上个月，一个做 PLC 的客户发来一段手机视频。夜里拍的，机房里灯没开全，一台壁挂工控机的侧面翘开两毫米，翘的正好是 DIN 导轨卡扣那一排。
他配了一句话："原来是阻燃 PC，这回换了个阻燃更好的料，结果装进柜子卡扣一按就响。"
工控机外壳换料，他动的是料，没动的是三件事：卡扣的韧性、导轨的受力方式、还有装配时那个季节的温度。
我问了他三句：换的是阻燃体系还是基材？卡扣和壳体是一套料还是两套料？低温冲击做过没有？
他停了两秒，说后两句他都没想过。
这个回答不奇怪。工控外壳的换料现场，多数人是从阻燃等级那一栏看起，也在那一栏结束。
下面这条时间线，就是他那个件的经过。
起点：新料打样件外观合格，无卤阻燃报告拿到 V-0，项目组松了口气，评审会上过了。
潜伏：货发到北方客户机房，冬天夜里机房温度到 5℃ 以下，机器每天开关机三四次，卡扣跟着做冷热循环。
爆发：装机第三周，售后反馈卡扣断。比例不高，十几台里有一两台，可客户是运营商，投诉直接报到项目组。
结算：800 台外壳返工，卡扣位置改料重打，跟着重跑的还有一份阻燃认证。
换料这笔账，很少记在料价那一栏，多半记在这种"没人重排的工序"上。
一、换料前的六维工况，四样得先落到数字
其一，温度和它的形状。工控柜内长期在 50–70℃，靠电源模块和变频器那一侧局部能到 85–95℃。这不是一个高数字，但它连续，一天二十四小时不停，机器开八年，这个温度就陪你八年。
宽温要求通常写 -20℃ 到 70℃；做轨道交通或户外柜的，会写 -40℃ 到 75℃。两头都要过，中间那个常温值反而最不重要。
其二，电。这一维在工控件上比机械载荷硬。外壳要 UL94 V-0，标准厚度 1.6 mm；靠近端子的薄壁件要 0.8 mm 也能 V-0。
阻燃只是入场券。带电回路附近还要看灼热丝，通常要 850℃ 级别；看 CTI，也就是相比漏电起痕指数，带电件一般要 400V 以上，高压侧往 600V 走。
CTI 这个数一旦从 400V 掉到 250V，爬电距离的要求就翻倍——结构上原来留 2 mm 的地方要留到 4 mm，机箱宽度定死了，等于结构重做。所以它不是报告上多一行字，是外壳尺寸的约束。
其三，机械。DIN 导轨卡扣要承受设备自重加振动加反复拆装，按 100 次拆装不裂来验收。听上去宽裕，可现场检修一天动一回，三个月就用完了。
其四，介质。现场是粉尘、油雾、凝露三样一起来。清洁时还会上醇类清洗剂，塑料壳表面被擦花的投诉，多半出在这里。
其五，寿命。工控设备的账期一般按 8 到 10 年算，比消费电子长得多，长期老化后的绝缘和力学底线要靠 RTI 那类指标兜着。
把年限换成小时：8 年 × 每年 8000 小时，接近 6.4 万小时。UL 746B 的 RTI 是以千小时量级的老化外推出来的，你手上的件却要跑 6 万小时，中间那段没有捷径，只能靠余量。
其六，功能附加项。屏蔽和散热。塑料本身不导电，要屏蔽就得做导电处理，涂层式能做到 50–70 dB；散热和防护等级是一对矛盾，IP54 可以靠散热筋自然对流，要 IP65 就得把热量用导热塑料或金属导出去。
还有一项容易被漏掉的隐性变量：振动。柜内接触器、继电器一动作就是冲击，壳体与导轨之间长期微动，螺丝连接会松、卡扣配合面会磨。这一项不写进工况表，供应商没法替你预留配合余量。
这六维里，温度、电、机械、寿命这四样要拿到数字，再谈换料。原件连服役温度区间都说不清的，换料就是拿批量去赌。
二、三条材料路线，并列摆开
换料不是换成指标最高的那条，是把三条路线的代价摆开，看哪条和你的工艺兜得住。
| 路线 | 薄壁阻燃 | 绝缘与爬电 | 韧性与卡扣 | 适合换自哪里 |
|---|
| 阻燃 PC | 1.6 mm 才稳 | 好，但耐溶剂一般 | 常温好、低温掉 | 原 ABS 壳体、外观优先件 |
| 无卤阻燃 PA66 | 0.8 mm 可达 | 可做高 CTI，配玻纤更稳 | 需增韧，卡扣要单独配 | 原 PC 壳体、要薄壁和爬电 |
| 阻燃 PA6 + 玻纤增韧 | 1.0–1.6 mm | 中等 | 韧性余量宽、成本低 | 原 PA66 成本压力件 |
三条没有谁更好，只有哪条和你这个机箱的壁厚、爬电距离、卡扣结构匹配。
