高压连接器换料，胶芯是最不能省的一块。这篇讲清爬电距离为什么替代不了 CTI、薄壁与调湿各卡在哪一步、判据表怎么读，以及打样试模三轮各验什么、留样多久。
高压连接器胶芯换料这件事，上周从一个电话开始。
一家做新能源高压连接器的厂，胶芯要从原来那支进口料换到我们这边的路线。
他开口第一句是："我要一个 V0 的料，你们有几种？"
我追问了三句：平台是 400V 还是 800V？长期工作温度多少？胶芯有没有薄壁位置？
他答得很干脆：800V，长期 150℃，胶芯最薄的地方零点六毫米。
换完以后出的问题不在耐温上。件能过炉，本体也不软，卡住的是两件事：薄壁位置填充不匀、表面有浮纤；整机那边按污染等级校核，爬电距离不够。
他电话里的第二句话是："我加了距离，怎么还是不行？"
这句话点出了高压件换料最容易踩的一个坑——爬电距离补的是路径长度，补不了材料的碳化倾向。
下面这条时间线，是他们那批胶芯的真实经过。
起点是干态 CTI 报告合格、外观也过得去，项目组认定换料成功；潜伏阶段是薄壁位置的浮纤一直存在，被当成外观小问题；爆发是湿热处理后 CTI 掉了一档，整机爬电距离校核跟着不通过，安规送检停住；结算是回查，基材换了、模温和料温没跟着调、调湿态数据一直没测。
高压件换料的账，最后往往记在两个词上：测试状态，和成型窗口。
一、换料前把工况六维落下数字
高压连接器的胶芯，工况比外壳严得多，六样要一项项过。
电压这条是第一位。400V 平台与 800V 平台的余量要求差一大截，1000V 的方向又不同；CTI 的判定还要绑定污染等级，同一支料在不同污染等级下的合规定位不一样。
温度这条要分两本账。长期工作温度常见 125–150℃；另一本是温度循环，从零下几十度到 125℃ 走几百到上千次，考的是嵌件与塑件界面的配合。
介质这条是高压件特有的。冷却液、湿气、盐雾、清洗剂几样一起上；湿气最要命——尼龙吸湿之后体积电阻率会往下走，同一个牌号干燥时合格，湿热环境里可能掉一档。
寿命这条按整车十年到十五年算。高压连接器的插拔次数比低压件少，但温度循环次数多，卡扣和嵌件要按循环后的状态看。
外观这条在胶芯上主要看薄壁。零点六毫米的壁，比一张银行卡厚一点点，浮纤、缺料、流痕都直接影响绝缘与装配。
合规这条最长：CTI 按 IEC 60112、阻燃按最小壁厚报 V-0、灼热丝、黄卡厚度，出口还要加无卤清单。
| 维度 | 换料时要问的数 | 漏了会怎样 |
|---|
| 电压 | 平台电压、污染等级 | CTI 档位选错 |
| 温度 | 长期温度、温循次数与范围 | 嵌件界面开裂 |
| 介质 | 冷却液、盐雾、湿度 | 湿态绝缘下滑 |
| 寿命 | 年限、插拔次数 | 卡扣强度不足 |
| 外观 | 最小壁厚、浮纤限值 | 薄壁绝缘失效 |
| 合规 | 壁厚、CTI 档、无卤清单 | 安规卡住重来 |
六样里，平台电压和最小壁厚这两项要先定，它们直接决定路线。
二、三条基材路线，并列摆开
换料不是往最耐温那一档冲，是把三条路的代价摆清楚。
| 路线 | 可覆盖平台 | 吸水与尺寸 | 加工窗口 | 适合换自哪里 |
|---|
| PA66-GF 阻燃 | 400V 通用件 | 吸水偏高，尺寸一般 | 宽、好做 | 原通用阻燃料 |
| PA6T-GF 无卤 | 800V 主流件 | 吸水低，尺寸较稳 | 需高料温与高模温 | 原进口料交期或成本受限 |
| PA9T-GF | 800V 高要求薄壁件 | 吸水更低，尺寸稳 | 窗口最窄、料温最高 | 原进口料停产或缺货 |
三条没有谁更好，只有哪条跟你的壁厚和结构兜得住。
一个常见误判是看到 CTI 不达标就直接往上跳一档。跳之前先做三件事：确认湿态 CTI 真的不达标，很多项目是被干态数据吓自己；确认结构侧的爬电距离已经优化到位；确认温升和装配温度在新料的舒适区。
三件事做完再跳，钱花得踏实。
另一个误判是只比吸水率，不看加工窗口。窗口最窄的那一档，模具和产线要跟着改；改不动的部分，就是换料后新的风险点。
三、换料判据表：这张表决定你复验哪几项
把前面的约束落成能核对的指标。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、你的结构和实测确定。
| 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 |
|---|
| 漏电起痕指数 CTI | 800V 平台件按 600V 档，湿热后测 | IEC 60112，调湿态 | 掉一档、安规卡 | 低吸湿基材 + 不易碳化 | 抗氧剂（抑制降解） |
