SMT Assembly Process Explained and Equipment Used: A Step-by-Step Guide to PCBA Manufacturing

Today's high-performance electronics, from pocket-sized smartphones and IoT sensors to sophisticated industrial control systems, are made possible by a manufacturing miracle: Surface Mount Technology (SMT). SMT is the backbone of electronics manufacturing, allowing us to have the incredible component density and miniaturization.

A bare circuit board is simply a substrate. The process of transforming it into a functional electronic circuit is known as Printed Circuit Board Assembly. This article outlines a detailed and technical description of the SMT PCB assembly process, designed for engineers, designers, and procurement specialists who want to understand the "how" and "why" of a high-quality, reliable PCB assembly.

The SMT process follows a systematic group of high-precision steps, where a mistake in phase one could lead to catastrophic failures at the end.

Basic steps of SMT Assembly

Phase 1: Pre-Assembly and Solder Paste Application

Statistically, a majority of all SMT defects can be traced back to this initial phase. Precision here is not just recommended; it is mandatory for a high-yield production run.

A. Data Preparation & DFM (Design for Manufacturability) Checks

Before any board enters the SMT line, a digital blueprint must be produced. This goes significantly further beyond Gerber files. A complete data package will be required:

● BOM (Bill of Materials): Lists every part, its manufacturer part number (MPN), and its reference designator on the board.

● Pick-and-Place File: A simple text file that lists the X-Y coordinates, its orientation (rotation), and board side for every single component.

This data is fed into an automated DFM (Design for Manufacturability) check. This software analysis is the first step, catching design-based errors before they become costly physical problems.

It checks for:

● Incorrect component footprints that don't match the physical part.

● Insufficient pad-to-pad or component-to-component clearance.

● Missing or poorly placed fiducial markers.

DFM Check Flaging Pad to Board Edge error

JLCPCB's SMT PCB assembly service streamlines this DFM check. By linking PCB fabrication and assembly data, they can catch potential assembly-halting errors, such as a footprint-to-drill-hit conflict, long before the stencil is ever cut.

B. Solder Paste Printing

Solder Paste Printing is the "screen printing" process of the electronics world. The objective is to deposit a specific, repeatable volume of solder paste onto each component pad on the PCB.

● The Medium: Solder paste is a thixotropic (viscous at rest, fluid when sheared) mixture of microscopic solder spheres in a flux medium. For modern lead-free processes, an alloy of the type SAC305 (96.5% Tin, 3% Silver, 0.5% Copper) is used paired with a 'no-clean' flux, which (as the name implies) does not require a post-assembly washing step.

● The Tool: A solder paste stencil, typically laser-cut from stainless steel, is used. The design of the stencil's apertures (openings) is a science in itself, requiring the design to account for aspect ratio and area ratio to ensure proper "paste release", especially for fine-pitch components (e.g., 0.4mm pitch BGAs).

● The Solder Paste Printing Process:

1) The bare PCB is loaded onto the line and clamped in place. A vision system uses the fiducial markers on the PCB to line it up precisely with the stencil.

2) A metal squeegee (or enclosed print head) then passes over the stencil, forcing the solder paste into the apertures.

3) The stencil "snaps off" the board (a controlled vertical separation), leaving behind precise, defined paste deposits on the component pads.

C. Solder Paste Inspection (SPI)

Solder Paste Inspection (SPI) is the first and most important quality gate. Immediately after printing, the board passes through a 3D SPI machine. This machine uses laser triangulation to build a complete 3D model of every solder paste deposit on the board.

It doesn't just check for presence; it measures:

● Volume

● Height

● Area

● X-Y Alignment

The machine flags any deposit that is "insufficient," "excessive," or "bridged" before a single, expensive component is placed, saving immense cost and rework time.

SMT solder paste printing process

Phase 2: High-Speed Component Placement

After the board is confirmed to have perfect paste deposits, it progresses into the main stage of the SMT line: the Pick-and-Place (PnP) machine. These are high-speed robotic machines that populate the board.

A. The "Pick-and-Place" (PnP) Machine

These highly complicated machines have several main components:

● Feeders: Which hold the components, typically in tape-and-reel format, trays for very large BGAs or QFNs, or tubes.

● Gantries & placement heads: These are the high-speed robotic arms (often with multiple nozzles) that travel between the feeders and the PCB.

