Two export quotes can both say “QC passed” and still leave you opening crates with dead lamps, loose power connectors, weak PSUs, or patchy brightness — because nobody named hours, patterns, temperature, pass/fail rules, or photo evidence. Factory aging (burn-in) is the boring step that should kill infant failures before freight, not during your first load-in.
Our CE / FCC / EMC export RFQ guide already owns market-access certificates. Our spare-parts / hot-swap guide owns inventory after the wall is hung. Our calibration guide owns D65 / gamma / coefficients after hang. This post owns only the pre-shipment burn-in story: what public OEM and FAT practice actually describe, which defects aging catches, and which RFQ lines force a real evidence pack before you compare prices.
Quick definitions (clear the name collision)
| Term | What it means | What it is not |
|---|---|---|
| Aging / burn-in (老化 / 烧机) | Continuous powered run of the wall (or agreed build) through stress patterns so early failures appear in the factory | Not a customs certificate |
| Factory acceptance test (FAT) | Buyer-witnessed or video-witnessed checks before ship — often includes aging evidence | Not the same as on-site SAT alone |
| Pattern cycle | Ordered colour / gray fields (commonly R → G → B → white → grayscale) with a named dwell | Not a single splash screen photo |
| Elevated ambient | Aging run at a higher room / chamber temperature than office comfort (OEM write-ups often cite roughly 40–50°C class language) | Not outdoor IP rating |
| Pass criteria | Written fail rules: dead lamps, clusters, brightness uniformity band, flicker / scan lines, PSU dropouts | Not “looks fine on WhatsApp” |
| Evidence pack | Time-stamped photos/video, start/end log, cabinet serial list, signer, retest after repair | Not a one-line “QC OK” on the packing list |
| Modules-only aging | Stressing loose modules or partial racks instead of the as-built wall | Not proof the shipped cabinets survived as a system |
| Infant mortality | Early-life LED / connector / PSU / solder faults that show under heat + continuous drive | Not wear-out after years of touring |
Industry language is blunt: aging is invisible insurance — especially for rental fleets that will see rough transport and long show hours.
Aging vs first power-on — fail in the factory, not at load-in
A wall that has never been burned in is asking the venue to be the test chamber. Dead pixels, intermittent power connectors, weak solder joints, and marginal PSUs love to appear after long continuous drive and thermal stress. That is exactly when your crew is on the clock and the client is watching.
Buyer-facing OEM and FAT writing through 2026 keeps teaching the same sequence:
- Build the wall (or the agreed FAT configuration) in the factory.
- Run continuous power through a named colour / gray cycle.
- Hold for a named duration (often many hours; see next section).
- Log pass/fail by cabinet serial, repair, retest, then pack.
- Hand the buyer photos/video + logs — not a slogan.
If your quote only says “100% QC” with none of those fields, you are comparing poetry, not products.
Worked illustration (literacy, not a certificate)
Imagine two identical-looking P2.6 rental bids for a 50-cabinet tour wall:
| Quote line | What you actually get | Risk at first show |
|---|---|---|
| “QC OK / aged” (no hours) | Unknown dwell; maybe a short smoke test | Early dead lamps and PSU drops on day one |
| Named cycle + 72h-class continuous run + serial log + photo pack | Failures pulled and retested before crate | Load-in is assembly + calibration, not first burn-in |
Treat multi-day burn-in hours as a planning example from public OEM practice, not as a universal law. Prefer writing your hours, patterns, and temperature on the PO.
What “72 hours” actually means
Public OEM aging articles in 2026 (for example a 16 Jun 2026 OEM burn-in explainer) describe a concrete package buyers can copy into RFQ language:
| Field | Typical public OEM framing | Buyer write-up tip |
|---|---|---|
| Duration | 72 consecutive hours of powered aging | Write your minimum hours; ask what they actually run |
| Patterns | Rotate red, green, blue, white, and grayscale | Demand order + dwell per field, not “colour test done” |
| Ambient | Elevated chamber / room (OEM text often cites about 40–50°C) | If they claim elevated ambient, ask how it is measured |
| Power | Continuous drive — not intermittent demo loops | Ban “we power it for photos only” |
| Outcome language | Zero dead pixels; brightness uniformity within a stated band (OEM examples use language such as ±2%); no flicker / obvious scan artefacts | Copy measurable rules onto the PO |
Honesty: 72 hours here is a common public OEM example, used so buyers have a concrete starting point. It is not an SFxDisplay lab certificate, and it is not a promise that every SFx quote includes a 72-hour chamber run. Some projects need longer; some fixed indoor walls agree shorter FAT with stricter photo evidence. The RFQ skill is to name the number, not to worship one vendor’s brochure hour count.