一个常见误判是"直接上 PA66 阻燃就对了"。PA66 的薄壁阻燃确实是它的长处，可它吸湿后的尺寸变化比 PC 明显，配合面在南方梅雨季会漂；不预留这个余量，装柜时就是另一类投诉。
另一个误判是"卡扣和壳体用一套料省事"。卡扣要韧性，壳体要刚性和阻燃，这两件事在一套配方里是互相让步的。BOM 多一个料号的管理成本，比售后批量换壳的成本低得多。
三、判据表：换料后你重验的是这几行
下表门限是方向性建议，不是验收标准。实际数值必须由你的件、你的工况和实测确定。
| 指标 | 方向性门限 | 验证方法 / 标准 | 换料后常见失效 | 通行解法 | 对应助剂体系 |
|---|
| 薄壁阻燃 | 0.8 mm 达 V-0 | UL94 / GB/T 2408 | 薄壁处续燃 | 改磷氮体系或加壁厚 | 阻燃协效 |
| 灼热丝 | GWIT 850℃ | IEC 60695-2-12 / GB/T 5169.12 | 端子附近起火 | 加玻纤与无机填料 | 阻燃协效 |
| 相比漏电起痕 | ≥400V（高压侧≥600V） | IEC 60112 / GB/T 4207 | 粉尘凝露后表面碳化 | 提 CTI 体系，避含卤 | 抗氧剂 |
| 长期使用温度 | 按 RTI 留余量 | UL 746B | 老化后脆化、绝缘下滑 | 提耐温档位 | 抗氧剂 |
| 低温冲击 | -20℃ 悬臂无脆断 | GB/T 1843 / ISO 180 | 卡扣冬季断 | 核壳结构增韧 | 增韧剂 |
| 卡扣拆装 | ≥100 次不裂 | 往复拆装台 | 卡扣根部裂 | 局部增韧、加大圆角 | 增韧剂 |
| 尺寸与吸水 | 配合面调湿后复测 | GB/T 1034 + 调湿后三坐标 | 装不进、异响 | 矿物填充降各向异性 | — |
怎么用这张表：先看阻燃和爬电这两行，这两行过不去，后面几行都不用谈，因为安规会先把项目拦住。再看低温冲击和卡扣，这两行决定售后。
也别一上来追 RTI 最高档。耐温档位往上提，往往要牺牲流动性和韧性，而柜内 85℃ 的位置通常只在电源侧的一小块区域——整机按最热那块配，是为用不上的余量买单。
四、换料后四种失效，和它们真正的原因
失效一：卡扣在冬天断，断口发白、位置在根部。最常见的误判是"韧性不够，给我加增韧"。
但增韧有类型之分。常温脆断，普通弹性体增韧就能解决；-20℃ 以下的低温脆断，普通弹性体自己先变脆，要上核壳结构才管用。不问一句"断在常温还是低温"，加进去的增韧剂可能只是白花钱。
失效二：安规测试过了，现场却爬电。实验室是干净环境，现场有粉尘和凝露，CTI 要按现场污染等级留余量，别按实验室数据贴线设计。这一条是我们最想提前讲的一类——它不在料的问题里，在"拿哪个环境的数去设计"里。
失效三：同一批外壳黄得深浅不一。这通常不是"料不稳定"，是抗氧剂分散不均，或者局部温度超过了助剂的耐温上限。看到黄变先查混料工艺和助剂耐温，别急着换基材。
这一条是助剂侧的归因：料没换错，是稳定化体系没跟这个件的工况配到位。
失效四：屏蔽效能用了一年往下掉。多数情况不是料衰减，是导电涂层的附着或衬垫压缩永久变形。先查涂层和衬垫，再考虑换料——顺序反了，成本翻倍。
五、加工与验证：先锁阻燃，再锁爬电
换料先动的是干燥和模温，这两样在 PA66 体系上最敏感。
PA66 含水率超过 0.15%，在料筒里就可能水解降解，出来是脆件加银纹。普通热风干燥机对尼龙基本无效，要用除湿干燥机。这一条我们每次都要说一遍，因为它成本偏低、也最容易被省掉。
模温是第二件。玻纤料表面发白、发毛，客户第一反应常常是"玻纤加多了"。把模温表一查，80℃。提到 115℃，同一批料、同一个模具，浮纤基本消失。
玻纤是被你加进去的，又被你冻在了表面上。这不是配方的事，是模温的事。
验证顺序建议这样排，顺序别换：
1. 材料级：含水率、阻燃与灼热丝报告、CTI（干态与调湿态各一组）
2. 工艺窗口：变模温与保压打对比件，看薄壁填充与浮纤
3. 件级：卡扣拆装 100 次、低温冲击、配合面调湿后复测
4. 安规：整机送检，阻燃加灼热丝加 RTI 一起报
5. EMC：开孔、缝隙、接地连续性先过一遍，再谈导电料
6. 整机：装柜做高低温循环，把现场那个冬天的温度补回来
