SMT连接器换料，件和模具都在手上，最容易被忽略的不是耐温，是取向和模温。这篇讲清换料后共面度为什么会漂、判据表怎么读、试模三轮各验什么，以及骨架件换料要重排的那几道工序。
SMT连接器换料这件事，上个月在一家做板对板连接器的厂里撞了个正着。
他们原来用一档进口高温料，交期太长，想换到我们这边的路线。
骨架件打出来，外观挑不出毛病，短射也没有缺口，车间当天就报了合格。
结果端子压进去，整排共面度超了，贴板以后有几颗虚焊。
他电话里的原话是："料看着一样，怎么过炉就歪了？"
我先反问了三句：换的只是基材，还是整套配方一起动？模温是不是照原来的档位设的？取向分布有没有重新看过？
他停了两秒，说模温调过，但没按新料重设。
这句话点出了SMT连接器换料最容易漏的一环——件的外观和耐温往往都能过，先出问题的是取向，以及取向带来的翘曲。
下面这条时间线，是他那个骨架件的真实经过。
起点是打样件外观合格、装配尺寸也过线，车间松了口气；潜伏阶段是量产件过炉后共面度慢慢漂，单看一件不觉得；爆发是整板贴装后端子歪斜，客户端开始退板；结算是回查，配方没大改，改的是模温低了二十度、冲填速度还是旧的，玻纤的排布方向跟着变了。
换料的账，常常不在配方单上，在成型窗口里。
开篇那三句问话，最后一节我再回收。
一、换料前把工况六维落下数字
SMT 连接器骨架的工况，头一条是炉子。
无铅工艺的峰值通常在 240–260℃，220℃ 以上的时间有 60–90 秒，整条曲线四到六分钟。
更要紧的是次数：双面贴装的板子过两次炉，返修还要再加一次。
把 260℃ 换成体感：PA66 的熔点就在 265℃ 附近，等于让这个件贴着熔点站两分钟，还没算第二次。
长期温度是第二条线。件装到板上以后，除回流焊那一瞬间，长期服役多在 105–125℃；汽车电子能到 125℃ 以上，还要跑十年。这个温度不算高，但它是连续的，跟炉子的短时峰值是两本账。
第三条是载荷。连接器的载荷不在本体，在端子和卡扣。端子压入力、插拔保持力，几千次插拔之后还得稳；骨架的作用是替端子把这些力接住。件本体强度再高，端子座一松，一切都白谈。
第四条是介质。助焊剂残留、清洗剂、湿气三样一起上。其中湿气最麻烦——尼龙吸湿之后尺寸会变，一个 30 毫米长的骨架，吸到一个百分点，长度方向就是零点零几毫米的变化，听着小，放进共面度公差里就不是小事。
第五条是外观。骨架件多是本色或黑色免喷涂，共面度、色差、浮纤三样都直接上判，没有涂层可以遮。
第六条是合规。阻燃按最小壁厚报 V-0，无卤看目标市场，湿敏等级要标到 MSL，随报告一起给。
六样里头两样给不出数字，换料就是拿批量去赌。把温度、回流次数、长期温度、寿命年限先问齐，方向自然会收窄。
| 维度 | 换料时要问的数 | 给不出会怎样 |
|---|
| 温度 | 实测峰值、220℃ 以上时长、过炉次数 | 档位选错，过炉才暴露 |
| 长期温度 | 服役区间与年限 | 老化后才显形 |
| 载荷 | 端子压入力与插拔次数 | 端子座先松 |
| 介质 | 助焊剂、清洗剂、车间湿度 | 湿态尺寸漂 |
| 外观 | 共面度、色差、浮纤限值 | 装配端退回 |
| 合规 | 阻燃、无卤、湿敏等级 | 认证一路卡 |
二、三条路线并列，先不急着分高下
换料不是换成最耐温的那一档，是把三条路的代价摆清楚，看哪条跟你的模具兜得住。
| 路线 | 可过的炉 | 取向与翘曲 | 干燥与流动 | 适合换自哪里 |
|---|
| PA46-GF30 | 中低峰值、单面件 | 中等 | 干燥要求紧、流动尚可 | 原 PA66-GF 余量不够 |
| PA6T-GF30 | 主流 SMT 件 | 可控，靠模温与浇口 | 需高模温、干燥紧 | 原 PA9T 交期或成本受限 |
| PA9T-GF30 | 高要求薄壁件 | 尺寸稳，窗口最窄 | 料温最高、流动性一般 | 原进口料停产或缺货 |
三条没有谁更好，只有哪条跟你的件匹配。
一个常见误判是"既然要换，就换到最高那一档"。耐温是够了，可是窗口最窄、料温最高、干燥最紧，模具和产线要跟着一起改，改不动的部分就成了新的风险点。
另一个误判是只比物性表上的耐温，不看取向。骨架件上，本体强度涨得平缓，取向差带来的翘曲却涨得快——换料之后让件装不上的，往往不是强度，是那零点几毫米。
再说到底层。玻纤是一根根短纤维，料在型腔里流动时，纤维会顺着流动方向躺下来。顺着流的那一侧收缩小，垂直的那一侧收缩大，两个方向的收缩率能差出好几倍。骨架件又长又有转折，这个差就被放大成整条的弯。
路线定下来之前，先把"原来用的是什么、为什么换"写成一页。是交期、是成本、还是原来那支料停产？原因不同，路线不同，后面的验证重点也不同。
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、你的炉、你的实测定下来。
| 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 |
|---|