Defects aging is meant to catch
A proper burn-in is not magic QA theatre. It stresses the build so infant faults appear while the factory can still swap parts:
| Symptom during / after aging | What it often points to | Why the factory is the right place |
|---|---|---|
| Dead or weak lamps | LED / solder / driver channel infant fail | Module swap + retest before freight |
| Colour-channel soft fields | One primary under-driven or failing | R/G/B fields make it obvious |
| Patchy brightness / blotchy white | Module mix, PSU sag, cable resistance | Uniformity check under continuous white |
| Flicker or scan lines on gray | Driver / receiving-card / power interaction | Gray ramp is harsher than a logo loop |
| Random reboot / black cabinet | Marginal PSU, connector seating, overheating | Heat + hours beat a 10-minute demo |
| Loose power / data connectors | Assembly torque / latch issues | Vibration later will finish the job on tour |
Aging does not replace: after-hang calibration, on-site dead-pixel sign-off (separate hang acceptance), IP weather design, or CE/FCC paperwork. It only owns early-life survival before the crate leaves.
Pass criteria buyers can write
Copy rules that a third party could score:
- Dead lamps — zero allowed on ship, or a written cluster rule (e.g. no adjacent dead pair; max N per cabinet) with mandatory module replace + re-age.
- Brightness uniformity — state a band on full white (OEM literacy often uses a tight percentage-class claim; write the number you will accept).
- No flicker / crawling scan artefacts on gray fields at the agreed refresh / performance profile.
- No PSU dropout / cabinet blackout across the full duration.
- Log by cabinet serial — every fail, repair part, and retest result tied to an ID that matches the packing list.
- Configuration lock — receiving-card performance settings, brightness mode, and firmware/parameter file used during the pass.
If two suppliers bid the same pitch and cabinet size but only one accepts these lines, the cheaper quote is not cheaper — it is unfinished.
Evidence pack before deposit (or before ship)
FAT checklists in the industry cluster around the same artefacts. Ask for:
| Artefact | Why it matters |
|---|---|
| Time-stamped photos / video of the wall on R, G, B, white, and gray | Proves the cycle happened |
| Start and end log (date/time, ambient if claimed, operator) | Proves duration |
| Cabinet serial list with pass / fail / retest | Ties evidence to crates |
| Repair record (what was swapped) | Stops silent module roulette |
| Signer (factory QC + optional buyer remote witness) | Accountability |
| Retest after repair | One pass before a swap is not a pass after |
Deposit leverage differs by deal structure — some buyers hold a slice until the evidence pack lands; others require video FAT before balance payment. Write the trigger once on the PO so sales chat cannot redefine it later.
Rental vs fixed — and the “modules only” caveat
| Buyer type | Why aging weight changes | Extra ask |
|---|---|---|
| Touring / rental fleets | Rough transport + long show hours; reputation dies on the first black cabinet | Prefer wall-as-built aging; spare modules from same aged batch (spare parts) |
| Fixed install (church, corporate, control) | Less vibration, but still hate early lamp death after scaffolding leaves | Can accept shorter FAT if evidence + on-site acceptance are strict |
| Export importers | You cannot easily return a container of infant fails | Evidence pack before balance; serial match to packing list |
Modules-only aging caveat: some factories age loose modules on racks, then assemble cabinets later. That catches many LED faults — and misses cabinet PSU, wiring, and connector issues that only appear when the wall is powered as a system. If the seller says “modules aged,” ask:
- Were cabinets powered as a wall (or agreed FAT grid)?
- For how long after final assembly?
- Do serials on the evidence match the shipped cabinets?
Traps that look like “QC ready”
| Trap | What goes wrong | What to ask |
|---|---|---|
| “Aged” with no hours | Ten-minute smoke test | Named continuous duration |
| Logo loop only | Hides primary/gray stress | R/G/B/white/gray cycle + dwell |
| Office-temperature photo | No thermal stress | Ambient / chamber condition if claimed |
| Modules only, then silent rebuild | Cabinet-level faults ship | Wall-as-built (or stated FAT grid) after assembly |
| Pass without serials | Cannot match crates | Cabinet ID log + packing list |
| Repair without re-age | New infant fail on the swap | Mandatory retest window |
| WhatsApp stills of a different wall | Evidence theatre | Time-stamped video of your serials |
| Aging waved as calibration | White point / gamma still unfinished | Keep calibration as a sister obligation |
| Aging waved as CE/FCC | Different paperwork | Keep CE/FCC/EMC separate |
| Comparing unequal bids | Cheap wins; fails on day one | Normalize hours + evidence before price |
RFQ checklist (copy onto the PO)
Ask every supplier the same lines so aging claims are comparable:
- Hours — continuous burn-in duration (use a public OEM example such as 72h only as a starting benchmark; write the number you require).
- Patterns — R / G / B / white / grayscale order and dwell per field.
- Temperature — ambient or chamber condition if elevated aging is claimed; how it is logged.
- Build under test — full wall, FAT grid size, or modules-only (state which — prefer as-built cabinets for rental).