前面不通过就往下走，后面测出来的数没有解释意义。为什么顺序不能换？因为 CTI 依赖含湿状态，含湿没锁住就去调模温，调出来的窗口只对那一模有效，批量一放又漂。
六、边界：这几种情况，工控外壳先别换
这一段可能比前面几段更值钱，因为它帮你在开工前止损。
其一，超大机箱、结构优先、成本压到极致的件。这类件钣金或压铸铝的单件成本往往低于改性尼龙，换料省下的余量抵不过批量价差，建议留在金属路线。
其二，要求结构性屏蔽、屏蔽效能指标又高的件。塑料走导电料或导电涂层是能做的，但一旦选定导电料，整批料价上浮，还牵连接地设计和电化学腐蚀，退路很少。这种情况更该考虑局部金属内衬，塑料壳只管结构和外观。
其三，电源仓那种长期稳定超过 105℃ 且要按长时间 RTI 考核的位置。PA6 和 PA66 在这个区间长期表现的余量不足，该看 PPS 或高温尼龙路线，普通阻燃换料填不上这个坑。
其四，只做几十台样机、认证周期又压不动的项目。样品顺、批量卡、整案回退的学费，比一开始不换高得多。把认证周期当独立科目排进项目表，比在料价上抠一点重要。
还有就是需要用回收料或副牌料顶住绝缘件的位置。介电强度批次波动大，耐压测试容易在抽样上打脸。绝缘件走正牌新料，批次附报告，这条没有例外。
七、换料风险清单（从原方案换到阻燃尼龙，要动的东西）
| 环节 | 要动什么 | 容易漏的点 |
|---|
| 模具 | 收缩率随玻纤含量变，卡扣与导轨配合面可能要修 | 只换料不修模，装柜时卡 |
| 分件 | 卡扣与壳体可能分成两套料 | 一套料管全机，卡扣先死 |
| 干燥 | 按实测含水率定窗口，除湿干燥机必备 | 热风干燥机对尼龙基本无效 |
| 模温 | 玻纤料模温往上提，浮纤才压得住 | 按通用建议值开 80℃，表面发白 |
| 料温 / 保压 | 薄壁件要看填充与飞边，联合调 | 只改料温不改保压，尺寸漂 |
| 色差 | 免喷涂深色件色板提前确认 | 不同批次基材底色有差 |
| 验证顺序 | 材料→工艺→件级→安规→EMC→整机 | 前一项未过就往下走 |
八、打样试模排程（几轮上机、每轮验什么、留样多久）
我们给工控外壳换料排的试模，通常三轮，轮次之间不跳步。
首轮·小样比对：用你的原模具打 3–5 模，只验含水率、外观、薄壁填充和卡扣能不能脱模。这轮不追性能，先把"料能不能填进去、件能不能下来"确认掉。留样两件，标批号与干燥参数，留到下一轮结束。
次轮·工艺窗口：固定料，变模温与保压打两组对比件。验薄壁阻燃的试样、调湿态 CTI、卡扣拆装 100 次与低温冲击。这轮定量产参数。留样按批次封存，留到量产稳定后三个月。
末轮·安规与整机：送检阻燃、灼热丝、RTI，同步装柜做 EMC 整改和高低温循环。这轮过了，才建议放量。留样封存覆盖首批量产，便于追因。
配方里的助剂体系按件的工况配——常规助剂常备现货，特殊型号按需配套；你报工况和牌号，料和助剂一次配齐。
读者常问的三句
问：换料后卡扣断，是料不行还是结构不行？先看断口和断的温度。常温断，查圆角与浇口；低温断，查增韧类型是不是核壳结构。两件事解法不同，别一上来换更高档的料。
问：阻燃等级一样的两个料，能互相替吗？不能只看等级。薄壁阻燃的测试厚度、灼热丝的门限、CTI 的数、RTI 的档，四项都是等级之外的东西。同一张 V-0 报告，背后可能不是同一条路。
问：塑料壳要不要做接地？要。塑料壳不等于免接地，内部带电部件的防护接地和 EMC 泄放路径都不能省，实现靠金属安装件和接地端子，结构设计时就要留位置。
给采购朋友的四个字，凑成一张可以向上汇报的表：
| 一张纸 | 结论 |
|---|
| 换什么 | 薄壁件看无卤阻燃 PA66，卡扣单独配增韧 |
| 动什么 | 干燥窗口、模温往上提、卡扣与壳体分件 |
| 验什么 | 薄壁阻燃、调湿态 CTI、低温冲击，三项一起测 |
| 什么时候能放量 | 三轮试模过、安规出报告、整机循环通过 |
这四行写清楚，评审会上就少有回头路。
这句话我们每周都听到。
"我换了更好的料，怎么反而出问题。"——问题常常不在料好不好，在换料时有没有把卡扣、爬电、散热这三处一起重排。回到开篇那三句问话：问换体系还是换基材，问卡扣有没有分件，问低温冲击做过没有。