| 爬电距离校核 | 按 CTI 档位反推设计值 | 结构图纸 + IEC 60664-1 校核 | 距离加长仍不通过 | 先对档位，再改结构 | —（属结构侧） |
| 薄壁填充 | 最小壁厚处无缺料与浮纤 | 短射取样 + 目视与切片 | 浮纤、缺料、绝缘薄弱 | 提料温模温、改浇口 | 润滑剂（改善流动） |
| 调湿后尺寸 | 槽口与嵌件位按调湿态定 | 调湿处理 + 三坐标复测 | 装配渗漏、插拔力漂 | 按调湿态出图与验收 | 偶联剂（尺寸稳定） |
| 嵌件耐温循 | 温循后界面不开裂 | 温度循环试验 + 切片 | 嵌件处开裂 | 咬合结构或近膨胀系数 | —（属结构侧） |
| 阻燃等级 | 按最小壁厚报 V-0 | UL94 / IEC 60695-11-10 | 薄壁不达标 | 换阻燃体系并复测 | — |
怎么读这张表：先看头两行。
CTI 和爬电距离是一对，但不是互相替代的关系。CTI 管的是表面会不会被碳化，爬电距离管的是表面路径够不够长。材料本身容易碳化，距离再长也只是把失效点挪个位置。
第三列是给采购的：CTI 报告要写明测试状态和对应厚度，厚度以下才有效，壁厚更薄要降档。
最后两行的位置，其实是提醒：有些项换料解决不了，得回到结构去改。
四、换料后四种失效，和它们真正的原因
失效一：干态 CTI 合格，湿热处理后掉一档。
根因是三件事叠在一起：吸潮、表面污染、电压下的碳化。过程大致是表面吸潮或沾上导电污渍，泄漏电流把表面碳化，碳化层导电，电流进一步集中，最后形成不可逆的通道。所以升高 CTI 只有两条路——让材料不吸潮，让材料不易碳化，而这两条本身又互相牵制。
失效二：CTI 不够就用爬电距离去补。
这是这一行最常见的一个做法，也是本篇最想改掉的一个惯性。爬电距离在设计中确实有用，但它替代不了材料：某处一旦被碳化，导电通道会从碳化区往外扩展，而不会沿你设计的那条路径走。正确顺序是先选对 CTI 档位，再用距离做设计余量；反了，就是拿结构去补材料的短板。
失效三：薄壁位置浮纤、缺料。
胶芯最薄处零点六毫米，料流到这里已经冷了一截。根因通常在料温、模温和冲填速度，不在基材本身；加了阻燃体系之后流动性又降一档，填充难度更大。这一项靠工艺窗口解决，换料号解决不了。
失效四：温循之后嵌件位置开裂。
金属和塑料的线膨胀系数差一大截，温度循环下界面处会产生应力。做温循试验时如果件在嵌件位置开裂，多半是这个原因，不是材料强度不够——处理办法是让嵌件有足够的咬合结构，或者选膨胀系数更接近的体系，而且这件事要在设计早期解决，改材料改不动它。
补一条容易漏的：密封槽的槽口尺寸。塑料件吸湿后会涨，槽口跟着变，密封圈的压缩率就跟着变。关键尺寸一定要按调湿后的数据来定，否则装配时是好的，用一段时间才开始渗。这一条属于调湿态尺寸管理的账，不是材料的账。
五、加工与验证：料温、模温和调湿是三件事
料温在高压件上有一层额外风险：熔体温度过高，材料局部降解，表面更容易碳化。同一个牌号，工艺没调好的件和调好的件，实测 CTI 可能不在一个档。
模温的影响走两条路。模温低，料流前端愈合差，熔接线弱；同时表面粗糙，更容易吸潮积污，CTI 跟着往下走。
干燥这件事在胶芯上被放大。这两类料吸水快，拆包后敞口放几个小时含水率就回升；上机前要用水分仪或露点确认，不凭手感。
调湿则是很多人漏掉的第三件。槽口、嵌件位、卡扣这些关键尺寸，都要按调湿态出图和验收；按干态尺寸做出来的件，装配时是好的，跑一阵才开始渗。
验证顺序建议这样排，不要换：
1. 材料级：调湿态 CTI、灼热丝、薄壁阻燃
2. 工艺窗口：变料温与模温，打对比件看薄壁填充
3. 件级：调湿后尺寸、嵌件界面、端子插拔力
4. 温循：按实际循环范围走完，切片看界面
5. 整机：按污染等级校核爬电距离与绝缘
为什么顺序不能换？因为 CTI 与尺寸都依赖吸湿状态。吸湿没锁住就去调模温，出来的数据只对那一批有效。
六、边界：这几种高压件，换料先收手
这一段可能比前面几段更值钱，因为它帮你在开工前止损。
其一，400V 平台的通用件，且湿态 CTI 只是"看起来不够"。先确认是不是被干态数据吓到了，再把爬电距离优化到位，很多项目到这里就不用换料了。
其二，最低装配温度很低的产线。半芳香族的韧性比脂肪族低一档，常温装配没问题，低温场景下卡扣脆断会冒出来。升级方案定稿前，把最低装配温度下的插拔试验加进清单。
其三，周边有强电弧的位置。这类位置要看耐电弧数据，普通阻燃体系顶不住，该往专门方案看。
其四，成本压到极致的大批量件。材料升一档的钱，未必能在整机上收回来，这笔账要一起算。
其五，失效点还没定位的件。绝缘不达标、件裂了，先分清是 CTI、是爬电、还是嵌件界面，三件事的解法完全不同。
把这五条写前面不是劝退，是省时间。
七、换料风险清单（从原路线换到这边，要动的东西）