- Pass/fail rules — dead-lamp / cluster policy; uniformity band; no flicker / PSU dropout.
- Serial log — cabinet (and critical module) IDs matching the packing list.
- Evidence pack — time-stamped photos/video, start/end log, signer, repair + retest record.
- Configuration lock — performance profile / firmware / parameter file used during the pass.
- Spares sister — same-batch aged spares so a hot-swap does not reintroduce infant risk (spare parts).
- Certs sister — model-matched CE/FCC/EMC package (export RFQ).
- After-hang sisters — calibration / coefficient handover; on-site pixel acceptance when that sister is live.
- Do not compare panel price alone until hours + evidence fields match.
Related reading (live SFxDisplay)
| Post | What it owns vs this article |
|---|---|
| CE / FCC / EMC export RFQ | Market-access certificates and test reports — not burn-in hours |
| Rental spare parts & hot-swap | After-hang inventory % and swap discipline |
| LED wall calibration: gamma & white balance | D65, gamma/EOTF, coefficient handover after hang |
| LED wall resolution & pixel map | Canvas / mapping literacy — not factory aging |
| Scan vs refresh vs frame rate | Timing clocks / camera flicker |
| 3840 Hz vs 7680 Hz | Refresh number literacy for filming |
| IP54 vs IP65 | Weather ingress — not burn-in |
| Front vs rear service | Access direction for maintenance |
| Soft vs hard connection rental cabinets | Cabinet connection style |
| Outdoor festival rental checklist | Event logistics checklist |
| How to choose an LED display | Broad selection after QC fields are written |
| Rental series | Product starting point after aging/evidence terms are agreed |
(pending articles — dead-pixel acceptance, low-brightness grayscale, wind/ballast, flight cases, 90° corners — stay unlinked until live.)
SFxDisplay honest boundary
We build and quote LED cabinets for buyers who need walls that survive first power-on after freight, not brochure poetry. We will help you write aging hours, pattern cycles, pass/fail rules, serial logging, and evidence-pack language when you ask. We will not invent a public SFxDisplay laboratory certificate that every pitch is burned for 72 hours. We will not pretend aging replaces CE/FCC paperwork, after-hang calibration, or on-site pixel acceptance. We will not name competitor factory brands as heroes. We will not treat “modules aged” as automatic proof that assembled cabinets passed as a system. If the wall is a touring rental fleet, an export container, or a fixed install you cannot easily re-scaffold, say so early — that changes FAT hours and evidence, not only the pitch number on the brochure.
Browse the rental series and how to choose, then paste the checklist onto the PO. Or contact SFxDisplay — WhatsApp +86 150 1911 0341 · molly@sfxdisplay.com.
Bottom line
“QC OK” is not a test plan. Factory aging / burn-in exists so infant LED, connector, driver, and PSU faults die under continuous colour and gray stress — ideally over a named multi-hour window such as the 72-hour package described in public OEM writing — before crates leave China. Write hours, patterns, temperature, pass/fail, serials, and photo/video evidence onto the RFQ; prefer wall-as-built aging for rental; retest after every repair. Keep certificates on the CE/FCC guide, spares on the hot-swap guide, picture setup on the calibration guide; keep pre-shipment burn-in honesty here — so load-in is assembly, not your first aging chamber.
FAQ
What is an LED display aging test / burn-in? A continuous powered run through stress patterns (commonly red, green, blue, white, and grayscale) so early-life faults appear in the factory. It is pre-shipment reliability literacy — not a CE/FCC certificate and not after-hang calibration.
Why do OEM guides mention 72 hours? Public OEM articles in 2026 (for example public OEM write-ups dated 16 Jun 2026) describe 72 consecutive hours with a full colour/gray cycle and often elevated ambient language. Use that as a concrete buyer benchmark, then write the hours you need on the PO. It is industry-practice literacy, not an SFxDisplay guarantee for every quote.
What patterns should I demand? At minimum: full-screen red, green, blue, white, plus a grayscale ramp or stepped gray fields, with named dwell times. Ask for the order — not only “colour test done.”
Is modules-only aging enough? It catches many lamp faults but can miss cabinet PSU, wiring, and connector issues. For rental and export walls, prefer aging the assembled cabinets as a wall (or a stated FAT grid), then log those serials.
What evidence should arrive before deposit or ship? Time-stamped photos/video of the pattern cycle, start/end log, cabinet serial pass/fail list, repair + retest notes, and who signed. Without serials, stills of “a” wall are weak.
How is this different from on-site dead-pixel acceptance? Aging tries to kill infant fails before freight. On-site pixel counts catch what survived shipping and assembly. Keep both — do not let factory aging language cancel your hang acceptance.
Does aging replace calibration? No. Burn-in stresses survival; calibration sets white point, gamma, and coefficients for the hung wall.
