Motionwell integrates SCARA robots for electronics manufacturing in Singapore. Delivered work includes a vision-guided sensor panel assembly system built on ABB SCARA at ±0.01 mm repeat accuracy, and a secondary packaging line in delivery through 2026 using four Yamaha SCARA robots. This guide covers why SCARA robots suit electronics assembly, how they compare with 6-axis, collaborative and cartesian machines, how to size one, and the cases where a SCARA is the wrong answer.
SCARA stands for Selective Compliance Articulated Robot Arm. The “selective compliance” means the arm is rigid in the vertical axis (Z) but compliant in the horizontal plane (X-Y). That makes SCARA robots suited to work where parts are picked from one horizontal surface and placed on another, which describes most electronics and semiconductor assembly operations.
What is a SCARA robot?
A SCARA robot has four axes: two rotary joints in the horizontal plane (J1, J2), one vertical linear axis (Z), and one rotary axis at the wrist (J4). Those four axes cover the shape of most electronics assembly tasks, pick a component up, move it horizontally, rotate it to the correct orientation, put it down.
| SCARA axis | Motion | Function in electronics assembly |
|---|---|---|
| J1 (shoulder) | Horizontal rotation | Large horizontal repositioning |
| J2 (elbow) | Horizontal rotation | Fine horizontal positioning |
| Z (vertical) | Linear up/down | Pick-up and place-down motion |
| J4 (wrist) | Rotation around Z | Component orientation correction |
The mechanical advantage is the split of duties. Horizontal loads are carried by the two rotary joints, while the vertical axis only deals with gravity and insertion force. That separation is what allows fast horizontal motion together with high vertical stiffness, and it is why the same arm that flies across a tray can still press a connector home without deflecting.
What a SCARA cannot do is tilt its tool. The wrist rotates about Z and nothing else, so the approach vector is always vertical. Every argument for and against SCARA comes back to that one constraint.
How does a SCARA compare with 6-axis, cobot and cartesian robots?
The figures below are the ranges published across vendor datasheets for each robot class, not measurements from any one machine, and they are there to show relative strengths rather than to specify a purchase.
| Factor | SCARA | 6-Axis | Cobot | Cartesian/Gantry |
|---|---|---|---|---|
| Relative speed (pick-and-place) | Fastest | Moderate | Slowest (force-limited) | Moderate |
| Datasheet repeatability | ±0.01-0.02 mm | ±0.02-0.05 mm | ±0.03-0.10 mm | ±0.01-0.05 mm |
| Payload range | 1-20 kg | 3-300+ kg | 3-25 kg | 5-100+ kg |
| Reach | 200-1000 mm | 500-3000+ mm | 500-1300 mm | Custom (frame-dependent) |
| Footprint | Very compact | Moderate | Moderate | Large (frame structure) |
| Relative cost at equal payload | Lowest | Highest | High | Varies with travel |
| Orientation flexibility | Limited (4 axes) | Full (6 axes) | Full (6 axes) | Limited (3-4 axes) |
| Best application | Pick-and-place, insertion, dispensing | Complex 3D paths, grinding, deburring | Human collaboration, flexible tasks | Long travel, heavy payload |
| Programming effort | Low-medium | Medium-high | Low (hand guiding) | Low |
| Safety fencing required | Yes (speed) | Yes (speed and force) | Not always (force-limited, after risk assessment) | Depends on speed |
Two cautions on reading that table. Repeatability is not placement accuracy: it describes how tightly the arm returns to a taught point, and it says nothing about gripper slip, part tolerance, fixture wear or thermal drift, which usually dominate what you measure on the finished assembly. And the cobot row is not a licence to skip safeguarding, force limiting still has to be proven by risk assessment, which is the subject of our guide to collaborative robot safety standards.
Choose SCARA when parts come from a tray, feeder or conveyor, move horizontally, and land on an assembly, test fixture or packaging tray, at part weights well under the arm’s rating, with cycle time under pressure.
