StartMade Blog Episode 168 ·

The "Flame Retardant" Curtains Caught Fire After Two Washes

Flame Retardant and Flammability Testing Guide for Curtains, Fabrics, Toys and Plastics

Hello, this is StartMade.

A company that mostly fits out cafés imported 2,000m of blackout curtain fabric from a curtain fabric mill in Shaoxing (绍兴), Zhejiang. The curtains were for twelve franchise stores, and the purchase order said "阻燃面料 (flame retardant fabric)." The mill sent a "B1级" test report, and when the sample was held to a lighter, only the spot touched by the flame shrank and the flame went out quickly. Inspection passed everything too: color, width, weight, even the blackout rate.

The problem surfaced at the fire inspection before a store opening. Curtains hung in multi-use businesses in Korea must meet the national flame-retardant performance standard, and the only document the company had was the Chinese mill's "B1级" report. A rush test at a lab showed that new fabric barely passed, but curtains that had been washed twice and rehung at a store that had already opened far exceeded the char length limit. The fabric was not made from inherently flame-retardant yarn. It had a flame retardant applied to the surface, and the chemical had washed out. The mill contact said: "阻燃是后整理的,洗了会掉一些,很正常 (The flame retardant is a finishing treatment, so some of it washing out is normal)."

Flame-retardant performance cannot be seen or felt, and one flick of a lighter tells you even less. Yet when it fails, it leads straight to sales stops, reinstallation and recalls, and if a real fire happens, to questions of liability. This post covers what "flame retardant" actually refers to, standards that differ by product and market, checking methods and their limits, approximate criteria, gaps in test reports, common supplier substitutions, and what to cover in the spec sheet and at inspection.

Words like "阻燃", "B1级" or "V-0" guarantee nothing on their own. Write the test standard for your market and use, the flame-retardant method (inherent yarn or finishing treatment), the performance required after washing, and for plastics the material grade and thickness into the spec sheet, then test samples pulled from mass production. That is what keeps you from getting stuck at fire inspections, customs and sales channel reviews.

Note This post is based on common practices of fabric, toy and electrical product factories in Shaoxing and Haining (Zhejiang), Nantong (Jiangsu), and Dongguan and Shantou (Guangdong) as of November 2026, widely used industry criteria, and public standards such as Korea's flame-retardant performance standard issued by the National Fire Agency, UL 94, EN 71-2, US 16 CFR 1610 and California TB 117-2013. The figures in this post are approximate values used in practice. Flammability rules differ by product, installation site and sales market and change often, so always confirm the requirements that apply with the relevant fire authority, certification body and test lab.

1. Why Fabric That Passed the Lighter Test Still Burns

The most common check, both at the factory and during inspection, is holding a lighter to the corner of the fabric. If it does not catch or goes out quickly, someone writes "FR OK." But synthetic fibers like polyester often melt, shrink away from the flame and self-extinguish even with no flame-retardant treatment at all. So a lighter does a poor job of telling flame-retardant fabric from ordinary fabric.

Standard tests are a different thing entirely. A specimen of a set size is mounted at a set angle, a set burner flame is applied for a set time, and after the flame is removed the lab measures how many seconds it keeps burning, how many centimeters are charred, and whether melting drips carry the fire. Several specimens are tested, so both the average and the worst result matter. Only these numbers show the difference between fabric that barely passes and fabric that passes with margin.

Fabric that passes the test can also change over time. Fabric with a surface-applied flame retardant may pass when new, but loses performance through washing, dry cleaning, sunlight and abrasion. Inspection only looks at new fabric, so the goods ship without anyone checking how they perform after a few washes. Fiber composition testing is covered in Episode 156; this post focuses only on fire performance.


2. What "Flame Retardant" and "阻燃" Actually Mean

In Korea, treating fabrics, curtains and carpets so they resist catching fire is usually called "bangyeom," while the property of plastics or cables that do not burn easily is called "nanyeon." Chinese factories call both "阻燃." The same "阻燃" can mean very different methods, and the method decides both the performance after washing and the price.