| 环节 | 要动什么 | 容易漏的点 |
|---|
| 模具 | 基材变了收缩率会变，槽口与嵌件位要复测 | 只换料不修模，密封先渗 |
| 干燥 | 换除湿干燥机，按实测含水率定窗口 | 热风干燥对这两类料基本无效 |
| 调湿 | 关键尺寸按调湿态出图与验收 | 按干态尺寸放行 |
| 料温 / 模温 | 料温过高会降解，表面更易碳化 | 照抄上一支料的档位 |
| 保压 / 脱模 | 薄壁与嵌件位要重定保压 | 嵌件处内应力集中 |
| 色差 | 本色与深色件分别对色板 | 批次间底色有差 |
| 验证顺序 | 材料→工艺→件级→温循→整机 | 前一项没过就往下走 |
八、打样试模排程（几轮上机、每轮验什么、留样多久）
我们给胶芯换料排的试模，通常分三轮，轮次之间不跳步。
头一轮·小样比对：用你的原模具打 3–5 模，验薄壁填充、浮纤、短射熔接线位置，同时确认干燥后含水率。这一轮先把"料能不能填满零点几毫米的薄壁"确认掉。留样两件，标注批号、颗粒干燥参数与模温。
第二轮·工艺窗口：固定料，变料温与模温，打两组对比件。送调湿态 CTI、薄壁阻燃；验调湿后槽口与嵌件位尺寸、端子插拔力。这一轮决定量产参数。留样按批封存，至少留到量产稳定后三个月。
第三轮·温循与整机：按实际循环范围做温度循环试验，切片看嵌件界面；再把件装到整机上，按污染等级校核爬电距离与绝缘。这一轮过了，才建议放量。留样封存周期覆盖首批量产，便于追因。
三轮之间为什么不能跳？因为每一轮的结论都是下一轮的前提：填充不确认，工艺窗口无从谈起；吸湿状态不定，CTI 与尺寸数据都没有解释意义。
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：CTI 不够，先加爬电距离还是先换料？先把距离按档位校核一遍，再看湿态 CTI 数据。顺序反了，钱就白花。
问：同一个牌号，为什么两家测出来的 CTI 不一样？多半差在测试状态和厚度上。把状态与厚度写进规格，比争谁的数据更可信有用。
问：调湿这道工序能不能省？关键配合尺寸不能省。按干态尺寸放行的件，渗漏往往出现在装配之后很长一段时间。
九、爬电距离与 CTI 的分工表（送给结构同事）
这张表建议直接抄进设计评审。
| 问题 | 归谁管 | 材料侧能做什么 | 结构侧能做什么 |
|---|
| 表面会不会被碳化 | 材料 | 选低吸湿、不易碳化的体系 | 无 |
| 表面路径够不够长 | 结构 | 无 | 按档位校核爬电距离 |
| 长期会不会掉档 | 材料与工艺 | 稳定化体系、控料温 | 避免积污死角 |
| 薄壁绝缘够不够 | 材料与工艺 | 提流动、改体系 | 局部加厚、改浇口 |
爬电距离解决的是路径，CTI 解决的是"表面会不会被碳化"。两个都要，不能拿一个顶另一个。
如果你要拿这篇去汇报，可以收成四行：
| 项 | 一句话结论 |
|---|
| 换什么 | 800V 胶芯往低吸湿的高温体系看，先确认湿态数据 |
| 动什么 | 料温模温重设、干燥换除湿、调湿后重新出图 |
| 验什么 | 调湿态 CTI、薄壁填充、调湿后尺寸、温循界面 |
| 什么时候能放量 | 三轮试模过、温循不开裂、整机爬电校核通过 |
前两天接了个电话。
电话里那句"我加了距离，怎么还是不行"，其实问的不是距离，是材料在湿热环境下会不会被碳化。
也回到开篇那三句问话：问平台电压、问长期温度、问最小壁厚——这三样答全了，胶芯往哪条路线走基本就定了。胶芯这块件换料，最贵的从来不是那袋料，是把测试状态、成型窗口、调湿尺寸这三件事一起摆上桌面。
When replacing materials in high-voltage connectors, the rubber core is the part that must not be skimped on. This article explains why the creepage distance cannot be replaced by CTI, which steps thin walls and moisture adjustment are critical at, how to read the criteria table, and what is checked in each of the three rounds of prototype molding tests, as well as how long samples are kept.
The matter of replacing the adhesive core in the high-voltage connector started last week from a phone call.
