Does a TVS Diode Manufacturer Surge-Test Every Production Lot?

Do not assume that every TVS diode, or every production lot, receives the same high-energy pulse shown in the maximum-rating section of a data sheet. Routine electrical screening, lot sampling, qualification testing, special upscreening, and customer application validation are different controls. A supplier audit should identify which one applies to the exact orderable part and shipment.

The question often arrives in an abbreviated form: “Is every lot surge tested?” A yes or no answer is almost useless. The test waveform, peak current, pulse count, interval, sample selection, pre- and post-stress measurements, temperature, and rejection rule determine what the record can support.

Separate a rated capability from a release test

A TVS data sheet may list peak pulse power, peak pulse current, clamping voltage, breakdown voltage, and leakage. Pulse values are tied to stated waveforms and conditions. They describe device behavior or capability under those conditions. They do not automatically describe the factory test flow for every shipped unit.

Breakdown voltage and reverse leakage can be measured quickly under controlled conditions and may be suitable for production screening. A high-energy pulse creates heating and may consume part of the device's stress margin. A manufacturer may use sampling, periodic reliability testing, qualification evidence, or a special screening flow instead of applying the maximum pulse to every device.

Some high-reliability TVS products are sold with additional screening. Those products demonstrate why the exact ordering code and procurement specification matter. An optional 100% screen on one series cannot be generalized to a manufacturer's full catalog.

Five records that buyers frequently treat as one

Production electrical test

This record shows whether measured parameters fall within release limits. It should identify the test temperature, limits, equipment or program revision, lot, and disposition. It may include breakdown voltage and leakage. It does not prove that every unit received the peak surge used for a published rating.

Lot acceptance or periodic sampling

A defined sample is tested to monitor a production lot or process. The sampling plan, acceptance number, test interval, and action after a failure are part of the evidence. A passing sample supports the lot under that plan; it is not the same as individual-unit screening.

Qualification or reliability report

Qualification records typically list stress methods, sample sizes, lots, duration, temperature, and electrical criteria. Public semiconductor reliability reports often show separate rows for temperature cycling, humidity, high-temperature bias, soldering, or other stresses. A report may cover a product family, a technology platform, or a named part. The coverage statement must be checked before applying the result to an orderable device.

Special screening or upscreening

An added procurement flow may include tighter parameter limits, additional electrical tests, surge screening, traceability, or other controls. It should appear in the order code, drawing, purchase specification, or certificate. A sales description without the process revision and acceptance criteria is not enough for incoming inspection.

Application validation

The customer's source impedance, waveform, PCB, ambient temperature, repetition rate, and protected circuit define the application result. This test answers whether the selected part works in that product. It cannot be replaced by a generic qualification report, and it should not be presented as a new factory rating.

Send a field list instead of asking for “the surge report”

A broad request tends to return a catalog page or a family-level presentation. A structured request is easier to review and easier for the supplier to answer.

Request the following information:

1. exact orderable part number, package, die or process revision, and report coverage;

2. routine production parameters and limits at each stated temperature;

3. lot-sampling plan, test frequency, acceptance rule, and failure response;

4. pulse waveform, source impedance, peak current, duration, interval, and count;

5. sample size, number of lots, lot identifiers, and sample selection method;

6. pre- and post-stress measurements, including leakage and breakdown conditions;

7. failure definition for short circuit, elevated leakage, parameter drift, or visible damage;

8. equipment identifier, method revision, date, and approval record;

9. change-notification rule for die, assembly, material, test program, or manufacturing site.

If a document lacks a part number and date, classify it as background information. If it contains clear conditions but uses a waveform unlike the application, retain it as device evidence while marking the application gap. Good document control preserves useful evidence without stretching its conclusion.

Why a passing breakdown-voltage check is incomplete

Breakdown voltage is measured at a specified current. A surge event is affected by dynamic resistance, current rise, pulse width, loop inductance, temperature, and repetition. Two lots can meet the same breakdown limits while showing different leakage shifts after stress or different clamping behavior under a fast pulse.

Incoming inspection based on breakdown voltage can find wrong parts and gross defects. It does not reproduce every transient characteristic. A layered control is more practical: use stable electrical parameters to identify incoming lots, apply pulse sampling when required by the control plan, measure parameters again after stress, and validate the installed part under the product's actual transient.

“The board still powers up” is also an incomplete post-stress check. A TVS may develop higher reverse leakage or a shifted breakdown characteristic while the protected circuit continues to operate. Post-stress measurements need a defined waiting time and temperature, since a hot device can produce a misleading comparison.

For repetitive pulses, record the interval and case or junction-temperature assumptions. Repeating a single-pulse rating without thermal analysis can turn a valid data-sheet number into an invalid test.

Connect supplier documentation to a controlled sample test

After document review, select samples from identifiable lots and run a controlled comparison. Keep the pulse generator, fixture, cabling, probe locations, ambient conditions, and measurement sequence fixed. Store the raw voltage and current records, not a cropped screenshot with only a pass label.

When qualifying a replacement, an A/B/A sequence is useful. Test the approved source A, candidate B, and a restored A condition. The restored measurement checks for fixture wear, calibration drift, rework damage, and order effects. If A changes after the test sequence, the comparison needs investigation before B is accepted or rejected.

Power calculations require time-aligned voltage and current measured for the same pulse and stated reference points. Multiplying a device voltage by the generator's nominal current can overstate or understate actual device stress when the network divides current through other paths.

The test report should also preserve negative results. A candidate that meets breakdown voltage but develops excessive leakage after the requested pulse has provided valuable selection evidence. Hiding that result behind a family qualification report makes the next investigation slower.

Write an approval statement with a limited scope

A defensible approval statement could read: the shipment passed the listed production electrical controls; pulse capability is supported by the named part-specific report and the attached application test conducted on the stated board revision. That sentence distinguishes lot release, component evidence, and system evidence.

Avoid statements such as “the manufacturer surge-tests every lot, so the part cannot fail.” Unless the control plan specifies the sampling or screening and the records are traceable to the shipment, the statement exceeds the evidence. Even a screened part can fail when the product applies a different waveform, repetition rate, thermal condition, or current path.

ASIM separates production-control records from application data by part number and revision. Buyers should preserve the same links in their approved-vendor file: order code, shipment lot, supplier record, PCB revision, waveform, and test outcome. The result is a chain that an engineer can audit after a field event, rather than a folder of unrelated certificates.

The practical answer to the title question is therefore conditional. A manufacturer may use unit screening, lot sampling, qualification tests, special upscreening, or a combination. The exact control must be stated for the exact part. Asking for the conditions is more useful than asking for a one-word promise.