| 熔点与炉内峰值余量 | 熔点高于实测峰值 ≥40℃ | ISO 11357 / 实测炉温曲线 | 本体软化、溢胶、引脚歪 | 往高温档位换 | 抗氧剂（耐温上限） |
| 过炉后共面度 | 漂移 ≤装配公差的二分之一 | 按实际曲线过炉 + 影像仪或三坐标 | 端子歪斜、虚焊 | 降取向差、重设模温 | — |
| 湿态弯曲保持 | 调湿后 ≥七成 | ISO 178 + 调湿处理 | 端子座发白、脆断 | 低吸水基材 + 调湿 | 偶联剂（玻纤界面） |
| 熔接线强度 | 按骨架根部应力另定 | 短射取样 + ISO 527 | 骨架根部开裂 | 提高模温、改浇口 | 润滑剂（影响熔接线） |
| 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 | — |
| 干燥后含水率 | PA6T / PA9T ≤0.05% | 水分仪或露点数据 | 银纹、气爆 | 除湿干燥、周转封闭 | — |
怎么读这张表：先看头一行和第二行。
耐温余量不够，是选料问题；共面度漂，是取向和工艺问题。两条线的解法完全不一样，别混着改。
第三列是留给采购的：验证方法要连标准号一起写进规格，写不清，后面就是扯皮。
也别一上来把玻纤含量往上加。含量涨上去，本体刚性涨得平缓，熔接线强度却掉得陡——骨架件的熔接线常常落在受力位置，这一项要单独取样。
四、换料后四种失效，和它们真正的原因
失效一：过炉后共面度超差。
头一个反应往往是"耐温不够，换更高档"。可换料之后玻纤含量和排布方向都变了，根因常在模温和冲填速度，不在牌号高低。先把模温按新料重设、把冲填速度降下来看取向，再谈要不要动料。这是本篇最想改掉的一个惯性。
失效二：端子座根部出现银纹，过炉后放大成微裂。
这类多半是干燥没跟上——这两类高温料的吸水速度比通用料快，拆包后敞口放几个小时，含水率就回来了。含水率超标，水分在料筒里把分子链剪断，件看着是好的，过炉那一遍才把账结清。
干燥这件事显形很晚。件做出来当天是好的，装到板上跑一阵才脆；追溯时料没变、模没变，变的是车间那几天的湿度。
失效三：同一批件黄得深浅不一。
这不是"料不稳定"，常见的原因是抗氧剂分散不均，或者料温超过了稳定体系的耐温上限。看到局部发黄先查混料、查料温，别急着换基材——这一条是助剂侧的账：料没选错，是稳定化体系没跟工况配到位。
失效四：骨架根部沿熔接线裂。
为什么熔接线这么脆？因为两股料流在型腔里相遇时，玻纤被推开、彼此没有缠结，界面就成了整条件上最弱的一根线。玻纤含量越高，这条线越弱。它要结构、浇口、模温一起动，光换料号解不开。
五、加工与验证：模温和干燥是先动的两项
模温在骨架件上比料温更值钱。
高温料的模温通常要往上抬一档；抬上去之后，料流冷却慢了，玻纤的取向差变小，共面度才收得住。具体抬到多少，按你的件和模具试出来，不能照搬上一支料的档位。
干燥要先换设备，再谈参数。这两类料要用除湿干燥，温度按体系定，时间按初始含水率定；上机前用水分仪或露点数据确认，不凭手感。南方梅雨季，拆包后的料在车间放几个小时含水率就能回升，干燥做得再好，周转环节敞口也是白做。
冲填速度是容易漏的第三项。速度一快，玻纤顺着流动方向排得更整齐，方向差更大；速度放下来，取向变匀。这一项和模温要联合调，不能单动一个。
验证顺序建议这样排，顺序不要换：
1. 材料级：含水率、干态与湿态强度保持率
2. 工艺窗口：变模温、变冲填，打对比件看取向
3. 件级：过炉后共面度、端子压入与保持力
4. 二次过炉：模拟双面贴装
5. 老化后过炉：把热老化后的样件再走一遍炉
为什么顺序不能换？因为共面度同时依赖含水率和取向。这两项没锁住就去调模温，调出来的窗口只对那一模有效，批量一放又漂。
还有一件换料时要顺带确认的事：浇口位置。浇口定了料流方向，也就定了玻纤的排布方向，也就定了收缩差的方向。换料之后如果熔接线正好落在高应力位置，先改浇口再看配方，通常比换料号省事。
六、边界：这几种骨架件，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，炉次已经排到三次以上的件。双面贴装加返修，材料要按三次峰值存活，高温料的余量会吃紧，这时该去看 LCP 或 PPS 的路线。
其二，共面度卡得极紧的长薄骨架。件越长越薄，取向差被放大得越狠；结构上又不能改浇口的时候，换料能救的空间很小，先动结构。
其三，单价已经压到很低的大批量件。这类件的账要算总，料价高出来的部分未必收得回来，留在原来的体系更稳。
其四，长期 150℃ 以上、或者周边有强电弧的位置。这类位置该往耐温更高的体系、或者热固性材料的方向看，普通换料填不上这个坑。
还有一类不该急着换的：失效点还没定位清楚的件。件裂了、歪了，先分清是熔化、是翘曲、还是气爆，三条线的解法完全不同。定位没做完就换料，换哪一支都是试。
把这四条写在前面不是劝退，是省时间。样品顺、批量卡、整案回退的学费，比一开始不换高得多。
七、换料风险清单（从原料换到这边的路线，要动的东西）
| 环节 | 要动什么 | 容易漏的点 |
|---|
| 模具 | 收缩率随基材改变，定位孔可能要修 | 只换料不修模，共面度先漂 |