● Nozzles: Use a vacuum to pick individual components from the feeders.

● Vision Systems: "On-the-fly" cameras inspect each component after it's picked to verify its orientation, and check for damage. A separate camera system utilizes the PCB's fiducials to determine the exact location of the board, enabling placement accuracy of better than 30 microns.

A pick-and-place machine accurately positions components onto the PCB.

B. The Pick and Place Machine Process

Using the pick and place file as its map, the machine's software optimizes a placement routine to minimize gantry travel time. The head picks a component, the vision system inspects and orients it, and the head precisely places it onto the tacky solder paste deposit. This happens at incredible speeds, with modern machines capable of placing over 100,000 CPH (Components Per Hour).

C. Special Component Considerations

For components like BGAs (Ball Grid Arrays) and QFNs (Quad-Flat No-leads), the solder connections are hidden underneath the part. The tackiness of the solder paste is the only thing holding these parts in perfect alignment before they enter the reflow oven, which is why the SPI check in Phase 1 is so critical.

4 different types of Solder Defect

Phase 3: Reflow Soldering

This is where the metallurgical bond is formed, turning the temporary assembly into a robust, permanent PCBA. The board, now populated with components, travels through a long reflow oven for reflow soldering.

A. The Reflow Oven

A modern convection reflow oven is a long tunnel with multiple (e.g., 8-12) independent heating zones, followed by cooling zones. This allows for meticulous control over the board's temperature as it travels through.

B. The Critical Reflow Profile (The "Thermal Recipe")

You cannot use one profile for all boards. Every PCBA design must have a custom-tuned thermal profile. To create it, a test board is run with thermocouples attached to different-mass areas (e.g., the leg of a tiny capacitor, the center of a large BGA, and a large ground plane). The oven is then tuned to ensure all parts of the board follow a precise temperature curve.

This curve is broken into four distinct zones:

1. Preheat: A gradual temperature ramp (for example, 1-2°C/sec). This prevents thermal shock, which can crack components or the PCB itself. It also begins to activate the flux in the solder paste slightly.

2. Soak (Thermal Stabilization): A short plateau (for example, 60-90 seconds) where the entire PCBA, regardless of mass, is allowed to reach a uniform temperature. This is very important on complex boards to ensure the BGA and a nearby resistor reach the reflow temperature at the same time.

3. Reflow (Peak): A rapid spike in temperature above the liquidus temperature of the solder (217°C for SAC305), which will peak out around 245-250°C. This is where the solder spheres melt, 'wet' the pad and the component lead, and the flux will aggressively clean oxides to ensure a good connection.

4. Cooling: A controlled, fast-as-possible ramp-down. This is equally critical as the heating. A rapid cooling is critical to induce the formation of a fine-grained, strong intermetallic compound (IMC) layer - the true metallurgical bond between the tin in the solder and the copper of the pad.

C. Key Reflow Profile Terminology

The most critical parameter in the reflow zone is TAL (Time Above Liquidus). This is the window, typically 45-75 seconds, where the solder is fully molten. If it's too short, the joints will be weak. If it's too long, it can damage components and cause excessive, brittle IMC growth, leading to long-term reliability failures.

D. Double-Sided SMT Assembly

For boards with components on both sides, the process is run twice. The "bottom" side (usually with smaller, lighter components) is assembled and reflowed first. The board is then flipped, solder paste is printed for the "top" side, and components are placed.

During the second reflow pass, the bottom-side components do not fall off. They are held in place by the surface tension of their already-solidified (and now molten again) solder joints. This property, however, limits the maximum size and weight of components that can be placed on the bottom side.

JLCPCB offers both single-sided and double-sided SMT PCB Assembly, ensuring high quality with a fully automated and reliable process.

Phase 4: Post-Reflow Inspection and Quality Assurance

The board is now fully assembled, but the process is not complete. A rigorous, multi-stage inspection is required to guarantee that the assembly matches the design and has no hidden defects.

A. Automated Optical Inspection (AOI)

The board immediately passes through an AOI machine. This machine utilizes high-resolution cameras and complex image comparison software to check the PCBA for:

● Component shift and rotation

● Correct component polarity (e.g., diodes, tantalum caps)

● Tombstoning or billboarding

● Solder bridging, and insufficient or excessive solder joints

However, AOI has one major limitation: it is a 'line of sight' inspection; it cannot see the critical solder joints that are hidden under components.

