


# PCB Impedance Test Report: How to Verify a PASS

_Published: 2026-11-16_

Table of Contents

Toggle

- PCB Impedance Test Report: Start With the Drawing Revision
- Impedance Test Coupon vs. a Trace on the PCB
- Controlled Impedance Stackup: Map Each Report Row to Its Trace Group
- PCB Impedance Tolerance: Calculate the Actual Limits
- TDR Measurement: Check the Window Behind the Number
- Controlled Impedance PCB Testing: Audit a Report Line by Line
- PCB Quality Control: Separate a Retest From a Process Investigation
- PCB Acceptance Criteria: What to Ask for Before Release
- PCB Impedance Test Report FAQs
- Technical References
- Custom Controlled-Impedance PCBs With Stackup Review

A PCB impedance test report is enough to support an impedance release only when each required trace group has a traceable result, was measured by the agreed method, and meets the drawing limits under the agreed decision rule. A PASS on a 100 Ω coupon on Layer 3 does not cover a second 100 Ω group on Layer 6. The two numbers may have the same target, but the copper, dielectric, and reference planes that determine the result can be different.

Use the drawing and the report side by side. Start with the groups the drawing requires, not with the rows the supplier happened to report. The worked example below shows how to distinguish an in-limit result, a group with no evidence, and a result above the upper limit.

![Green bare PCB production panel with straight impedance coupon traces on its edge rail](https://www.bestpcbs.com/blog/wp-content/uploads/2026/11/pcb-impedance-test-report-hero.jpg)Illustration of a production panel with test traces on its rail. A released panel drawing is needed to identify which structure each trace represents.

## PCB Impedance Test Report: Start With the Drawing Revision

Match the report to the released board part number and revision, approved stackup revision, and the lot or panel identifiers defined on the order. A PDF named after the correct part is not enough: the filename says nothing about which revision was measured or whether the tested panels were included in the shipment.

For each controlled structure, enter the identifiers that let both parties match a result to a requirement. For example, copy **DP3 | L3 | L2/L4 | differential** from the drawing, then find DP3 or its documented equivalent in the coupon map. If the report uses a panel label and the shipment uses a serial range, ask for the panel-to-board mapping. Do not guess from the production date.

## Impedance Test Coupon vs. a Trace on the PCB

A panel coupon is useful for checking repeatable fabrication of a particular line structure. An on-board structure is useful when the decision depends on a location or geometry the rail cannot represent. IPC-TM-650 2.5.5.7A (March 2004), §3.1, gives examples of both. Neither automatically verifies an assembled link.

**Specimen**
**What it can verify**
**What to check before accepting it**

Same-panel coupon
The sampled line construction when it matches the approved group.
Match signal layer, reference plane, trace width and pair spacing, dielectric, copper, and relevant solder mask condition.

Dedicated test line inside the board outline
Local production geometry at that board location.
Confirm an accessible launch, reference connection, and enough uniform line for the measurement.

Functional production trace
The accessible segment of the actual route.
Check the scope and window before interpreting the result. Branches, vias, connectors, or an inaccessible end can prevent a clean estimate of uniform line impedance.

For an RF microstrip ordered with a specified mask condition, model and test the agreed condition. A rail line tested without mask does not, by itself, verify a masked production line. If the question is a dip at a layer transition, a matching straight coupon and a [local via impedance measurement](https://www.bestpcbs.com/blog/2026/11/pcb-via-impedance-tdr/) address different parts of that problem.

## Controlled Impedance Stackup: Map Each Report Row to Its Trace Group

Join the drawing and test report using the trace group, not the impedance value. A coupon identifier such as C07 is traceable only if the coupon drawing or a signed test map establishes what it represents. Two groups can both specify 100 Ω, yet have different reference planes and trace spacing. A 100 Ω stripline between two planes is not interchangeable with a 100 Ω outside-layer microstrip.