MethodChinese factory termCharacteristicsWeakness
Surface finishing treatment后整理阻燃, 浸渍阻燃Ordinary fabric dipped in or sprayed with a flame-retardant solution, the cheapest optionPerformance easily lost in washing and dry cleaning, can stiffen the hand feel
Durable finishing耐久阻燃整理, 耐水洗阻燃Reactive flame retardant bonded to cotton or cotton blendsDemanding process with large quality variation, reduces fabric strength
Inherently flame-retardant yarn本质阻燃, 永久阻燃涤纶Fabric woven from polyester yarn spun with flame-retardant components, keeps performance after washingHigher fabric price, indistinguishable from ordinary fabric without yarn documents
Flame-retardant plastic阻燃料, V0料Resin grade compounded with flame retardants, rated by UL 94Rating changes with grade, thickness and color, drops when recycled material is mixed in
Flame-retardant foam阻燃海绵Polyurethane foam with added flame retardants or a high-density formulationThe flame retardant itself may be restricted

The "B1级" rating Chinese factories offer comes from China's building material fire performance classification (GB 8624) or its standard for flame-retardant products in public places (GB 20286). Its test methods and pass criteria differ from Korea's flame-retardant performance standard, the US NFPA 701 or the European EN 13773. "B1 means it is flame-retardant fabric in Korea too" simply does not hold.


3. Standards That Change by Product and Market

Flammability requirements change completely depending on what you sell, in which country, and where it will hang. The same polyester fabric faces different tests as a café curtain, a child's pajama or a costume. Here are the typical combinations.

ProductKoreaUSEU / UKWhat Chinese factories usually offer
Curtains, blinds, carpets (commercial)Fire flame-retardant performance standard, KFI flame-retardant performance inspectionNFPA 701, state and city fire codesEN 13773, EN 13501-1, UK BS 5867GB 8624 B1, GB 20286
Apparel, children's sleepwearCheck product-specific safety standards16 CFR 1610 (apparel), 16 CFR 1615/1616 (children's sleepwear)EN 14878 (children's nightwear)GB/T 5455 vertical burn results
Toys, costumes, play tentsFlammability items in the toy safety standardASTM F963 flammability itemsEN 71-2In-house EN 71-2 results
Sofa and mattress foam and coversCheck by productCalifornia TB 117-2013, 16 CFR 1633 (mattresses)UK furniture fire safety regulations (BS 5852 etc.)GB 17927
Plastic housings for electrical productsKC electrical safety standards (IEC-based)UL 94, product-specific UL standardsIEC 60695 glow-wire etc.Resin supplier's UL yellow card, GB/T 2408

Most problems arise in the gap between the far-right column and the rest. The factory sends whatever documents it has, and the importer believes those documents stand in for the sales market's standard. In Korea, curtains and carpets installed in places subject to flame-retardant rules, such as multi-use businesses, lodging and performance venues, must be products that passed the Korea Fire Institute (KFI) flame-retardant performance inspection, and a Chinese mill's report cannot replace it. Product certification in general is covered in Episode 10, and import requirements for children's products in Episode 65.


4. Checking Methods — From a Lighter to Lab Burn Tests

Ask "阻燃吗?" and the factory almost always answers "阻燃的." The only way to verify that is to test, and each method has clearly different limits.

MethodHow it worksWhat it tells youLimits
Lighter or match testHold a flame to a fabric corner or plastic pieceOnly whether a material obviously burns wellSynthetics melt and go out easily, so it barely discriminates; burn and fire risk
Standard burn test (fabric)Mount a specimen in a 45° or vertical tester and ignite with a burnerAfterflame and afterglow time, char length and area, melting dripsNeeds a lab; testing only new fabric says nothing about durability
Retest after washing or dry cleaningSame burn test after a set number of washesWhether flame-retardant performance survives washingWash conditions must be fixed for comparison; more time and cost
UL 94 burn test (plastics)Bar specimens mounted vertically or horizontally, flame applied twiceV-0, V-1, V-2, HB ratingsResults differ by specimen thickness; a finished housing cannot be tested as is
Glow-wire testA heated nichrome wire pressed against the product or a specimenWhether a housing ignites from an overheating partTest temperature and criteria must come from the product standard
Furniture ignition testA cigarette or small flame placed on the cushionWhether the foam and cover combination resists an ignition sourceTesting foam and cover separately loses much of the meaning
Flame retardant analysisXRF screening for bromine, chlorine and phosphorus; GC-MS for specific substancesWhether halogenated or restricted flame retardants are usedSays nothing about burn performance; targeted analysis is expensive

Burn It Again After Washing

For curtains, bedding, apparel and anything else that gets washed, results on new fabric are not enough. When you send samples to a lab, specify "after ○ water washes" or "after ○ dry cleanings" as well. Surface-treated fabric drops sharply at this stage, while inherently flame-retardant fabric barely changes. A retest after washing is also the surest way to confirm the flame-retardant method the factory claims.

A UL 94 Rating Comes With Thickness and Color Attached

The listing a resin supplier registers with UL, often called the "yellow card," gives a rating together with a minimum thickness, such as "V-0 at 1.5mm," and sometimes a color condition. A grade that is V-0 at 1.5mm can drop to V-2 or HB when molded at 0.8mm. When a factory sends a yellow card, first check that the grade name, thickness and color match the actual product. Plastic grade verification is covered in Episode 125.