| 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 | 强制调湿 + 称重判定 + 复测尺寸 | 按壁厚估时间，厚壁没吸透 |
| 料温 / 模温 | 模温是取向差的第一道闸，要重设 | 照抄上一支料的档位 |
| 保压 / 脱模 | 端子座与骨架根部要重新定保压 | 玻纤高时熔接线更脆 |
| 色差 | 本色与黑色件分别对色板 | 不同批次基材底色有差 |
| 验证顺序 | 含水→工艺→件级→二次过炉→老化 | 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）
我们给这类骨架件排的试模，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，只验含水率、外观、短射熔接线位置。这一轮先确认料能不能把骨架的细筋填满。留样两件，标注批号与干燥参数，至少留到第二轮结束。
第二轮·工艺窗口：固定料，变模温与冲填速度，打两组对比件。验过炉后共面度、湿态强度保持、端子压入与保持力。这一轮定量产参数。留样按批封存，至少留到量产稳定后三个月。
第三轮·过炉实测：按实际炉温曲线走两次炉，再把热老化后的样件走一遍。中途复测共面度与端子。这一轮过了，才建议放量。
三轮之间为什么不能跳？因为每一轮验的东西都是下一轮的前提：填充不确认，工艺窗口无从谈起；工艺窗口不定，共面度数据就只对那一模有效。
留样要连干燥参数、模温、冲填速度一起记，不能只写批号。真出问题回溯时，缺了参数记录，留样也只是一块塑料。
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：换料后件歪了，是耐温不够，还是取向变了？先量过炉前后的共面度曲线，再决定动不动料。这两件事的解法不一样。
问：原来的料只是交期长，不是不好，配方能照抄吗？物性能照抄，模温、冲填、干燥这三样跟着设备和模具走，照搬必踩。
问：模温抬高了会不会拉长周期？会，周期变长是抬模温的代价。这笔账要和共面度收住的收益一起算，别只看一头。
九、一页纸汇报表（给采购向上汇报用）
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 | 从原高温料换到 PA6T-GF 路线，先确认耐温余量 |
| 动什么 | 模温重设、冲填降速、干燥换除湿，三件一起动 |
| 验什么 | 过炉后共面度、湿态强度、熔接线，三项分工 |
| 什么时候能放量 | 三轮试模全过、老化后过炉复测通过 |
这四行写清楚，评审会上就少有回头路。
料有人卖，判断不一定有人给。
装到板上的那几颗端子、过炉后那零点几毫米的漂移，都是选料时就得留出的余量。回到开篇那三句问话：问件、问失效形态、问成型窗口——把这三样问清，换料才不是碰运气。
SMT connector material replacement: the parts and molds are on hand. The most easily overlooked factors are not temperature resistance, but orientation and mold temperature. This article explains why coplanarity drifts after material replacement, how to read the criteria table, what to check in the three rounds of trial molding, and the specific processes in frame parts that need re-sequencing after material replacement.
The matter of replacing SMT connectors happened last month when I ran into it at a factory that makes board-to-board connectors.
They originally used a grade of imported high-temperature material, but the delivery time was too long, so they want to switch to our route.
The frame parts were produced, and the appearance showed no defects. There were no short shots or gaps, and the workshop declared them qualified on the same day.
The terminals were pressed in, and the coplanarity of the whole row exceeded the limit. After mounting the board, a few solder joints were weak.
