another case where SCARA is the wrong pick: 3D welding, painting, deburring, or any path that needs an arbitrary wrist pose; these are 6-axis work and trying to force them into a SCARA cell by adding rotary stages just adds BOM cost and cycle time without solving the geometry problem [S1][S5]. Buyers cross-referencing a SCARA shortlist against construction machinery and equipment cells should also note that SCARAs are not rated for outdoor or heavy-vibration duty, and any cell with a washdown or cleanroom requirement must be matched to a cleanroom-rated SCARA variant rather than a standard model.
Application Spread and Market Sizing
The 2026 SCARA market is sized at USD 12.45 billion, up from USD 11.36 billion in 2025, with a forecast USD 18.08 billion by 2030 at a 9.74% CAGR, driven by electronics, automotive EV, pharmaceuticals, and food processing [S3]. Electronics and semiconductors are the dominant vertical, holding 42.1% of the market in the most recent segmentation, which is consistent with the architecture's natural fit for 3C assembly, semiconductor handling, and connector insertion [S3].
Other named verticals are automotive and EV manufacturing, pharmaceuticals and medical devices, food and beverage, and metal and machinery, with pick-and-place, precision assembly, dispensing, gluing, soldering, screwdriving, fastening, and vision-guided inspection as the recurring application list [S3]. For integrators already running a SCARA robot fleet, the next planning step is usually to map the existing part-and-gripper mass against the 3 / 6 / 10 / 20 kg bands and then map the cell footprint against the 400 / 600 / 700 / 1,000 / 1,200 mm reach tiers, because that is the only sizing pass that lines up cycle, payload, and reach against actual production data in one sheet.
Trackable signals for the next 6–12 months: tighter 1,200 mm-reach SCARA SKUs entering the catalog as Tier-1 EV battery and solar panel lines spec them for tray handling, and integrated servo vendors pushing higher-resolution encoders into the J1 and J2 joints to defend ±0.01 mm repeatability against 6-axis cobot cells that quote comparable accuracy on lighter payloads. The 2026 release of Codian's SCARA kinematics inside the B&R MCR Performance Analyzer also points to workspace-simulation and joint-space planning tools becoming a standard part of the SCARA selection workflow, not a custom integrator add-on [S6].
See also our earlier report, Steam Trap Types: Operating Principles, Spec Limits, and Selection Map.