Why are SCARA robots so common in electronics assembly?
Electronics manufacturing has a few characteristics that line up with what this arm is good at.
Speed
Electronics cycle times are counted in seconds. SCARA robots do not compete with dedicated SMT placement heads for chip-scale components, and it is a mistake to buy one expecting that. Where they win is the tier above: connectors, shields, sub-assemblies, carriers, boards and finished units moving between stations.
Beware the datasheet cycle time. It is measured on a fixed gate-to-gate move with a dummy load and nothing else happening. Your real cycle adds vision acquisition and processing, settling time before the vacuum releases, gripper actuation, the handshake with the PLC, and any wait on the conveyor or fixture. Those additions, not the arm, are usually what decides whether the station makes rate, which is why we time them on the actual parts before quoting a throughput figure.
Precision
Electronics assembly usually asks for placement in the hundredths of a millimetre. A SCARA’s mechanical repeatability is comfortably inside that band, so the error budget is spent elsewhere: on how well the part is located before the pick, on gripper repeatability, and on how well the camera is calibrated to the robot frame. That is why vision guidance, covered under machine vision inspection, matters more than another decimal place on the robot spec sheet.
Compact footprint
Floor space in Singapore is expensive. A small SCARA mounts on the machine frame itself and occupies little more than its own base plus its swept envelope, which is why several stations fit side by side on one bench-scale line. On a special purpose machine built around a rotary or shuttle index, that difference decides whether the cell fits the space the customer has.
What actually drives the cost of a SCARA cell?
The robot is one line on the bill among several. Cost drivers, roughly in order of how often they surprise people:
- End-of-arm tooling, especially if the cell handles several part variants and needs quick-change tooling
- Part feeding, engineered against your specific part, not ordered from a catalogue
- Vision: camera, lens, lighting, and the engineering hours to make it repeatable under real factory light
- Fixturing and nests, which is exactly the cost vision guidance is meant to reduce
- Safeguarding, guarding structure and the assessment work behind it
- Integration, commissioning and, in regulated work, validation documentation
Against that list, choosing between two comparable arms changes the project total less than the choice appears to deserve. Specify the cell, then choose the robot.
Where does a SCARA earn its place on an electronics line?
Panel assembly and component insertion
Our sensor panel assembly project is the reference implementation: ABB SCARA robots pick components, align them under vision feedback and insert them into panel assemblies, with ±0.01 mm repeat accuracy at the arm. The build is written up in the SCARA panel assembly case study.
Vision guidance is what removes the precision tray. Parts can be presented in approximate positions, because the camera measures the actual location and orientation for every pick and the controller corrects the move. That lowers tooling cost and shortens changeover between product variants.
How does a pick-and-place robot work with feeders and trays?
The typical SCARA job in electronics is moving components from a feeding system, vibratory feeder, tray stacker, tube feeder, flexible vision feeder, to an assembly position.
| Parameter | Typical range | What determines it |
|---|---|---|
| Cycle time | Sub-second to a few seconds per pick-place | Travel distance, settling time, vision processing, PLC handshake |
| Part weight | 1-500 grams | Component and carrier weight |
| Placement accuracy | Hundredths of a millimetre | Part tolerance, fixture quality, camera-to-robot calibration |
| Part presentation | Tray, vibratory feeder, tape and reel, conveyor | Part geometry, feed rate requirement, cost |
| Tooling type | Vacuum cup, vacuum pad, mechanical gripper | Part surface, weight, shape |
Can a pick-and-place robot dispense or solder?
Placing a part and making the joint are separate operations. The placement robot puts the component where it belongs, and the joint is formed after that by a reflow, wave or selective soldering process, so soldering runs as its own station rather than inside the placement move. A SCARA can carry a dispense valve, solder iron or wire feeder along a path that stays in the horizontal plane. For dispensing, what matters is Z-axis accuracy and how the arm settles, vibration during the bead shows up directly as inconsistent width or dot size. Direct-drive joints, with no belt or gear backlash in the path, give the smoothest motion here, for the same reason discussed in our comparison of servo and stepper motor drives.