His exact words on the phone were: 'The material looks fine, so why does it warp when it goes through the oven?'
I first countered with three questions: Are we just changing the substrate, or adjusting the entire formula? Is the mold temperature set according to the original settings? Have we reviewed the orientation distribution again?
He paused for two seconds and said that the mold temperature had been adjusted, but the reset for the new material was not done.
This sentence points out the part of changing materials in SMT connectors that is most likely to be overlooked — the appearance and heat resistance of the components often pass, but the first problems arise with orientation and the warping caused by orientation.
The following timeline is the true course of that skeletal part of his.
The starting point was that the prototype part looked good and the assembly dimensions passed inspection, so the workshop breathed a sigh of relief; the latent phase was when the mass-produced parts slowly showed changes in coplanarity after going through the furnace, though a single piece didn't seem affected; the outbreak was when the terminals skewed after the entire board was mounted, and the client started returning boards; the settlement was a review, finding that the formula hadn't changed much, but the mold temperature was 20 degrees lower, the punching and filling speed remained the same, and the orientation of the fiberglass had changed accordingly.
The accounts for material changes are often not on the formula sheet, but in the molding window.
The first three questions at the beginning, I will revisit the last section.
1. Record the six-dimensional process data before changing the material
The working condition of the SMT connector skeleton, the first one is the furnace.
The peak temperature for lead-free processes is usually between 240–260°C, with a duration above 220°C of 60–90 seconds, and the entire profile lasts four to six minutes.
More importantly, it's the number of times: double-sided mounted boards go through the oven twice, and rework adds another time.