The Flame Retardant Itself May Be Restricted

Resisting fire is not the end of it. Some brominated flame retardants that used to be common, and chlorinated phosphate flame retardants such as TCEP and TDCPP, are restricted in many countries. The EU sets limits for TCEP, TCPP and TDCP in toys, and some US states ban certain flame retardants in children's products and furniture. If you ask for fire resistance without setting chemical conditions, the factory will use the cheapest halogenated flame retardant. Hazardous substance testing in general is covered in Episode 165.


5. Approximate Criteria

Below is a summary of public standards and values commonly referenced in the industry. The criteria that actually apply depend on product, installation site and sales market, so before sending anything to a lab, decide which grade of which standard you need.

StandardMain pass itemsApproximate criteria
Korea flame-retardant performance standard (thin fabrics, curtains etc.)Afterflame time, afterglow time, char area and length, flame contact count, maximum smoke densityAfterflame within 20s, afterglow within 30s, char area within 50cm², char length within 20cm, 3 or more flame contacts (melting materials), maximum smoke density 400 or less
UL 94 V-0Afterflame time per specimen and in total, flaming drips10s or less per specimen, 50s or less for 5 specimens combined, no drips igniting the cotton below
UL 94 V-1 / V-2Same items30s or less per specimen, 250s or less in total; V-1 allows no igniting drips, V-2 does
UL 94 HBHorizontal burn rate40mm per minute or less for 3mm and thicker etc., the lowest rating
16 CFR 1610 (US apparel)45° burn timeClass 1 (normal) can be sold, Class 3 (rapid and intense burning) cannot
EN 71-2 costumesFlame spread rate30mm per second or less; above 10mm per second requires a "keep away from fire" warning

Criteria always need test conditions attached. Writing finished-product requirements with sample condition and pass criteria, as below, makes them usable straight away at inspection and when ordering lab tests.

Using These Figures The figures above are approximate values condensed for practical use from public standards and industry practice as of November 2026. Korea's flame-retardant performance standard uses different test methods and criteria for different items (thin fabrics, thick fabrics, synthetic resin boards, carpets and so on), and US fire codes vary by state and city. Before testing, confirm the exact standard and grade with the relevant fire authority, certification body or test lab.

6. Gaps in Flame Retardant Test Reports

Ask a factory for proof and it usually sends a report from the fabric mill or resin supplier. The documents themselves are rarely fake. The link between what the document covers and the product being made right now is simply broken.

GapWhat it meansWhat to check
Another country's standardResults under a standard other than the sales market's, like GB 8624 B1Results retested to the sales market's standard
Greige fabric reportFabric tested before dyeing, coating or blackout finishingA test on fabric after all finishing
Results before washing onlyResults for new fabric only, nothing after washingA retest with specified washing conditions
Different color or weightResults from white fabric applied to every colorMatch sample color and weight to the actual item number
Resin supplier's yellow cardA rating for the raw material grade, not your productMatch grade name, thickness and color; raw material purchase records
Missing componentsOnly the face fabric tested; lining, thread, tape and foam left outA test on the finished product as assembled
Old reportsResults from 2–3 years ago; flame retardant or raw material changed sinceIssue date and change history of materials and processes
Get Yarn Documents and Photos of the Flame Retardant Drums If the mill says the fabric is inherently flame-retardant, get the yarn supplier's delivery note and yarn test report. If it is a finished fabric, get the product name and lot number on the flame retardant drums and photos of the dipping equipment on the day of processing. For plastics, one photo is enough to confirm that the resin bag label next to the molding machine matches the grade name on the yellow card.

7. Common Supplier Substitutions

Inherently flame-retardant fabric costs considerably more than ordinary fabric, and flame-retardant plastic grades cost more than general grades. Cut the flame retardant concentration a little or mix in recycled material, and costs drop right away. Nobody knows the difference until a flame touches it. These are the types we have actually seen in consultations.