A factory that produces high-voltage connectors for new energy wants to switch the rubber core from the previously imported material to our supply route.
The first thing he said was: 'I want a V0 material, how many types do you have?'
I asked three follow-up questions: Is the platform 400V or 800V? What is the long-term operating temperature? Does the adhesive core have thin-walled areas?
He answered very straightforwardly: 800V, long-term 150°C, the thinnest part of the cable core is 0.6 millimeters.
After the replacement, the problems are no longer about temperature resistance. The parts can go through the furnace, the body is not soft either, and the issues are twofold: uneven filling in thin-walled areas and floating fibers on the surface; on the whole machine side, when checked according to the pollution level, the creepage distance is insufficient.
His second sentence on the phone was: 'I added distance, but it still doesn't work.'
This sentence points out one of the most common pitfalls when changing materials for high-voltage parts—the creepage distance compensates for the path length, but it cannot compensate for the material's tendency to carbonize.
The timeline below is the true course of that batch of glue cores.
The starting point was that the dry-state CTI report was qualified and the appearance was acceptable, so the project team considered the material change a success; the latent phase was that the floating fibers at the thin-walled areas persisted and were regarded as minor appearance issues; the outbreak occurred after the damp-heat treatment, when the CTI dropped by one level, causing the overall creepage distance verification to fail, and safety inspection submissions were halted; the settlement was a retrospective review, which found that the substrate had been changed, but the mold temperature and material temperature were not adjusted accordingly, and the damp-state data had never been measured.