Last month, a client working in PLCs sent me a video on their phone. Filmed at night, the lights in the computer room were off, and the side of a wall-mounted industrial control computer was flipped open by two millimeters, right where the DIN rail clip was raised.
He added, "So it's a flame-retardant PC, but this time I switched to a better flame-retardant material, but when I put it in the cabinet, the latch made a sound." "
The industrial computer shell was replaced. What he changed was the material, but the three things he didn't change were the toughness of the clips, the way the guides bear the load, and the temperature during assembly.
I asked him three questions: Is it the flame-retardant system or the base material? Are the clips and casing made of the same or two layers? Have you ever done low-temperature shock treatments?
He paused for two seconds, but hadn't thought about the last two sentences.
This answer isn't strange. At the material change site for industrial control casings, most people start by the flame-retardant rating section and end there.
The timeline below is the process of his part.
Starting point: The new material sample parts met the appearance standard, the halogen-free flame-retardant report received V-0, the project team breathed a sigh of relief, and the review meeting was approved.
Undercover: The goods were shipped to the northern client's data center. On winter nights, the temperature dropped below 5°C, and the machine was turned on and off three or four times a day, with the latch cycling hot and cold accordingly.
Outbreak: In the third week of installation, after-sales feedback says clips break. The proportion is low—one or two out of a dozen units—but the customer is the operator, so complaints are reported directly to the project team.
