Quick answer: use stress measurement as process evidence, not as a stand-alone safety promise

Surface stress measurement helps an OEM judge whether heat-treated appliance glass received a controlled tempering process. A useful specification identifies the glass construction, measurement method, instrument range, measurement side, locations, sampling plan, reporting unit and acceptance rule. The result should then be reviewed together with thickness, geometry, edge condition, flatness, fragmentation or impact evidence, and the final appliance application.

Stress data alone cannot guarantee that every panel will survive installation or service. Breakage can still originate from edge damage, contact with metal, an incorrect cutout, assembly preload, thermal gradients or handling after factory inspection. Conversely, a fragmentation test on a few broken specimens does not show how stable the non-destructive stress result is across routine production. The two forms of evidence answer different questions.

ASTM C1279-23 describes non-destructive photoelastic measurement of edge and surface stresses in annealed, heat-strengthened and fully tempered flat glass. Its official scope also states important limitations, including the need for light-transmitting glass and the fact that the described surface measurement applies to the tin side of float glass. An appliance project should confirm that the chosen method is applicable to its printed, coated, shaped or darkened panel before copying any value into a purchase specification.

Why tempering creates measurable stress

During thermal tempering, glass is heated and then cooled rapidly. The surfaces cool and stiffen before the interior. As the core continues to cool, a residual stress profile forms: the surfaces are placed in compression while the interior is placed in tension. Surface compression helps the glass resist tensile stress introduced by use, handling and moderate impact.

This simplified explanation is useful, but the actual distribution depends on:

  • glass thickness and composition;
  • furnace temperature and residence time;
  • convection and radiation balance;
  • quench pressure and nozzle condition;
  • panel size, shape, holes and cutouts;
  • printed areas that absorb heat differently;
  • loading pattern and panel spacing;
  • orientation in the furnace;
  • cooling uniformity near edges and corners.

Therefore, the buyer should treat surface stress as a characteristic of a specific part and process window—not as a generic certificate attached to any tempered glass.

Surface stress, edge stress and central tension are not interchangeable

The terms are sometimes used loosely in supplier discussions. They should be separated.

Surface stress

Surface compressive stress is measured at or near the glass surface by an applicable photoelastic method. It is often used as non-destructive evidence of heat treatment. The method, optical contact fluid where applicable, refractive-index assumptions and measurement side affect the result.

Edge stress

Edge stress is measured through the edge using a compatible instrument and prepared optical path. It can provide useful information near a region where damage often initiates, but edge finish, bevel geometry, printing and accessibility may restrict where a valid reading can be taken.

Central tension

Central tension exists inside the thickness and influences fragmentation behavior. It is not simply the same number as surface compression with a different sign. Inferring it requires an appropriate stress model, thickness information and measurement approach.

If a report lists only “tempering stress” without defining which quantity was measured, the report is incomplete for technical approval.

Start with the product application and risk

A surface-stress plan should reflect how the glass is used. Consider four examples:

Application Stress-related concern Additional evidence needed
Oven door or fascia Thermal gradient, edge support and hot-zone printing Thermal-cycle or thermal-shock validation, assembly clearance, edge inspection
Washing-machine lid Repeated closing, local hardware load and large area Impact/load validation, flatness, hinge and bumper review
Microwave door panel Large printed mask, viewing window and frame retention Print durability, impact/fragmentation evidence, fit and edge protection
Display/control cover glass Optical appearance, bonding flatness and cutouts Optical inspection, bonding validation, dimensional and functional checks

The OEM should define which failure is being prevented. “High stress” is not a complete design objective. The intended result is a stable process that meets the released product’s safety, optical and assembly requirements.

Confirm method applicability before measuring

Photoelastic instruments rely on light interacting with stressed glass. Several part features can interfere:

  • dense black ceramic or organic ink;
  • mirror or conductive coatings;
  • textured or anti-glare surfaces;
  • curved geometry;
  • very small panels or narrow borders;
  • holes and closely spaced cutouts;
  • laminates or bonded assemblies;
  • unknown float-glass orientation;
  • surface contamination or protective film.