The cost of changing high-pressure parts is often ultimately recorded under two terms: test status and molding window.
1. Record the six-dimensional process data before changing the material
The gel core of the high-voltage connector operates under much harsher conditions than the casing, so each of the six items must be checked one by one.
Voltage comes first. The margin requirements for the 400V platform and the 800V platform differ greatly, and the direction for 1000V is also different; CTI determination also needs to be tied to the pollution level, and the compliance positioning of the same material varies under different pollution levels.
The temperature aspect needs to be accounted for in two separate ways. The long-term operating temperature is commonly 125–150°C; the other is temperature cycling, ranging from several tens of degrees below zero to 125°C for several hundred to over a thousand cycles, which tests the fit between the inserts and the plastic parts.
The dielectric is specific to high-voltage components. Coolant, moisture, salt spray, and cleaning agents all affect it together; moisture is the most critical—after nylon absorbs moisture, its volume resistivity will decrease. The same grade may pass when dry but could drop a level in a hot and humid environment.
The lifespan is calculated as ten to fifteen years for the whole vehicle. The number of plug-in and pull-out cycles for high-voltage connectors is less than that of low-voltage components, but the number of temperature cycles is higher, and the latches and inserts should be checked according to their condition after the cycles.
Regarding appearance, this mainly concerns the thin walls on the glue core. A wall thickness of 0.6 millimeters, slightly thicker than a bank card, means that floating fibers, material deficiencies, and flow marks all directly affect insulation and assembly.
Compliance takes the longest: CTI according to IEC 60112, flame retardant rated V-0 based on minimum wall thickness, glow wire, yellow card thickness, and exports also require a halogen-free list.
| Dimension | The numbers to ask when changing materials | What happens if it leaks? |
|---|
| Voltage | Platform voltage, pollution degree | CTI gear selected incorrectly |
| Temperature | Long-term temperature, number of temperature cycles, and range | Insert interface cracking |
| Medium | Coolant, salt spray, humidity | Insulation degradation under wet conditions |
| Lifespan | Years of use, number of insertions and removals | Insufficient buckle strength |
| Appearance | Minimum wall thickness, floating fiber limit | Thin-wall insulation failure |
| Compliance | Wall thickness, CTI rating, Halogen-free list | The safety regulation card got stuck, redo |
Among the six items, the platform voltage and the minimum wall thickness must be determined first, as they directly determine the route.
Two or three substrate routes, laid out side by side
Changing materials is not about pushing to the most heat-resistant setting, it's about clearly laying out the costs of the three options.
| Route | Platforms that can be covered | Water Absorption and Size | Processing window | Where is it suitable to change from? |
|---|
| PA66-GF Flame Retardant | 400V Standard Component | High water absorption, average size | Wide, easy to make | Original general-purpose flame retardant |
| PA6T-GF Halogen-free | 800V mainstream components | Low water absorption, relatively stable dimensions | Requires high material temperature and high mold temperature | Original imported material delivery time or cost is limited |
| PA9T-GF | 800V high-demand thin-walled parts | Lower water absorption, stable dimensions | Narrowest window, highest material temperature | Original imported materials are discontinued or out of stock |
None of the three is better; it only depends on which one can handle your wall thickness and structure.
A common misjudgment is to immediately jump to a higher level when the CTI is found to be substandard. Before jumping, do three things: confirm that the wet CTI is indeed substandard, as many projects are misled by the dry-state data; confirm that the creepage distance on the structural side has been properly optimized; confirm that the temperature rise and assembly temperature are within the comfort zone for the new material.
Finish three things before jumping, and you'll spend money with peace of mind.
Another misjudgment is comparing only the water absorption rate without considering the processing window. For the grade with the narrowest window, the mold and production line need to be adjusted accordingly; the parts that cannot be adjusted become new risk points after changing the material.