Settlement: 800 units reworked on shells, clip positions were reworked and replaced, and along with the re-operation came a flame-retardant certification.
Material replacements are rarely recorded in the material price column; most are noted in these "processes that no one reschedules."
1. The six dimensions before material replacement, four must first be counted as numbers
One, temperature, and its shape. Inside the industrial control cabinet, the temperature stays at 50–70°C for a long time, but on the side of the power module and inverter, localized temperatures can reach 85–95°C. That's not a high number, but it runs continuously, 24/7, and the machine runs for eight years, so this temperature will accompany you for eight years.
Wide temperature requirements usually state -20°C to 70°C; For rail transit or outdoor cabinets, it says -40°C to 75°C. Both ends must be used, and the room temperature value in the middle is actually the least important.
Second, electricity. This dimension is harder than mechanical loads on industrial control components. The enclosure requires UL94 V-0, with a standard thickness of 1.6 mm; Thin-walled parts near the terminals can be 0.8 mm or V-0.
Flame retardant is just the ticket to entry. Near live circuits, you also need to check the hot wire, usually at 850°C; Check CTI, which means compared to the leakage trace index, live parts generally need to be above 400V, and the high-voltage side should go toward 600V.
CTI Once this number drops from 400V to 250V, the creepage distance requirement doubles—the original 2mm space should be left at 4mm, and the chassis width is fixed, essentially a structural rework. So it's not just an extra line in the report, but a constraint of enclosure size.
Third, mechanical. DIN rail clips must withstand the equipment's own weight, vibration, and repeated disassembly and assembly, and be accepted as 100 times without cracking. It sounds generous, but you can only do one daily maintenance and use it up in three months.
Fourth, medium. On site, dust, oil mist, and condensation are all collected. Alcohol-based cleaners are also applied during cleaning, and complaints about scratches on plastic shell surfaces mostly come from this.
Fifth, lifespan. Industrial control equipment usually has a lifespan of 8 to 10 years, much longer than consumer electronics. After long-term aging, insulation and mechanical bottom lines rely on indicators like RTI.
Convert lifespan into hours: 8 years × 8,000 hours per year, close to 64,000 hours. UL 746B's RTI is designed for thousand-hour aging, but your part needs to run 60,000 hours. There's no shortcut in between—you can only rely on margin .
Sixth, additional functional items. Shielding and heat dissipation. Plastic itself is non-conductive; if you want to shield, you need conductive treatment. Coated types can achieve 50–70 dB; Heat dissipation and protection levels are contradictory: IP54 can be naturally convected by heat sinks, but IP65 requires heat to be dissipated using thermal plastic or metal.
There's another hidden variable that's easy to overlook: vibration. Contactors and relays inside the cabinet are immediately impacted, and the shell and guide rails move for long periods, causing screw connections to loosen and the latch surfaces to wear. If this item isn't written in the operating schedule, the supplier can't reserve a matching margin for you.
Of these six dimensions, temperature, electricity, mechanics, and lifespan need to be obtained before discussing material replacement. You can't even specify the service temperature range for the original parts; replacements are like gambling on mass production.
Two or three material routes, arranged side by side
Material replacement isn't about switching to the highest indicator; it's about laying out the costs of the three routes and seeing which one fits your process.
| Route | Thin-walled flame retardant | Insulation and creepage resistance | Toughness and snap-fit | Suitable replacement from |
|---|
| Flame retardant PC | 1.6 mm is more stable | Good, but solvent resistance is average | Good at room temperature, drops at low temperatures | original ABS housing, preferred appearance parts |
| halogen-free flame retardant PA66 | 0.8 mm can achieve | high CTI, more stable with fiberglass | requires toughening, buckle must be separately paired with | original PC housing, thin-walled and creepy-electric |
| flame-retardant PA6 + glass fiber toughened | 1.0–1.6 mm | Medium | Wide toughness margin, low cost | Original PA66 cost pressure part |
Of the three, none is better; the only choice is which one matches your case's wall thickness, creepage distance, and snap-fit structure.