TypeHow it worksHow to prevent it
Inherent FR in samples onlySamples use inherently FR fabric; production uses ordinary fabric with a finishing treatmentYarn documents, retest after washing
Less flame retardantLower concentration or shorter dipping time to save costBurn test on production goods, check margin against the baseline
Only some colors treatedOnly the main color treated; other color lots skippedTest samples of each color
Switched FR gradeA general grade or recycled material instead of the V-0 gradeResin bag photos, burn test on production goods
Thinner wallsHousing molded thinner, outside the yellow card's thickness conditionMeasure minimum wall thickness, state the thickness condition
Switched foam or componentsFR foam replaced with ordinary foam, FR lining with ordinary liningSpecify each component, test the finished combination
Halogenated flame retardantsOrdered halogen-free but brominated flame retardants usedXRF bromine and chlorine screening, a flame retardant clause

What we see most often is samples made with inherently FR fabric and production switched to finished fabric. Both pass the test when new, so without the flame-retardant method and washing condition in the spec sheet there is nothing to hold the factory to. Requiring prior approval for process changes is covered in Episode 119.


8. The Cost When Flame Retardancy Is Missing

Flammability problems show up all at once at an inspection or after an accident, not gradually after sales, and they apply to the entire lot.

Recalls and product liability are covered in Episode 53, and claims and dispute resolution in Episode 36.


9. What to Put in the Spec Sheet

The two characters "阻燃" mean different things at different factories. Put the following items in a table and attach it to the spec sheet.

ItemExample
Applicable standard and gradeStandard name and grade for the sales market and use (e.g. Korea flame-retardant performance standard, UL 94 V-0, EN 71-2)
Flame-retardant methodSpecify inherent FR yarn / durable finishing / surface finishing, with yarn or flame retardant maker and product name
Wash durabilityPasses the same standard after ○ water washes or dry cleanings
Plastic materialResin maker and grade name, UL listed thickness and color conditions, no recycled material
Minimum thicknessMinimum housing wall ○mm or more (at or above the listed thickness)
ComponentsFlame-retardant conditions for lining, sewing thread, tape, foam and filling
Flame retardant chemistryNo halogenated flame retardants added, or flame retardants banned in the sales market not detected
Marking and labelsFlame-retardant marks, warnings and care labels per the sales market's rules
Change controlPrior written approval for any change of yarn, flame retardant, resin grade, thickness or components

The most effective items are the flame-retardant method and the wash durability condition. With those two lines, goods switched to finished fabric get caught right away in the post-wash retest. How to write spec sheets and BOMs is covered in Episode 59.


10. Adding Flammability Checks to Your QC Checklist

Sample Stage

The sample stage sets the baseline. Run the sales market's burn test on the approved sample, and for washable products, the post-wash retest as well. For plastic housings, compare the yellow card's grade, thickness and color against the actual sample, and if needed, cut specimens from a finished housing and send them for burn testing. Record actual figures rather than just "pass," so you can see how far production goods have slipped from the baseline. Sample management is covered in Episode 34.

Production Stage

On the days fabric finishing or molding runs, get photos or video of the yarn delivery note, flame retardant drum labels, dipping and drying equipment, and resin bag labels. If there are several colors, also get the processing schedule and lot numbers for each color. Production tracking is covered in Episode 102.

Pre-Shipment Inspection Stage

The inspector picks samples from packed finished goods, seals them and sends them to a lab. For fabric, pick samples by color; for plastics, by lot, and measure housing wall thickness at the same time. Do not use an on-site lighter check as grounds for any decision. Pre-shipment inspection is covered in Episode 100.

Sample Control Terms for Commercial Flame-Retardant Curtain Fabric The fabric is woven from inherently flame-retardant polyester yarn of the maker and product name stated in the spec sheet. It may not be replaced with surface-finished flame-retardant fabric. Fabric after all finishing (dyeing, blackout coating) must pass the sales market's flame-retardant standard both new and after 5 water washes. Any change of yarn, flame-retardant method, coating chemicals or processing conditions for each color requires prior written approval. Before shipment, burn tests are run on samples randomly picked for each color. If the standard is not met, the entire lot is remade or returned, and testing and logistics costs are borne by the supplier.

11. Sample Request Wording for the Factory

Putting the following wording in your purchase order or spec sheet takes the flame-retardant method out of the factory's discretion and makes it easier to assign responsibility if something goes wrong.

"本产品面料须使用规格书指定厂家及型号的本质阻燃涤纶纱线织造,不得改用普通面料加后整理阻燃。完成染色及遮光涂层后的成品面料,新布及水洗5次后均须符合销售国阻燃标准(以规格书注明的标准为准)。GB 8624 B1级报告不能代替上述测试。如需更改纱线、阻燃方式、涂层药剂或各颜色的加工条件,须提前书面确认。生产时请提供纱线送货单、纱线检测报告及涂层加工现场的照片。出货前由我方按颜色从成品中随机抽样送第三方检测,如不合格,整批重做或退货,检测及物流费用由供应商承担。"


12. Three Real Cases

These are anonymized versions of patterns we often encounter in consultations. Amounts and figures are examples.

