PCB Sample Boards: Types, Materials and Ordering Guide

What a PCB Sample Is For

A sample board is a small quantity of boards built to engineering requirements before any production commitment is made. Quantities are typically between one and twenty pieces, and the important detail is that they are fabricated on the production process rather than on a simplified one.

That last point is what gives the sample its value. A board that works because it was built with extra attention, on a different stackup, using a material the production line will not carry, has demonstrated nothing about the product. It has demonstrated that a board can be made by hand. The purpose of the sample is to answer the questions that production will otherwise answer for you: does the circuit function, does the signal integrity hold, does the thermal path work, do the connectors and the enclosure fit, and can the design actually be manufactured and assembled.

The economics are straightforward. A sample order costs a few hundred dollars. Discovering the same problem after tooling and production setup costs far more, and discovering it after a product launch costs more again.

Sample Against Prototype

The two terms are often used interchangeably, and there is a useful distinction between them.

• Quantity: samples usually run to one to twenty pieces; prototypes typically from five to one hundred.
• Purpose: samples exist for engineering validation and measurement. Prototypes often serve as a pilot run for a small production batch or a market test.
• Unit price: higher for a sample, because the fixed setup is spread over fewer boards.
• Process: a sample should use the full production process, and a pilot prototype should be identical to what follows.

In practice a sample is an engineering grade prototype, and the same discipline applies to both: build them on the construction that production will use.

What Samples Are Used For

• Functional verification: circuit logic and power stability under real conditions rather than in simulation.
• Signal integrity testing: high speed and high frequency measurement, where the board geometry is part of the circuit.
• Thermal evaluation: confirming that the heat path works and that the copper distribution is adequate.
• Mechanical and fit checks: connectors, interfaces and whether the board actually fits the enclosure.
• Manufacturability and assembly review: finding the design features that the process cannot hold reliably before committing to volume.

The last one is overlooked most often and is often the most valuable. A sample run is the cheapest manufacturability test available, because the fabricator builds the actual design rather than reviewing a drawing.

Common Sample Types

• Single sided: simple control or power boards with a minimal circuit.
• Double sided: the general purpose case for most consumer and industrial electronics.
• Multilayer, four to ten layers: high density designs, where the stackup itself needs validation.
• Flexible circuits: wearable devices and camera modules, where the board has to bend or fold.
• Rigid flex: medical and aerospace designs that combine a rigid section with a flexible interconnect.
• High frequency: RF, microwave and antenna systems, where the material properties are part of the design.

Which of those applies is decided by the application rather than by preference, since the construction determines what the sample can prove. A single sided sample cannot validate an impedance controlled high speed design.

Materials and Stackups

• Standard FR-4: the default for general electronics and the most economical option.
• High Tg FR-4: industrial and automotive work where thermal stability through assembly and operation matters.
• Halogen free FR-4: compliance driven projects where the material has to meet the emission and content requirements.
• High frequency laminates: hydrocarbon ceramic and PTFE materials for RF and microwave work, and the hybrid stacks that combine them with FR-4.

Typical stackups run from a two layer FR-4 board with one ounce copper, through a four layer impedance controlled stack, to six layer mixed signal designs and hybrid high frequency stacks. The material and stackup have to be chosen for the production build, not for the sample alone, because a sample that validates a material the production line will not run is worse than no sample at all. Working through that decision belongs in the design and layout phase, before the sample order is placed.

Fabrication Flow

Sample boards follow the same process as production boards.

• Design file review: Gerber, drill data and stackup information, followed by a design for manufacture review.
• Inner layer imaging and etching, with inspection of the internal layers before they are laminated into the stack.
• Lamination and drilling, with the layer registration that multilayer construction requires.
• Plating and outer layer formation, producing the final conductor geometry.
• Surface finish: ENIG, hot air solder levelling, or OSP, chosen for the application.
• Full electrical test, not a sampled test, since every sample board is going to be evaluated.

Where the sample includes assembly, surface mount, through hole and functional test are added to the sequence. Getting the first article through prototype assembly at the same specification as production assembly surfaces the component placement and stencil issues that a bare board cannot reveal.