A common misjudgment is "just use PA66 flame-retardant directly." PA66's thin-walled flame retardant is indeed its strength, but its dimensional changes after moisture absorption are more noticeable than PC, and the surface will float during the southern rainy season; If you don't reserve this margin, it becomes a different type of complaint during container loading.
Another misjudgment is "using the same material for the clip and shell, which is easier to handle." The clip must be tough, the shell needs rigidity and flame retardancy; these two things are mutually compromised within one formula. The management cost of adding one part number to the BOM is much lower than the cost of bulk after-sales shell replacement.
3. Specification Table: After re-retesting, you re-inspect these few lines
The threshold in the table below is directional advice, not acceptance standard. The actual values must be determined by your part, your working conditions, and actual measurements.
| Indicator | Directional Threshold | Verification Method / Standard | Common Failure After Material Replacement | Common Solution | Corresponding Additive System |
|---|
| Thin-Walled Flame Retardant | 0.8 mm Reach V-0 | UL94 / GB/T 2408 | Continued flame at thin-walled area | Modified phosphorus-nitrogen system or wall thickness added | Flame retardant synergistic |
| Scorching wire | GWIT 850° C | IEC 60695-2-12 / GB/T 5169.12 | Fire near terminal | With glass fiber and inorganic fillers | Flame Retardant Synergy |
| Compared to leakage-induced tracking | ≥400V (high voltage side ≥600V) | IEC 60112 / GB/T 4207 | Surface carbonization after dust condensation | Promote the CTI system, avoid halogens | Antioxidant |
| Long-term use temperature | Reserve according to RTI | UL 746B | Brittleness after aging, insulation deterioration | Temperature adjustment setting | Antioxidant |
| Low-temperature shock | -20℃ cantilever no brittle fracture | GB/T 1843 / ISO 180 | Buckle breaks in winter | Core-shell structure toughening | Toughening agent |
| Clip removal and installation | ≥100 times without cracking | Reciprocating assembly and disassembly platform | Crack at the base of the buckle | Local toughening and increased fillet radius | Toughening agent |
| Size and Water Absorption | Re-test after adjusting surface moisture | GB/T 1034 Three-coordinate measurement after humidity adjustment | Doesn't fit, abnormal noise | Mineral-filled reduced anisotropy | — |
How to use this table: First look at the rows for flame retardancy and creepage; if these two rows don't pass, there's no need to worry about the following rows, because safety regulations will block the project first. Then look at low-temperature impact and buckles; these two rows determine after-sales.
Don't rush to chase the highest RTI right away. Increasing the temperature rating often comes at the cost of flowability and toughness, and the position inside the cabinet that reaches 85℃ is usually only a small area on the power supply side—designing the whole machine according to the hottest spot means paying for extra margin that won't be used.
4. Four types of failures after material replacement, and their real causes
Failure 1: The buckle breaks in winter, with a white fracture at the base. The most common misjudgment is 'not tough enough, add toughness for me.'
But there are different types of toughening. For brittle fracture at room temperature, ordinary elastomer toughening can solve it; for brittle fracture below -20°C, ordinary elastomers themselves first become brittle, and only core-shell structures are effective. If you don't ask whether the fracture occurs at room temperature or low temperature, the added toughening agent may just be a waste of money.
Failure 2: Safety tests passed, but electricity creeps on-site. The laboratory is a clean environment with dust and condensation on site. CTI should be designed with margins based on on-site contamination levels, not based on lab data lines. This is the type we most want to discuss in advance—it's not about materials, but about "designing based on the values of which environment."
Failure Three: The same batch of casings have uneven degrees of yellowing. This is usually not due to 'unstable material,' but rather because the antioxidant is unevenly dispersed, or local temperatures exceed the thermal limit of the additives. When you see yellowing, first check the mixing process and the heat resistance of the additives, and don't rush to change the base material.
This one is the additive side's attribution: the material wasn't changed incorrectly; it's that the stabilization system wasn't matched to the operating conditions of this part.