Create a measurement feasibility map during sample development. Mark accessible transparent zones, the tin side, edge locations, printed boundaries and fixture restrictions. If the production part cannot be measured at the desired location, agree on a representative witness coupon only when correlation to the real part has been demonstrated. A coupon should not silently replace production-part evidence.

Define a repeatable measurement plan

The plan should be controlled like any other critical inspection method.

Identify the instrument and method

Record the instrument type, model, measurement range, calibration or verification status, software version if relevant, contact fluid, reference standard and operating instruction. State whether the result is surface stress, edge stress or another photoelastic quantity.

Define the measurement side

For float glass, the air side and tin side are not interchangeable for every method. Marking or identifying orientation before printing and tempering can prevent confusion later. The report should state the side actually measured.

Fix the locations

Use drawing coordinates or a controlled sketch. Typical plans may include the center, quadrants, near-edge positions and zones around major cutouts, but the correct pattern depends on part geometry. Avoid vague instructions such as “measure five points.” Two inspectors can choose very different five points.

Control surface condition

Define whether protective film is removed, how the surface is cleaned, allowable temperature, stabilization time and how remeasurement is handled. Residue, scratches or a coating boundary can disturb optical contact or reading quality.

Record individual data

Do not report only an average. Keep part identification, lot, furnace run, rack position when available, point-by-point readings, invalid readings, operator, date and instrument ID. Averages can hide a low corner or a systematic left-to-right difference.

Validate the measurement system before using capability statistics

An instrument display with several decimal places is not proof of measurement capability. Before applying Cpk, control limits or supplier acceptance to stress data, examine repeatability and reproducibility on representative parts.

A practical study can include:

  1. parts spanning normal process variation;
  2. multiple trained operators;
  3. repeated measurements at clearly marked locations;
  4. removal and repositioning between trials;
  5. the same cleaning and contact procedure used in production;
  6. review of invalid or unstable readings;
  7. assessment of resolution relative to the engineering tolerance.

This does not require inventing a universal Gage R&R threshold. The customer and supplier should agree how much measurement variation is acceptable for the decision being made. For a deeper framework, see our measurement system analysis and Gage R&R guide.

Build acceptance criteria around the released design

Do not copy a surface-stress number from architectural glass, a different thickness or an unrelated supplier report. Establish the rule from:

  • the applicable customer drawing and material specification;
  • legal and safety requirements for the final appliance market;
  • the selected glass type and thickness;
  • validated prototype and design-verification results;
  • the approved measurement method;
  • capability demonstrated on production-intent tooling;
  • correlation to fragmentation, impact and field-relevant tests.

Where no product-specific standard gives the acceptance value, the drawing can state a customer-approved range rather than an unsupported “minimum only.” An excessively broad range may allow process drift; an unnecessarily tight range may add inspection cost without improving product function.

Correlate stress data with destructive evidence

Non-destructive stress measurement is valuable because production panels remain usable. Destructive tests remain necessary when the program must verify break pattern, impact behavior or complete assembly performance.

Use correlation lots during qualification:

  • measure stress at defined locations;
  • record thickness, flatness, edge condition and print configuration;
  • select identified specimens for destructive tests;
  • photograph and classify the result under the agreed method;
  • compare results without assuming one-to-one prediction;
  • investigate outliers and location effects;
  • preserve raw data and furnace traceability.

Our impact and fragmentation testing guide explains how to prevent a broken-sample test from becoming an undocumented pass/fail ritual.

Understand what stress measurement cannot detect

A conforming reading does not reveal every defect. It may not detect:

  • a fresh edge chip created after measurement;
  • deep handling scratches outside the measurement point;
  • point contact caused by the appliance frame;
  • an undersized assembly gap;
  • uneven adhesive or fastener preload;
  • a later thermal event beyond the validated range;
  • inclusion-related spontaneous breakage risk;
  • local distortion that affects display bonding;
  • incorrect glass chemistry or thickness unless separately checked.

This is why shipment inspection must combine dimensional, cosmetic, edge, print, flatness and traceability evidence. Stress measurement is one control in the plan, not the whole control plan.

