First-pass yield, or FPY, is one of the most useful metrics in SMT manufacturing because it shows how many boards move through a defined process stage correctly the first time. In simple terms, FPY measures how much of the factory’s output is produced without rework or repair.
That sounds straightforward, but FPY is often misunderstood. Different factories calculate it at different points in the process, and the number can be misleading if teams do not define clearly what counts as a first-pass success. Used properly, FPY is not just a quality percentage. It is a signal of how well printing, setup, placement, reflow, inspection, and execution work together.
What FPY means in practice
At its core, FPY answers this question: what percentage of boards passed the selected stage without needing correction?
The phrase "selected stage" is important because companies may calculate FPY at:
- SPI
- AOI
- end of SMT
- electrical test
- finished assembly release
These are not equivalent. A board that passes AOI first time may still fail later in test. A board that reaches final shipment may still have consumed rework effort along the way. That is why the metric only becomes meaningful when the organization is clear about where it is measured.
Why FPY matters
FPY matters because a first-pass failure affects more than defect counts. Each failure also consumes resources such as:
- operator time
- review time
- rework capacity
- engineering attention
- queue time
- schedule margin
A low FPY therefore usually means more than weak quality. It often means weaker throughput, less predictable delivery, and more labor cost.
FPY is different from final yield
One of the most common mistakes is to confuse FPY with final yield. Final yield measures how many units eventually become acceptable, including those that needed repair or rework. FPY measures how many passed without that recovery step.
This distinction matters because a line can have strong final yield but weak FPY. In that case, the factory may be recovering product successfully, but the underlying process is still unstable.
Why the definition must be clear
FPY is only useful if pass and fail conditions are defined consistently. For example:
- Does manual review count as a failure?
- Do AOI false calls reduce FPY?
- Does a corrected setup error affect the metric?
- Is a board considered first-pass good if it passes AOI but fails later in ICT?
Different factories use different rules. The key requirement is internal consistency so that the number can support real decision-making.
The basic calculation
A simple FPY formula is:
FPY = first-pass good units / total units entering the selected stage
The formula is simple, but interpretation depends on what qualifies as a first-pass good unit in the real workflow.
In many SMT environments, a first-pass good board is one that:
- completes the chosen stage without repair
- does not need touch-up or rework
- meets release criteria immediately
- does not require repeated handling because of a true defect
Some factories count false-call review separately rather than mixing it into FPY loss. That can be sensible as long as the method is clear.
What a high FPY usually tells you
When FPY is consistently high and the definition is sound, it usually suggests that:
- printing is stable
- setup control is effective
- material loading errors are limited
- placement is reliable
- reflow is reasonably matched to the product
- inspection is not creating excessive unnecessary disruption
It does not mean the process is perfect. It means the line is avoiding many preventable failures.
What a low FPY usually tells you
A low FPY means the process is losing control somewhere before the board exits the selected stage cleanly. Common causes include:
- unstable paste printing
- wrong materials or feeder loading
- nozzle or placement issues
- recurring solder defects
- weak changeover discipline
- high false-call burden in inspection
That is why FPY should be interpreted as a process signal, not merely a quality statistic.
Why the number alone is not enough
FPY becomes much more useful when it is paired with loss categories. A single percentage cannot tell you whether the line is mainly failing because of:
- bridging
- tombstoning
- wrong polarity
- print variation
- placement offset
- inspection noise
Two product lines with the same FPY may need different corrective actions. One may need stencil-printing control. The other may need setup verification or AOI refinement.
Why stage-level FPY is often better than one site-wide number
A single plant-wide FPY figure can hide important patterns. More useful analysis often breaks the metric down by:
- line
- product family
- process stage
- defect type
- board location
- shift
- material lot when relevant
This makes the data more actionable. If one product family loses FPY only after changeover, a plant-wide average may not show the real issue.
False calls can distort interpretation
Inspection systems complicate FPY because not every failure event represents a real manufacturing defect. A poorly tuned AOI program may create high review burden while the actual assembly process remains relatively stable.
For that reason, many factories separate:
- true defect-related first-pass loss
- review burden caused by false calls
Both matter operationally, but they do not mean the same thing. If false calls dominate, the next action may be inspection optimization rather than process correction.
FPY often points upstream
Boards usually fail at one stage for reasons that began earlier. A solder defect seen after reflow may have started in printing. A polarity issue found at AOI may have started in setup control. A placement shift may trace back to feeder or nozzle condition.
That is why FPY should be read as a connected-process metric. The most useful question is where the process first lost control.
How to read FPY trends
Trend analysis is often more valuable than a single daily number. Useful questions include:
- did FPY drop after a material change?
- does one product family consistently run lower?
- does the loss appear after changeover?
- did a new inspection program increase review burden?
- are the same defects repeating in one board area?
These questions help turn FPY from a dashboard number into an engineering tool.
What FPY does not tell you by itself
FPY is valuable, but it does not reveal everything. On its own, it does not show:
- severity of each failure
- repair time per unit
- customer risk if a defect escaped
- cost of each failure category
- whether the loss came from real defects or mainly review noise
That is why FPY should be read alongside defect Pareto data, rework analysis, and inspection trends.
Common mistakes when reading FPY
Several habits reduce FPY’s value:
- comparing numbers that use different definitions
- looking only at the percentage without loss categories
- ignoring false-call influence
- assuming final yield makes FPY less important
- reacting to one short-term dip without checking the trend
These mistakes usually create argument rather than improvement.
What good FPY analysis looks like
Useful FPY analysis usually does a few things well:
1. defines pass and fail clearly
2. separates true defect loss from inspection noise where needed
3. links recurring losses to specific process causes
4. reviews trends by product, line, and event
5. checks whether corrective actions actually improved first-pass performance
That is what makes FPY useful operationally.
Key takeaway
First-pass yield in SMT measures how many boards pass a defined process stage the first time without rework or correction. It should be read as a process-performance metric, not just a quality percentage. A strong FPY usually indicates better control across printing, setup, placement, reflow, and inspection. A weak FPY usually means the factory is spending too much effort recovering from preventable instability. To read FPY correctly, define it clearly, separate true defects from review noise where needed, and connect the number to recurring process losses behind it.