Failure four: Shielding effectiveness declines after a year of use. In most cases, it is not material degradation, but the adhesion of the conductive coating or permanent deformation of the gasket compression. First check the coating and gasket, then consider replacing the material — if the order is reversed, the cost doubles.
5. Processing and verification: first lock flame retardancy, then lock tracking resistance
When changing materials, the first things to adjust are drying and mold temperature, as these two are the most sensitive in the PA66 system.
If the moisture content of PA66 exceeds 0.15%, it may hydrolyze and degrade in the barrel, resulting in brittle parts with silver streaks. A regular hot air dryer is basically ineffective for nylon; a dehumidifying dryer must be used. We have to mention this every time because the cost is relatively low and it is the easiest to be skipped.
Mold temperature is the second issue. The surface of the glass fiber material turns white and fuzzy, and the customer's first reaction is often "too much glass fiber." When checking the mold temperature, it is 80°C. When increased to 115°C, with the same batch of material and the same mold, the floating fibers basically disappear.
The fiberglass was added by you, and then it was frozen on the surface by you. This is not a matter of the formula; it’s a matter of the mold temperature.
It is recommended to arrange the verification in this order; do not change the sequence:
1. Material level: Moisture content, flammability and glow wire report, CTI (one set each for dry and conditioned states)
2. Process window: Compare mold temperature and holding pressure to see thin-wall filling and floating fibers
3. Component level: Snap-fit disassembly and assembly 100 times, low-temperature impact, retest after moisture adjustment of mating surfaces
4. Safety regulations: Submit the entire machine for inspection, report together with flame retardant, hot wire, and RTI
5. EMC: Go through openings, gaps, and grounding continuity first, then talk about conductive materials
6. Complete machine: Load into the cabinet for high and low temperature cycling, restoring the winter temperature on site.
If it doesn't pass at the front, just proceed to the next step; the numbers tested later have no explanatory significance. Why can't the order be changed? Because CTI depends on the moisture content. If the moisture content isn't locked, adjusting the mold temperature will only create a proper window for that specific mold, but when the batch is produced, it will drift again.
6. Boundaries: In these cases, don't replace the industrial control enclosure yet
This section may be more valuable than the previous few sections because it helps you stop losses before starting work.
First, components with extra-large cases, structure-priority designs, and costs pushed to the extreme. For these components, the unit cost of sheet metal or die-cast aluminum is often lower than modified nylon, and the savings from changing materials do not make up for the difference in batch pricing, so it is recommended to stick with the metal route.
Secondly, for parts that require structural shielding and have high shielding performance requirements. Using conductive plastic or conductive coating is feasible, but once a conductive material is chosen, the price of the entire batch rises, and it also involves the design of grounding and electrochemical corrosion, leaving little room for alternatives. In this situation, it is better to consider local metal lining, and let the plastic shell handle only the structure and appearance.
Thirdly, for positions like the power compartment that remain stably above 105℃ for a long time and require long-term RTI evaluation. PA6 and PA66 do not have enough long-term performance margin in this range; PPS or high-temperature nylon options should be considered, as ordinary flame-retardant material replacements cannot fill this gap.
Fourth, projects that only produce a few dozen prototypes and whose certification cycles cannot be shortened. The learning costs from smooth sample production, batch delays, and the need to roll back the entire plan are much higher than not upgrading at the start. Treating the certification cycle as an independent item in the project schedule is more important than saving a little on material costs.
Also, recycled material or secondary material needs to be used to hold the position of the insulators. The dielectric strength fluctuates greatly between batches, and the voltage-withstand test can easily fail during sampling. Insulators should use standard new material, with batch reports; there are no exceptions to this rule.