Test and inspection station loading
SCARA robots load and unload test fixtures and measurement stations: pick from an input conveyor or tray, seat the part in the fixture, wait for the test, retrieve it, sort to pass or fail. Loading is often the easy half, the hard half is seating a connector or contact repeatably enough that a test failure means a bad part rather than a bad insertion.
Board handling and conveyor transfer
Vacuum tooling contacts the board surface without clamping, which avoids stressing populated areas. Where the line cannot stop, conveyor tracking lets the arm pick from a moving belt.
Which SCARA platforms does Motionwell actually run?
Motionwell is a system integrator and machine builder, not a robot manufacturer. We buy arms and build the machine around them, so the platform question is settled by the application and by what the customer already supports.
ABB
ABB SCARA is the platform behind the sensor panel assembly cell, and ABB is the robot brand we integrate most across the shop, from SCARA up to the 6-axis arms used in our aerospace force-controlled grinding work. Where a customer already runs ABB, staying on one controller family and one offline programming environment keeps the maintenance and spares story simple for their team.
Yamaha
Yamaha SCARA is our other running platform. Four Yamaha SCARA arms are being installed on a secondary packaging line in delivery through 2026, work that sits close to our pharmaceutical packaging practice. Packaging duty is a different problem from panel assembly: reach and speed are set by carton and case geometry rather than by placement accuracy, and multiple arms have to be sequenced against the line rather than optimised individually.
How we choose between them
We do not lock customers into a brand. Selection runs on criteria, in this order:
- Cell geometry. Furthest pick to furthest place, plus the intrusion of fixtures and feeders, sets minimum reach, measured on the layout, not estimated.
- True payload. Part plus gripper plus manifold plus cable plus anything else on the arm, with margin for the moment of inertia at full extension, which is what limits acceleration in practice.
- Z stroke and mounting. Insertion depth, tray stack height and whether the arm is floor, bench or ceiling mounted.
- Controller fit. Whether the arm talks cleanly to the customer’s PLC and network, most of our lines run Allen-Bradley, Siemens, Omron, Mitsubishi or Beckhoff control.
- Vision toolchain. How the camera calibrates to the robot frame and how offsets are sent. We integrate Keyence and Cognex.
- Local support. Spares lead time and service coverage in Singapore, which decides recovery time on a line that is down.
Motionwell has integration experience on customer-supplied robots from other brands where the customer already owns the arm. What we will not do is tell you a platform is our standard when it is not on our shop floor.
When is a SCARA the wrong choice?
The failure mode is buying a SCARA on speed alone, then discovering the geometry does not work. It is the wrong choice when:
- The tool must approach from changing angles. Four axes only present the tool vertically. Tilted insertions, contoured surfaces and multi-face work need a 6-axis arm.
- The payload or reach is out of band. Cases, cartons and layer picking belong on a heavier arm, see our palletizing systems page for that class of work.
- The travel is long. Metres of straight travel are cheaper and stiffer on a cartesian or gantry frame than on a rotary arm sized to reach that far.
- An operator has to share the space. If people work inside the envelope through the shift, a force-limited arm with a proper risk assessment fits better than a fenced high-speed SCARA.
- The parts are chip-scale at SMT rates. A placement machine exists for that job and will beat any general-purpose robot on it.
- The cycle is dominated by process, not motion. If the station waits four seconds on a cure, test or dwell, a faster arm buys nothing. Fix the process bottleneck first.
What does a SCARA cell need besides the robot?