3. Material Change Evaluation Form: This form determines which items you will re-inspect
Translate the previous constraints into verifiable metrics. The thresholds in the table below are directional suggestions, not acceptance criteria; the actual values must be determined by your parts, your structure, and actual measurements.
| Indicator | Directional Threshold | Verification Method / Standard | Common failures after material change | Common solution | Corresponding auxiliary agent system |
|---|
| Tracking Index CTI | 800V platform component tested at 600V setting after damp heat | IEC 60112, conditioned state | Drop one gear, safety regulation card | Low moisture-absorbing substrate, not easy to carbonize | Antioxidant (inhibits degradation) |
| Creepage distance verification | Derive the design value in reverse according to the CTI level | Structural Drawing IEC 60664-1 Check | The extended distance still does not pass | First adjust the gear, then change the structure | — (possessive structure side) |
| Thin-wall filling | No missing material or floating fibers at the minimum wall thickness | Short-range sampling Visual inspection and slicing | Floating fibers, material shortage, weak insulation | Material heating temperature and mold temperature, modify the gate | Lubricant (improves flow) |
| Dimensions after moisture conditioning | The groove and insert positions are determined according to the conditioned state | Moisture conditioning Re-measurement with CMM | Assembly leakage, plug-in force drift | Drawings and acceptance based on moisture conditioning | Coupling agent (size-stable) |
| Insert Temperature Cycling | The interface does not crack after thermal cycling | Temperature cycling test Slice | Cracking at the insert | Occlusion structure or near expansion coefficient | — (Belongs to the structural side) |
| Flame retardant rating | Report V-0 according to minimum wall thickness | UL94 / IEC 60695-11-10 | Thin wall does not meet the standard | Change the flame retardant system and retest | — |
How to read this table: first look at the first two rows.
CTI and creepage distance are a pair, but they are not interchangeable. CTI controls whether the surface will carbonize, while creepage distance controls whether the surface path is long enough. If the material itself easily carbonizes, no matter how long the distance is, it only shifts the point of failure.
The third column is for procurement: the CTI report must specify the test status and corresponding thickness; it is only valid below the thickness, and if the wall is thinner, it must be downgraded.
The position of the last two lines is actually a reminder: some items cannot be solved by changing materials and need to be modified at the structural level.
4. Four types of failures after material replacement, and their real causes
Failure 1: Passes dry CTI, drops one level after damp heat treatment.
The root cause is a combination of three factors: moisture absorption, surface contamination, and carbonization under voltage. The process roughly goes like this: the surface absorbs moisture or gets coated with conductive dirt, leakage current carbonizes the surface, the carbonized layer conducts electricity, the current becomes further concentrated, and eventually an irreversible path is formed. So there are only two ways to increase CTI—either make the material resistant to moisture absorption or make it resistant to carbonization, but these two aspects inherently counteract each other.
Failure 2: If the CTI is insufficient, use the creepage distance to make up for it.
This is the most common practice in this line, and also an inertia that this article most wants to change. Creepage distance is indeed useful in design, but it cannot replace the material: once a spot is carbonized, the conductive path will expand outward from the carbonized area, rather than following the path you designed. The correct sequence is to first choose the right CTI level, and then use distance to provide design tolerance; if reversed, it means using the structure to make up for the shortcomings of the material.
Failure 3: Floating fibers and material deficiency in thin-walled areas.
The thinnest part of the glue core is 0.6 millimeters, and by the time the material reaches here, it has already cooled considerably. The root cause usually lies in the material temperature, mold temperature, and injection speed, not in the substrate itself; after adding the flame retardant system, the fluidity drops another level, making filling even more difficult. This issue can be resolved through the process window, not by changing the material grade.
Failure 4: Cracking at the insert position after thermal cycling.
The coefficients of thermal expansion for metal and plastic differ significantly, so stress will occur at the interface under temperature cycling. If a part cracks at the insert location during a thermal cycling test, it is most likely due to this reason, not because the material strength is insufficient. The solution is to provide the insert with sufficient interlocking structure, or to choose a system with a thermal expansion coefficient that is closer, and this issue needs to be addressed early in the design stage; changing the material later won’t fix it.