7. Material Change Risk List (Things that need to be changed when switching from the original plan to flame-retardant nylon)
| link; segment; part | What do you want to move? | Points that are easy to overlook |
|---|
| Mold | The shrinkage rate changes with the glass fiber content, and the mating surface between the buckle and the guide rail may need to be adjusted. | Only replace the material without repairing the mold, it gets stuck when loading the container |
| Parcel | The buckle and the housing may be made from two sets of materials | A set of material pipes for the whole machine, the clip is stuck first |
| Dry | Set the window according to the actual measured moisture content; a dehumidifying dryer is essential. | Hot air dryers are basically ineffective for nylon |
| Mold temperature | The mold temperature of the fiberglass material needs to be raised for the floating fibers to be pressed down. | Set to 80°C according to the general recommended value, the surface turns white |
| Material Temperature / Pressure Holding | For thin-walled parts, attention should be paid to filling and flash, and coordinate adjustments accordingly. | Only change the material temperature without changing the holding pressure, the dimensions drift. |
| Color difference | Pre-confirm the color swatch for dark-colored parts without spraying | There are differences in the base color of different batches of substrate |
| Verification order | Materials → Process → Component Level → Safety Standards → EMC → Whole Machine | If the previous item fails, just move on. |
8. Proofing and mold testing schedule (number of machine runs, what is checked in each run, how long samples are kept)
We replace the material for the industrial control enclosure trial mold, usually in three rounds, without skipping steps between rounds.
First round · Sample comparison: Use your original mold to make 3–5 samples, only testing moisture content, appearance, thin-wall filling, and whether the snap fit can be demolded. This round does not pursue performance; first confirm whether the material can be filled in and the parts can be taken out. Keep two samples, mark the batch number and drying parameters, and keep them until the end of the next round.
Second round · Process window: fix the material, vary mold temperature and holding pressure to produce two sets of comparison samples. Test thin-wall flame-retardant specimens, adjust moisture-conditioned CTI, perform snap-fit assembly/disassembly 100 times, and conduct low-temperature impact. This round determines the quantitative production parameters. Samples are sealed by batch and kept for three months after mass production stabilizes.
Final round · Safety regulations and complete machine: Send for inspection of flame retardance, glow wire, and RTI, while simultaneously performing EMC corrections and high-low temperature cycles when loading the cabinet. Only after passing this round is it recommended to increase the volume. Keep sample storage to cover the first batch of mass production, facilitating traceability.
The auxiliary system in the formula is matched according to the working conditions per item — conventional auxiliaries are kept in stock, and special models are matched as needed; you report the working conditions and grade, and the materials and auxiliaries are prepared together at once.
Three questions readers often ask
Q: After changing the material, the buckle breaks. Is it the material or the structure that's the problem? First, look at the fracture surface and the temperature at which it broke. If it breaks at room temperature, check the fillet and the gate; if it breaks at low temperature, check whether the toughening type is a core-shell structure. The solutions for these two cases are different, so don't immediately switch to a higher-grade material.
Q: Can two materials with the same flame-retardant rating be substituted for each other? You can't just look at the rating. The test thickness for thin-wall flame retardancy, the glow-wire ignition temperature threshold, the CTI number, and the RTI level are all factors beyond just the rating. Even if they have the same V-0 report, the underlying compliance path might not be the same.
Q: Do plastic cases need grounding? A: Yes. Plastic cases do not mean grounding is not required. Protective grounding for internal live parts and the EMC discharge path cannot be omitted. This is achieved through metal mounting parts and grounding terminals, so space must be reserved for them during structural design.
Four words for purchasing friends, combined into a table that can be reported upwards:
| A piece of paper | Conclusion |
|---|
| Change what | Thin-walled halogen-free flame-retardant PA66, snap-fit separately equipped with toughening |
| Move what | Dry the window, raise the mold temperature, separate the buckle from the housing |
| Test what | Thin-walled flame-retardant, moisture-regulating CTI, low-temperature impact, all three tested together |
| When can the volume increase? | Three-stage mold trial passed, safety certification report issued, complete machine cycle passed |
If these four lines are written clearly, there will be fewer revisions during the review meeting.
We hear this sentence every week.
"I switched to better materials, so why are problems still occurring?" — The problem often isn't about whether the material is good, but whether, when changing materials, the three areas of clips, creepage, and heat dissipation were all rearranged. Going back to the three initial questions: ask whether the system is being changed or just the base material, ask whether the clips have separate parts, and ask whether low-temperature impact tests have been done.