Showing how an Automated Optical Inspection (AOI) works

B. Automated X-Ray Inspection (AXI)

This is the "see-through" solution for post-reflow quality assurance. AXI is used to inspect the hidden solder joints of components like BGAs, QFNs, and large-bottom-terminated parts. The X-ray can see through the silicon of the chip to inspect the individual solder balls underneath, checking for:

● Voids: Bubbles inside the solder joint.

● Shorts: Bridged balls under the BGA.

● Opens: Balls that did not properly connect.

JLCPCB integrates these quality assurance steps (SPI, AOI, and AXI) throughout the line, not just at the end. This creates a real-time feedback loop. If the AOI machine starts to see a component shift, it can feed that data back to the PnP machine for an immediate micro-adjustment.

C. Electrical Testing: Flying Probe Test, In-Circuit Test (ICT) & Functional Testing (FCT)

The final step is to confirm the board works.

● Flying Probe Test (FPT) method is ideal for prototypes and low-volume runs. It uses 2-6 robotic probes to move around the board and touch test points, checking for shorts, opens, and correct component values (resistors, capacitors). It is highly flexible and requires no custom fixture, though the test time per board is slower.

● ICT (In-Circuit Test) often uses a "bed-of-nails" fixture to press pogo pins against test points on the board. It checks for shorts, opens, and verifies that passive components (resistors, capacitors) have the correct values.

● FCT (Functional Test) is a custom-built test jig that powers up the board and simulates its real-world operation. It confirms that the board does what it was designed to do—that LEDs light up, signals are sent, and data is processed correctly.

JLCPCB supports Flying Probe Testing (FPT) and Functional Test as part of its comprehensive PCBA testing services.

D. Conformal Coating

This step is applicable for high-reliability applications, such as those in the automotive, aerospace, and medical fields.

Conformal Coating is a thin, non-conductive polymer layer that is sprayed onto the entire PCBA. This coating protects the assembly from moisture, dust, chemicals, and thermal shock, dramatically increasing its lifespan in harsh environments. This is one reason "no-clean" flux is critical, as residue can prevent the coating from adhering properly.

Conclusion

The SMT (Surface-Mount Technology) process is a high-tech symphony. It is a single, interconnected system where the quality of each step, from the DFM (design for manufacturing) check and solder paste printing to the carefully profiled reflow and inspection with an X-ray, impacts the next.

The reliability of your final electronic product is not only a factor of its design; it is forged in the precision, control, and robust quality assurance of the assembly process. For your next prototype or production run, ensure your design is brought to life with a high-quality, reliable SMT assembly process.

Explore JLCPCB's SMT PCB Assembly service and see how we employ fully automated SMT lines and integrated quality assurance systems to deliver reliability and quality to meet your project specifications.

FAQs

Q1: What is the difference between SMT and THT (Through-Hole Technology)?

SMT and THT are two types of component mounting.

● SMT (Surface Mount Technology): Components are soldered onto the surface of the PCB pads. SMT allows for higher component density, more automation, and is easier to use when assembling onto both sides of the board.

● THT (Through-Hole Technology): Components have leads (wires) that are inserted through drilled holes in the PCB and then soldered on the opposite side, and usually have a stronger mechanical bond. THT is still widely used for bulky components that endure mechanical stress, such as large connectors or power relays

Q2: Can you mix SMT and THT components on the same board?

Yes, this is a mixed-technology assembly and is incredibly common. Typically, the process flow starts by assembling and reflowing the SMT components (as discussed in this article) and, after the SMT pass, inserting the THT components and soldering the through-holes leads using a separate process such as wave soldering or selective soldering (robotic, high-precision soldering iron) without disturbing the already-soldered SMT parts.

Q3: How are moisture-sensitive components (MSL) handled?

These components absorb moisture, which can cause "popcorning" (damage) during reflow. They are stored in "dry bags" and must be "baked" in an oven to remove moisture if their "floor life" (time exposed to ambient air) is exceeded.

Q4: What is "no-clean" flux, and does it really not need to be cleaned?

Its residues are non-corrosive and non-conductive, so they can be left on the board for most applications (IPC Class 1/2). Mission-critical (Class 3) or coated boards may still require cleaning.