Add one that is easily overlooked: the groove opening size of the sealing groove. Plastic parts swell after absorbing moisture, causing the groove opening to change, and consequently, the compression rate of the sealing ring changes. The critical dimensions must be determined according to the data after humidity adjustment; otherwise, the assembly may be fine initially, but after a period of use, leakage may start. This point belongs to the management account of dimensions in a humidity-adjusted state, not to the material account.
5. Processing and Verification: Material temperature, mold temperature, and humidity control are three separate matters
There is an additional risk of melt temperature on high-pressure parts: if the melt temperature is too high, the material may locally degrade, and the surface is more prone to carbonization. For the same grade, parts with improperly adjusted process and properly adjusted process may have CTI measurements that are not in the same range.
The impact of mold temperature goes in two ways. When the mold temperature is low, the front end of the material flow heals poorly, and the weld lines are weak; at the same time, the surface is rougher, making it easier to absorb moisture and accumulate dirt, causing the CTI to decrease.
The issue of drying is amplified with the glue core. Both types of material absorb water quickly; after unsealing, leaving them open for a few hours will cause the moisture content to rise again. Before running the machine, it is necessary to confirm with a moisture meter or a dew point meter, not by feel.
Humidity adjustment is the third thing that many people overlook. Key dimensions such as slots, embedding positions, and clips must be drawn and inspected according to the humidity-adjusted state; parts made to dry-state dimensions may fit well during assembly, but leakage will start after a while.
It is recommended to arrange the verification sequence like this, do not change it:
1. Material grade: moisture-conditioned CTI, glow wire, thin-wall flame retardant
2. Process window: change material temperature and mold temperature, make comparison parts to observe thin-wall filling
3. Item level: dimensions after moisture adjustment, insert interface, terminal insertion and extraction force
4. Gentle circulation: Complete the actual circulation range, and examine the interface with slices
5. Complete machine: Check creepage distance and insulation according to pollution level
Why can't the sequence be changed? Because both CTI and size depend on moisture absorption status. If moisture absorption isn't locked, adjust mold temperature, and the data only applies to that batch.
6. Boundary: For these high-voltage components, stop handling them before changing materials
This section may be more valuable than the previous few sections because it helps you cut losses before starting work.
First, the 400V platform's general-purpose components, and the CTI in wet conditions just 'looks insufficient.' First, confirm whether it was intimidated by the dry-state data, and then optimize the creepage distance. Many projects don't need to change materials at this point.
Secondly, production lines with very low minimum assembly temperatures. The toughness of semi-aromatic types is one level lower than that of aliphatic types. Assembly at room temperature is fine, but brittle snapping of the clips can appear in low-temperature scenarios. Before finalizing the upgrade plan, include plug-in and pull-out tests at the minimum assembly temperature in the checklist.
Third, locations with strong surrounding electric arcs. For these locations, the arc resistance data should be checked; ordinary flame-retardant systems can't withstand it, and specialized solutions should be considered.
Fourth, large-volume parts with costs pushed to the extreme. The extra money spent on upgrading materials may not necessarily be recovered on the whole machine, and this calculation must be considered together.
Fifth, parts whose failure points have not yet been identified. If the insulation does not meet the standard or the part is cracked, first determine whether it is a CTI issue, a tracking issue, or an insert interface issue, as the solutions for these three problems are completely different.
Putting these five points at the beginning is not to discourage, but to save time.
7. Material Change Risk List (From the original route to this one, things that need to be moved)
| link; segment; part | What do you want to move? | Points that are easy to overlook |
|---|
| Mold | If the substrate changes, the shrinkage rate will change, and the groove and insert positions need to be re-measured. | Only change the material without repairing the mold; sealing will leak first |
| Dry | Replace the dehumidifying dryer and set the window according to the measured moisture content. | Hot air drying is basically ineffective for these two types of materials |
| Humidity control | Key dimensions are drawn and accepted according to the moisture-conditioned state | Release according to dry-state dimensions |
| Material Temperature / Mold Temperature | If the material temperature is too high, it will degrade and the surface will carbonize more easily. | Copy the gear setting from the previous batch |
| Pressure Holding / Demolding | The thin walls and insert locations need to have holding pressure reset | Stress concentration at the insert |
| Color difference | Color matching the natural color and the dark parts separately with the color board | There are differences in the base color between batches |
| Verification order | Material → Process → Component Level → Thermal Cycle → Complete Machine | If the previous item fails, just move on. |
8. Proofing and mold testing schedule (number of machine runs, what is inspected in each run, how long samples are kept)
When we change the material for the glue core and arrange the test mold, it is usually divided into three rounds, and steps are not skipped between rounds.