Record the signal layer, reference plane or planes, single-ended or differential mode, and the coupon geometry. Use the _approved_ CAM dimensions or finished dimensions as appropriate; original CAD width and finished etched width are not the same field. When the coupon differs intentionally from the product geometry, request the engineering basis and its acceptance by the purchaser before treating it as equivalent evidence.

![Two differential stripline cross-sections on different signal and reference layers](https://www.bestpcbs.com/blog/wp-content/uploads/2026/11/pcb-impedance-coupon-groups.jpg)Both groups can target 100 Ω. They must still be matched independently to the report because they occupy different layers. Cross-sections are schematic.

## PCB Impedance Tolerance: Calculate the Actual Limits

A 100 Ω target with a specified tolerance of ±8% has a 92–108 Ω acceptance interval; a separate 50 Ω target at ±10% has a 45–55 Ω interval. Use the actual limits on the drawing, including whether they are absolute, percentage-based, or asymmetric. Do not copy a default tolerance from a supplier’s report template.

For symmetric percentage limits, multiply the target by **1 − tolerance** and **1 + tolerance**, with the percentage written as a fraction. For 100 Ω × 0.08, the allowance is 8 Ω. A measured 109.1 Ω is 1.1 Ω above the upper limit. Preserve the reported precision and apply the agreed decision rule for uncertainty or rounding. This example assumes comparison to the stated limits with no separate guard band; neither the numbers nor the rule are IPC defaults.

## TDR Measurement: Check the Window Behind the Number

Time-domain reflectometry (TDR) launches an electrical step and observes reflections. A change in impedance changes the reflection. A straight coupon section can provide a usable measurement region, but its probe launch and far-end termination are different structures and may dominate a poorly placed window.

Ask for the trace and marked measurement window when a result is near the limit, the value shifts between panels, or the order requires waveform evidence. Check that the window stays within the uniform line and excludes the initial launch transient and the far-end reflection. If no usable flat region exists, get the lab to explain its method and limitations or agree on a more suitable specimen. Do not move the window after seeing the result just to rescue a PASS. IPC-TM-650 2.5.5.7A describes the measurement method and its limitations; it does not set your lot release limits.

If labs disagree about a differential measurement and the difference appears near the launch, ask for the two raw leg traces and the reference check. The [differential TDR deskew guide](https://www.bestpcbs.com/blog/2026/11/differential-tdr-deskew/) explains how to separate launch timing from receive-path mismatch before attributing the difference to the coupon.

![Schematic TDR trace with probe launch, marked measurement window, and far-end reflection](https://www.bestpcbs.com/blog/wp-content/uploads/2026/11/tdr-report-measurement-window.jpg)Place the agreed window on the uniform line. This schematic shows the separation of events, not measured values or a usable time setting for every probe.

## Controlled Impedance PCB Testing: Audit a Report Line by Line

For this hypothetical Rev C board, compare the three drawing groups below against a report for panel P04. Assume the order calls for one eligible result per group on this panel and direct comparison to the drawing limits. Use the worksheet after confirming the document revisions and panel ID. Fill one row for every drawing group—even if the report has no corresponding row.

**Drawing group and limits**
**Report evidence**
**Decision and next step**

DP3 · L3/L2–L4 · differential; 92–108 Ω
C07 maps to DP3; 101.2 Ω
Matches and is in limits. Record C07 and the accepted measurement window.

DP6 · L6/L5–L7 · differential; 92–108 Ω
No row or mapped specimen
Not verified. Request a P04 result for DP6. Do not copy DP3’s value.

SE1 · L1/L2 · single-ended; 45–55 Ω
C09 maps to SE1; 54.0 Ω
Matches and is in limits under the example rule. Confirm the mask condition in the C09 map.

If the supplier adds an eligible DP6 result of 109.1 Ω, change its status from “not verified” to “above the upper limit by 1.1 Ω” under the example rule. The disposition remains on hold, but now the next action is a measurement review and a documented nonconformance—not a request to fill a blank row. A second test must follow an agreed process; do not average it with another trace group.