Cost

Sample pricing is dominated by setup rather than material, which is why the unit cost at five pieces is several times the unit cost at a thousand.

• Two layer FR-4, five pieces: roughly 20 to 50 dollars per board.
• Four layer FR-4, five pieces: roughly 60 to 120 dollars per board.
• Six layer FR-4, five pieces: roughly 120 to 250 dollars per board.
• High frequency, two to five pieces: roughly 150 to 400 dollars per board.
• Rigid flex, two to five pieces: roughly 300 to 800 dollars per board.

The main factors within those bands are layer count and board size, the material, the surface finish, and whether impedance control or another special process is required. A first article on standard material with a simple stackup sits at the low end; a hybrid high frequency stack with controlled impedance and a fine line requirement sits at the top, and the difference is in the process rather than in the board area.

Lead Time and Expediting

• Standard samples: five to seven working days.
• Expedited samples: twenty four to seventy two hours, subject to material availability.
• High frequency and rigid flex: seven to ten working days, because of the material and the additional lamination steps.

Expediting shortens the development cycle at a cost, and it is most useful when a design iteration is on the critical path. It is not a substitute for having the design correct, since a fast board built to a wrong stackup simply fails sooner.

Quality and Test Requirements

• Full electrical test on every board rather than a sample, since each one will be evaluated.
• Automated optical inspection for conductor and solder mask defects.
• Impedance testing where the design specifies controlled impedance.
• Confirmation of board thickness, copper weight and material against the specification, which matters when the sample is going to be used for certification or for a design freeze.

The point of that testing on a sample is that it establishes the baseline the production run has to match. The same reasoning carries into the assembled product, where the PCBA testing regime applied to the sample should be the one that will be applied in volume, so that the test coverage is validated along with the design. Building the sample under a supplier’s documented quality management system is what makes the measurements, and the records behind them, usable as evidence later.

Sample, Then Small Batch

The sequence from sample to production has three stages. The sample validates the design at one to twenty pieces and low flexibility. The small batch, typically fifty to five hundred pieces, serves market testing and process validation with a lower unit price and less flexibility. Production follows, with the lowest unit price and the least flexibility of all.

The trap is a design that can be built as a sample but not as a production order, usually because it relies on a process window that only holds at small quantities. Reviewing manufacturability at the sample stage, and confirming the material supply, is what prevents the design from being returned to the drawing board after the pilot run.

Ordering

Have the following ready before requesting a quotation: Gerber files, drill data, a stackup and impedance specification, and the quantity and lead time the program needs. Clear engineering communication at this point prevents the rework that otherwise happens after the boards arrive, and it allows the fabricator to return a design for manufacture report with the quotation rather than sending the boards and letting the design speak for itself.

Frequently Asked Questions

How many samples should be ordered? Five to ten is the useful range, allowing for measurement, spares and destructive testing.

Can a sample be used for certification testing? Yes, provided the material and process match production exactly. That is the reason to build a sample on the production construction.

Is there a minimum order quantity? Many fabricators, including high volume manufacturers, accept orders from a single board, though the unit price reflects the setup cost.

What is the fastest available turnaround? Twenty four to seventy two hours for standard constructions where the material is in stock.

What usually goes wrong on a first sample? Assumed impedance values that the stackup does not deliver, poorly designed pads or solder mask openings, unbalanced copper distribution causing warp, and fabrication notes that leave the finish or tolerance to interpretation.

Summary

A PCB sample is a small quantity of boards built to the production process, used to validate function, signal integrity, thermal behaviour, mechanical fit and manufacturability before any volume commitment. The distinction from a prototype is mainly one of quantity and intent; the discipline is the same, which is to build on the construction that production will use.

The costs are dominated by setup, running from roughly twenty to fifty dollars per board for a two layer sample to several hundred for rigid flex and high frequency constructions, with lead times from twenty four hours expedited to ten working days for the more complex materials.

The value of the exercise depends on what is checked. Full electrical test, impedance verification where relevant and confirmation of the stackup and materials give a baseline that production can be held to. Skipping the sample to save a week is almost always the more expensive decision, because the same problems will be found later, when the tooling exists and the components are already attached.