Case 1 — Blackout Curtains for a Café Franchise

This is the case from the opening. The purchase order said only "阻燃面料," with no flame-retardant method, applicable standard or washing condition. The mill held its ground, saying "We gave you a B1 report, so we did our part," and in the end the company refitted all twelve stores with curtains that had passed Korea's flame-retardant performance inspection. Most of the imported fabric was sold off cheaply for home use, where flame-retardant rules do not apply. Now the spec sheet requires inherently flame-retardant fabric and a post-wash retest, and the domestic inspection schedule is built into the order timeline.

Case 2 — Children's Costumes Shipped to Europe

A seller making Halloween costumes in Nantong (南通), Jiangsu, for a European online channel was asked for EN 71-2 results during the channel's review. The report the factory sent covered only the cape's face fabric. When the finished costume was tested, flame spread quickly through the thin polyester lace trim sewn onto the cape. The factory said "面料是合格的 (The fabric passed)," but what gets tested is not the fabric but the whole costume a child wears. Removing the lace and retesting cost the seller most of the Halloween season.

Case 3 — Power Strip Housings With Recycled Material

A company importing power strips from Dongguan (东莞), Guangdong, had specified V-0 flame-retardant PC for the housing. On the third lot, the inspection report noted that the housing color was slightly yellowish. Specimens cut from the housing kept burning long after the flame was removed, and melting drips fell and ignited the cotton below, a result that did not even reach V-2. A factory visit revealed that, to hit the price, the factory was mixing nearly 30% recycled material of unknown origin into the flame-retardant PC. Had the color not looked slightly off, the lot would have gone straight to market.


13. Common Mistakes

The most common mistake is stopping at a lighter test. Synthetic fibers melt and go out easily even without treatment, so a lighter test tells you practically nothing. We also often see Chinese GB reports mistaken for the sales market's standard, only new-fabric results checked with no post-wash performance, and a resin supplier's yellow card used as the test result for the product. Testing only the face fabric and leaving out lining, trim or foam is common too, yet fires usually start at the weakest component.


14. Flame Retardancy Control Checklist

Sample Stage

Production Stage

Pre-Shipment Inspection Stage


Final Thoughts - Check Which Flame It Was Tested With, Not Just Whether It Burns

Fabric leaving the factory looks the same whether or not it was treated, whether or not the yarn is flame-retardant. Held to a lighter, it goes out much the same way. The difference shows up after a few washes, at a fire inspection or sales channel review, or in the worst case in a real fire. Writing the sales market's standard, the flame-retardant method and washing conditions into the spec sheet, and testing samples pulled from mass production, costs far less than reinstallation or a recall.

There is something you can do right now. Ask the factory currently producing for you whether the fabric is inherently flame-retardant or finished, and whether it has post-wash test results. If the only answer is "放心,有B1报告 (Don't worry, we have a B1 report)," draw up a flammability control sheet and send it before your next order.

StartMade supports you from confirming the sales market's standards and writing spec sheets with flame-retardant methods and materials, through reviewing supplier reports and checking processing sites during production, to pre-shipment inspection including sampling by color and lot and arranging burn tests. If you have already had a flammability problem, we can start by testing remaining stock; if you are preparing a new product, we can start together with confirming standards and the spec sheet.

Will Your Curtains and Housings Really Hold Up Against Fire?

From confirming market standards and specifying the flame-retardant method to report review, post-wash burn tests and pre-shipment inspection
We check before the fire inspection and the sales review

Phone   010-9980-9959
Email   deanpark@greenfrogseoul.com
KakaoTalk   pf.kakao.com/_NzSNX
Website   www.startmade.co.kr

Frequently Asked Questions

What is the difference between flame resistant and flame retardant?
Flame resistant usually describes materials that resist burning by nature, such as inherently flame-retardant polyester or aramid fibers. Flame retardant usually means a chemical added to or applied on a material to slow ignition and burning. Chinese factories call both "阻燃," so the word alone tells you little. What matters is which standard of which country the product was tested to and passed.
Do flame retardant curtains lose their effect after washing?
Fabric with a surface-applied flame retardant often loses performance through water washing or dry cleaning. Fabric woven from inherently flame-retardant yarn keeps its performance after washing. For curtains that will be washed, ask for retest results after washing as well.
What is the difference between UL 94 V-0 and V-2?
Both test vertical bar specimens with two flame applications. V-0 requires each specimen to stop burning within 10 seconds, while V-2 allows up to 30 seconds. V-2 permits melting drips that ignite the cotton below, while V-0 and V-1 do not. The rating depends on specimen thickness, so check that it applies at the thinnest wall of your actual product.