## PCB Quality Control: Separate a Retest From a Process Investigation

Check identification and measurement repeatability before attributing an out-of-limit number to the board. Confirm the coupon map, probe contact, reference and calibration status, test mode, and accepted window. If a repeat on the same coupon with the same agreed method reproduces the excursion, compare a matched section with the released stackup: copper profile and pair spacing, relevant dielectric thickness, and coating condition. Document any deviation from the planned sampling before extending the result to the lot.

For a panel-rail process check, acceptance comes from the agreed groups and sample coverage, not the count of green entries in the PDF. For prototype debug, use an accessible board segment or a dedicated test line if the rail result cannot answer a board-local question. For an assembled high-speed interface, verify the interface by its own agreed criteria—such as channel loss or eye-mask testing where called for. A coupon impedance result does not certify the connector-to-receiver link.

## PCB Acceptance Criteria: What to Ask for Before Release

A release reviewer should be able to name the drawing revision, tested population, covered groups, measurement method, and decision rule without reverse-engineering the supplier’s file. A compact request is: “Please supply the panel or lot map, relate every coupon trace to its released impedance group, and provide the agreed measurement window and result for each required group.”

If the revision or sample coverage is missing, hold the impedance release and ask for that evidence. If a compliant retest confirms an out-of-limit result, record the nonconformance and follow the purchase agreement for containment, replacement, or an approved concession. If the evidence matches and is within limits, sign off the impedance requirement only; it is not a substitute for bare-board continuity or assembled function tests.

## PCB Impedance Test Report FAQs

### Does a 101.2 Ω measurement qualify two 100 Ω trace groups?

No. It qualifies only the mapped group and sampled condition, assuming it meets the agreed rule. Separate signal or reference layers require their own documented coverage or an explicitly agreed equivalence.

### Is a 50 Ω single-ended reading the same as 100 Ω differential?

No. In a coupled pair, differential impedance depends on odd-mode impedance, not necessarily the impedance of a single isolated trace. Verify the mode and fixture used for the specified test.

### Can an impedance report identify which finished boards were tested?

Only when the identifiers support that claim. A rail coupon usually identifies a panel sample, not a test on each board. Ask for the panel-to-board or panel-to-lot mapping when serial-level traceability matters.

### Why might two lots give different but in-limit readings?

Finished dimensions and material properties can vary. First check that both reports use comparable coupon revisions, mask conditions, and measurement windows. Then trend like-for-like results within the agreed sampling scheme.

### What if the report lists percent deviation but no ohms?

Confirm whether the deviation is signed and what target it uses. For example, +3.2% relative to 100 Ω means 103.2 Ω; an unsigned 3.2% could indicate either 103.2 Ω or 96.8 Ω.

### Can a passing report replace an insertion-loss measurement?

No. Characteristic impedance and insertion loss answer different questions. Use the frequency-domain or other agreed channel measurement when loss is part of the requirement.

### Should an auditor require an IPC class label on every report?

Check the contract for the required product specification, revision, and class. The TDR method reference alone does not assign a product class or substitute for the order’s documented acceptance criteria.

### Is a coupon report still useful after the boards have been depanelized?

Yes, if the coupon remains linked to the shipped lot or board population and the test conditions are known. Without that mapping, its values may describe a process run but cannot substantiate the particular shipment.

## Technical References

- IPC-TM-650 2.5.5.7A: Characteristic Impedance of Lines on Printed Boards by TDR, March 2004; test specimens, measurement setup, and limitations.
- Polar Instruments AP124: Testing Controlled Impedance Boards with Test Coupons; representative line construction and panel context.
- Polar Instruments AP8502: Test Coupons or Built-In Test Traces?; practical access and measurement limitations.

## Custom Controlled-Impedance PCBs With Stackup Review

Send EBest Circuit your fabrication data, released stackup, layer-by-layer impedance table, and expected report format. A stackup and coupon review before fabrication can flag which structures need their own test trace and which requirements call for separate validation. Request a manufacturing quote at [sales@bestpcbs.com](mailto:sales@bestpcbs.com), and include your quantities and approval date.