How should the camera be mounted?
| Mounting | Advantages | Disadvantages | Best for |
|---|---|---|---|
| Fixed overhead | Simple calibration, no effect on cycle time | Limited to one viewing area | Pick station with a consistent part area |
| Robot-mounted (eye-in-hand) | Flexible viewing, follows the arm | Adds weight, usually must stop to capture | Variable pick locations, large work area |
| Fixed upward-looking | Checks the part after pick, from underneath | Adds travel time to pass over the camera | Verifying orientation in the gripper |
Our usual arrangement is a fixed overhead camera at the pick station for part location, plus a fixed upward-looking camera for picked-part verification where the process needs it. That gives position data without hanging weight on the arm.
How do parts get to the robot?
The feeder has to match the robot’s cycle and handle the part without damaging it. Costs below are relative, not quotations.
| Feeding method | Suited part types | Feed rate | Relative cost |
|---|---|---|---|
| Vibratory bowl feeder | Small symmetrical parts (screws, caps, pins) | High, part-dependent | Medium, rises sharply with awkward geometry |
| Tray stacker/unstacker | Fragile or orientation-sensitive parts | Tray-dependent | Medium-high |
| Tape and reel feeder | SMD components, connectors | Component-dependent | Low |
| Flexible vision feeder | Mixed shapes, small batches, variable parts | Moderate | High |
| Conveyor with vision | Any shape the camera can find on the belt | Line-speed dependent | Medium |
For high-mix electronics assembly, a flexible vision feeder paired with a vision-guided SCARA handles changeover in software: the feeder spreads parts on a flat surface, the camera identifies type, location and orientation, and the robot picks what it can reach. No mechanical retooling between variants.
Does the line have to stop for the pick?
Not if the controller supports conveyor tracking. The robot reads a conveyor encoder, predicts where the part will be, and adjusts its path while the belt keeps moving. It needs an encoder on the conveyor drive, a vision or sensor trigger upstream of the pick point, controller support for tracking, and enough reach and speed to finish the pick inside the tracking window. If any of those is missing, index the conveyor and accept the stop.
What safeguarding does a SCARA cell need?
A SCARA runs fast enough that it is not a collaborative machine. The cell needs safety-rated perimeter guarding, fence, light curtains or safety scanner zones, with interlocked access for maintenance and replenishment, E-stops reachable from every operator position, and enough separation between the robot envelope and where people stand. The robot itself falls under ISO 10218-1 and the integrated cell under ISO 10218-2.
Where floor space is tight, safety-rated area scanners can define speed-limited and stop zones instead of a full physical fence. Either way the layout has to be justified by a risk assessment and the stopping distance has to be measured, not assumed. That work is described on our machine safety and compliance page.
Which SCARA projects has Motionwell delivered?
Sensor panel assembly
Vision-guided ABB SCARA assembly of sensor panels, with ±0.01 mm repeat accuracy at the arm. Components are picked, aligned under vision feedback and inserted into the panel assembly. This is the cell we point customers at when the requirement is accurate placement of small parts presented with position variation.
Secondary packaging line (Yamaha SCARA, 2026)
Four Yamaha SCARA robots on a secondary packaging line, in delivery through 2026. Different demands from panel assembly: the arms are sized for carton and case handling and sequenced against line rate rather than tuned for placement accuracy.
Related SCARA-adjacent work sits on our medical device automation pages, where rotary assembly platforms combine indexed stations with vision inspection.
How do you scope a SCARA cell for your line?
Send us enough to size the cell rather than guess at it:
- Part geometry and weight, drawings or samples
- The current process and the cycle time you need to hit
- Placement accuracy requirement, and how it is measured on the finished assembly
- Production volume, in parts per hour and shift pattern
- Existing PLC, network and control standards
- Clean environment or washdown requirements, if any
We evaluate the application, propose a cell configuration and platform, and give a budget range. Where vision guidance is involved we run a feasibility study on your actual production parts before anyone commits to a specification, because part surface, colour and presentation decide whether the vision works far more than the camera model does.
If you are still comparing suppliers, our guide on choosing a system integrator covers what to ask before the quotation stage. To discuss a SCARA application directly, contact the engineering team.