First round · Sample comparison: Use your original mold to produce 3–5 pieces, check thin-wall filling, floating fibers, short shot weld line positions, and also confirm the moisture content after drying. In this round, first verify whether the material can fill a thin wall of a few tenths of a millimeter. Keep two samples, and mark the batch number, granule drying parameters, and mold temperature.
Second Round · Process Window: Fix the material, change the material temperature and mold temperature, and make two sets of comparison parts. Send for humidity-conditioned CTI and thin-wall flame retardant; after conditioning, check the slot and insert dimensions, and the terminal insertion and extraction force. This round determines the mass production parameters. Samples should be sealed and stored by batch, and kept for at least three months after mass production stabilizes.
Third round: Temperature cycling for thermal circulation and the complete machine: perform temperature cycling tests according to the actual circulation range and inspect the insert interface on cross-sections; then install the parts on the complete machine and check creepage distance and insulation according to contamination level. Only after passing this round is mass production recommended. Samples should be preserved for a period covering the first batch of mass production to facilitate root cause tracing.
Why can't you skip between the three rounds? Because the conclusion of each round is the premise for the next: if the filling is not confirmed, there is no way to discuss the process window; if the moisture absorption state is uncertain, CTI and dimensional data have no explanatory significance.
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: If CTI is insufficient, should we increase the creepage distance first or change the material first? First, check the distance according to the rating, and then look at the wet-state CTI data. If the order is wrong, the money will be wasted.
Q: For the same grade, why do two companies get different CTI results? The difference is mostly due to test conditions and thickness. Writing the conditions and thickness into the specifications is more useful than arguing about whose data is more reliable.
Question: Can the humidity adjustment process be skipped? The critical fit dimensions cannot be skipped. Parts released according to dry-state dimensions often experience leaks long after assembly.
9. Creepage Distance and CTI Division Table (for structural colleagues)
It is recommended to directly copy this table into the design review.
| Question | Who is in charge of this? | What can the materials side do? | What can the structural side do? |
|---|
| Will the surface become carbonized? | Material | Choose a system that is low in moisture absorption and not prone to carbonization | None |
| Is the surface path long enough? | Structure | None | Check creepage distance according to the gear position |
| Will it slip gears over the long term? | Materials and Processes | Stabilization system, material temperature control | Avoid areas where dirt accumulates |
| Is the thin-wall insulation enough? | Materials and Processes | Increase liquidity, reform the system | Local thickening, modify the gate |
Creepage distance addresses the path, while CTI addresses whether the surface will carbonize. Both are needed; one cannot replace the other.
If you are going to use this article for a report, you can summarize it in four lines:
| item | A one-sentence conclusion |
|---|
| Change what | For the 800V gel core, from the perspective of low moisture absorption in high-temperature systems, first confirm the wet-state data |
| Move what | Material temperature and mold temperature reset, replace drying with dehumidification, redraw after adjusting humidity |
| Test what | Moisture-conditioned CTI, thin-wall filling, dimensions after moisture conditioning, thermal cycling interface |
| When can the volume increase? | Three-round mold trial passed, no cracking during thermal cycling, full machine creepage verification passed |
I received a phone call a couple of days ago.
That sentence on the phone, 'I increased the distance, but it still doesn't work,' was actually not about the distance; it was asking whether the material would carbonize in a hot and humid environment.
Also, back to the three opening questions: asking about the platform voltage, asking about the long-term temperature, asking about the minimum wall thickness—if you can answer all three, the route for the gel core is basically determined. When changing materials for the gel core, the most expensive part is never that bag of material; it's putting on the table the three things of testing state, molding window, and moisture-adjusted dimensions.
