SMT Assembly Service Types: Matching Board to Supplier

Two assembly houses can own the same model of placement machine and still be unable to build the same board. What separates them is not the equipment list but the way the line is configured, the staff around it, and the class of work the business was built to accept. Choosing a partner is therefore a question of matching the board to the shop rather than ranking shops in the abstract.

This article sets out the categories of SMT assembly service that exist in practice, the kind of work each one is set up to do well, and the questions that separate a shop that can build a prototype from one that can only run volume.

Why SMT Shops Are Not Interchangeable

SMT assembly is a chain of operations: paste printing, placement, reflow, inspection, rework and often through-hole soldering on the same panel. A shop that has optimised that chain for one product family has tuned its stencils, its nozzle inventory, its feeder setup and its reflow profile around that family, and those choices do not transfer.

Advertising adds to the confusion, because almost every shop describes itself with the same words. The useful signal is not the service list but the evidence: which products the shop builds every week, how its line is staffed, and whether it holds inventory for the parts it claims to place. A shop that cannot describe its last ten jobs in the same product family is telling you something about how it will treat yours.

Specialist Shops Built Around One Product Family

The first category is the specialist. These shops build one class of product and little else: mobile boards, automotive electronics, flexible circuits, or LED lighting and display panels. Their machines, fixtures, test tools and operators are all selected for that class, and the result is a line with very few surprises for a board that fits the profile.

The limitation is equally clear. A board outside the class will not benefit from any of that tuning, and the shop may not have the stencil frames, nozzle sizes or reflow capability to handle it. If your product does not resemble what the shop builds every day, another supplier is the better answer regardless of price.

Volume Shops: Precision Tier and Value Tier

The second category is the volume producer, and it usually splits into two tiers. The precision tier runs finer pitch, more placement positions, tighter paste control and full inspection, and prices accordingly. The value tier runs simpler boards in large quantities with the minimum of touch time, and its cost advantage comes from exactly that simplification.

Matching a product to a tier is mostly a matter of complexity and volume. A simple, high-volume board belongs in the value tier, where the process is tuned for throughput. A dense board with fine-pitch devices and a high layer count belongs in the precision tier, where the investment in process control is what keeps the yield acceptable.

The Mid-Size Shop and the Prototype Problem

The mid-size shop is the most common arrangement and the one where prototype work is most often mishandled. Several placement machines, a full staffing structure with planners, in-process inspectors, engineers and operators, and a shift pattern that keeps the line running: all of that is built for repeat production, not for a one-off panel.

A prototype or a small sample run does not fit that structure. It interrupts a scheduled line, needs components that may not be in stock, and may require parts that cannot all be placed by machine, so someone has to finish the job by hand. Where no operator is assigned to that work, quality depends on whoever is free, and the order in which the parts are placed stops following a plan.

Machine-Only Workshops and What They Cannot Do

At the other end is the shop that owns a machine and little else. It can place parts at a low unit price, and it can do so quickly, but it typically has no stores function, no inspection step and no engineering support. Anything outside the machine recipe, including hand placement, solder paste control and first-article confirmation, is outside its scope.

That arrangement produces a distinctive failure mode: boards that look finished and fail later. Paste that was handled without control, components placed without confirming orientation, and parts placed but never inspected all appear at final test or in the field. The causes of components shifting during reflow are a reminder of how much process control sits behind an ordinary-looking board.

Small-Batch and Prototype-Oriented Assembly

A fourth category exists for the gap the others leave: shops built deliberately around prototypes and small batches. Their distinguishing feature is not the number of machines but the presence of hand-soldering technicians who can place fine-pitch devices and chip parts down to the smallest standard case sizes, and who can remove and rework anything with accessible leads.

These shops also tend to keep the other functions that volume work does not need. They clean up the bill of materials before ordering, generate the placement and coordinate files from the board data, and offer component sourcing so that a short or obsolete part does not stall the build. The wider stages of turning a design into an assembled board describe where this support fits.

Selection Criteria: Capability, Materials and Traceability

Three questions do most of the work when screening an assembly supplier. The first is capability: the finest pitch and the smallest case size the line handles routinely, the largest board it can accept, and the highest placement count it wants to see on a single panel. Equipment lists are easy to copy from a brochure; a description of where the line stops coping is harder to write and much more informative. Answers that are given as a range rather than a claim are worth more.

The second is materials control. Ask how solder paste is stored, thawed and logged, whether a stencil is inspected before use and how often it is cleaned, and how first-article inspection is recorded. The third is traceability: whether the shop can tell you which paste lot and which component reel went into a given panel.

Prototype Runs Before Volume

A prototype assembly run is the cheapest test of a supplier that exists, and it should be treated as one. Sending one panel of a representative board shows how the shop handles a short build, whether it comes back with questions or with guesses, and whether the documentation that returns with the boards matches what was sent.

Where a supplier passes that test, the hand-off to volume is straightforward, because the process has already been proven on the actual design. Where it fails, the cost of finding out is one panel rather than a production run. For boards that are still changing, the route through prototype requirements for multilayer boards is the same one that leads to volume.

FAQ

Should a full turnkey assembly order include the bare board? Usually yes. Buying the board and the assembly from one supplier removes the argument about who is responsible when a pad will not wet, and it simplifies the logistics of a small build.

Is a low unit price a reliable guide? No. A quotation that omits paste control, first-article inspection and rework will be lower, and the difference will appear later as yield loss or field returns rather than on the invoice. The comparison that matters is the landed cost of a good panel, not the price of a placement.

How much documentation should be sent? The bill of materials with manufacturer part numbers, the board data in a neutral format, and a placement list. A shop that reformats that data itself is a shop that does the preparation work rather than passing it back to you.