Use process data to find furnace patterns

Point-by-point readings become more useful when linked to production context. Review trends by:

  • furnace recipe;
  • load position;
  • orientation;
  • part family and thickness;
  • print coverage;
  • shift and operator;
  • quench maintenance interval;
  • measurement instrument;
  • raw-glass lot where relevant.

A stable lot average can coexist with a recurring low zone on one rack position. Heat maps and run charts can reveal this pattern earlier than a shipment-level certificate. If capability indices are used, confirm distribution, stability and measurement adequacy first; see our Cpk and SPC guide for appliance glass.

Supplier evidence package for OEM review

A credible qualification package may include:

  • released drawing and stress characteristic definition;
  • method reference and project-specific work instruction;
  • instrument identification and verification record;
  • measurement-side confirmation;
  • location map;
  • MSA or repeatability evidence;
  • production-intent sample results;
  • furnace lot and rack traceability;
  • correlation to fragmentation or impact tests;
  • nonconformance and reaction plan;
  • change-control triggers;
  • shipment certificate format.

The package should distinguish measured facts from assumptions. For example, “surface stress measured on the tin side at six mapped positions” is auditable. “Optical-grade tempering” is not auditable unless the term is defined by measurable requirements.

Buyer’s RFQ checklist

Send the supplier:

  1. glass composition, thickness and finished dimensions;
  2. complete CAD and drawing with datums;
  3. hole, cutout, edge and corner requirements;
  4. print and coating artwork;
  5. application, support and thermal environment;
  6. applicable standard or customer method;
  7. stress quantity, unit, range and measurement locations;
  8. sampling frequency and lot definition;
  9. destructive validation requirements;
  10. report and traceability format;
  11. approved-deviation process;
  12. engineering-change notification requirements.

If the value is not yet known, request a feasibility study rather than forcing the supplier to guess. Tairong can review drawings for custom oven glass, washing-machine glass and control-panel glass before the inspection agreement is released.

Common specification mistakes

Writing only “fully tempered”

This does not define the evidence, locations, sampling or applicable product standard.

Measuring through unsuitable printing or coating

The reading may be unavailable or method-dependent. Confirm feasibility on the real decoration stack.

Reporting only lot averages

The average can hide local nonuniformity and outliers.

Treating stress as guaranteed product life

Final reliability depends on edges, geometry, assembly and environment as well as tempering.

Changing thickness or artwork without review

Thermal absorption and cooling response can change. The approved recipe and evidence may no longer apply.

FAQ

Is surface stress measurement destructive?

The photoelastic approach described by ASTM C1279 is non-destructive, but method applicability and surface access must be confirmed for the actual part.

Can surface stress replace fragmentation testing?

No. Stress measurement provides non-destructive process evidence; fragmentation evaluates break behavior on sacrificed specimens. A qualification plan may need both.

Where should stress be measured on printed appliance glass?

Use mapped transparent locations that are compatible with the instrument, including risk-relevant zones where feasible. Do not choose locations after seeing the result.

Should every panel be measured?

Not automatically. Frequency should reflect product risk, process capability, measurement time, customer requirements and the reaction plan. Some critical programs may require higher frequency than stable, mature parts.

Does a higher number always mean better glass?

No. The target must match the material, thickness, geometry, distortion limits and validated appliance design. Process stability and overall product evidence matter more than pursuing an isolated maximum.

What should a supplier certificate contain?

It should identify the part and lot, method, instrument, measurement side and locations, individual or agreed summary results, acceptance rule, date and responsible inspector.

Conclusion

Surface-stress measurement is most valuable when it is specific, repeatable and connected to the finished appliance. Define the quantity, method, side, locations and acceptance rule; validate the measurement system; correlate non-destructive readings with destructive and assembly evidence; then monitor trends by furnace lot. This approach turns an ambiguous “tempering certificate” into decision-ready OEM evidence.

Send your drawing, thickness, printing stack, application and customer requirements through our contact page. Tairong’s engineering team can prepare a measurement-feasibility review and a project-specific inspection proposal before quotation.