Convert 260℃ to what it feels like: The melting point of PA66 is around 265℃, which is like having this piece stand right at its melting point for two minutes, not counting the second time.
The long-term temperature is the second line. After the component is mounted on the board, except for the moment of reflow soldering, it usually operates at 105–125°C for long-term service; automotive electronics can reach above 125°C and still need to run for ten years. This temperature is not considered high, but it is continuous, which is a different matter compared to the short-term peak in the oven.
Article 3 is about load. The load on the connector is not on the body itself, but on the terminals and latches. The terminal pressing force and insertion/removal retention force must remain stable even after thousands of insertions and removals; the function of the framework is to take these forces on behalf of the terminals. No matter how strong the body of the part is, if the terminal housing loosens, everything is meaningless.
Article 4 is about the medium. Flux residue, cleaning agents, and moisture all come into play together. Among them, moisture is the most troublesome—after nylon absorbs moisture, its dimensions change. For a 30 mm long frame, if it absorbs one percent of moisture, the change in length is a fraction of a millimeter. It sounds small, but when placed within coplanarity tolerances, it is not a trivial matter.
Article 5 is appearance. The frame parts are mostly in their natural color or black without painting. Flatness, color difference, and fiber protrusion are all judged directly, as there is no coating to cover them.
Article 6 is about compliance. For flame retardancy, report V-0 according to the minimum wall thickness; for halogen-free, refer to the target market; moisture sensitivity level must be marked up to MSL and provided together with the report.
Out of the six items, two cannot provide numbers, and changing materials is basically gambling with bulk quantities. First ask about temperature, number of refluxes, long-term temperature, and service life; the direction will naturally narrow.
| Dimension | The numbers to ask when changing materials | Can't say what will happen |
|---|
| Temperature | Measured peak value, duration above 220°C, number of passes through the oven | The gear was chosen incorrectly, only revealed after passing through the furnace. |
| Long-term temperature | Service Interval and Duration | Only becomes visible after aging |
| Load | Terminal crimping force and insertion/removal cycles | Loosen the terminal block first |
| Medium | Flux, cleaning agent, workshop humidity | Wet state dimensional drift |
| Appearance | Coplanarity, color difference, floating fiber limits | Assembly End Return |
| Compliance | Flame retardant, halogen-free, moisture sensitivity level | Authentication All-in-One Card |
The three routes run side by side, no need to rush to rank them.
Changing the material doesn’t mean switching to the most heat-resistant option; it’s about laying out the costs of the three options clearly and seeing which one your mold can handle.
| Route | Passable furnace | Orientation and Warping | Drying and Flow | Where is it suitable to change from? |
|---|
| PA46-GF30 | Medium-low peak, single-sided part | Medium | Drying requirements are strict, flow is acceptable | The remaining amount of original PA66-GF is insufficient |
| PA6T-GF30 | Mainstream SMT components | Controllable, dependent on mold temperature and gate | Requires high mold temperature and dry tightness | Original PA9T delivery time or cost constrained |
| PA9T-GF30 | High-precision thin-walled parts | Size is stable, window is the narrowest | Highest material temperature, average fluidity | Original imported materials are discontinued or out of stock |
None of the three is better; it depends on which one matches your item.
A common misjudgment is 'If you are going to change, change to the highest level.' The temperature resistance is sufficient, but the window is the narrowest, the material temperature is the highest, and drying is the tightest. The mold and production line need to be modified accordingly, and the parts that cannot be changed become new risk points.
Another misjudgment is only comparing the heat resistance on the material property table without considering the orientation. For frame parts, the increase in the main body strength is gradual, but the warping caused by poor orientation rises quickly — after changing the material, what often prevents the part from fitting is not the strength, but those fractions of a millimeter.
Now let's talk about the base layer. Glass fibers are short fibers, and when the material flows in the mold cavity, the fibers lay down following the flow direction. The side along the flow shrinks less, while the side perpendicular to it shrinks more, and the shrinkage rates in the two directions can differ by several times. Structural parts are long and have bends, and this difference gets amplified into bends along the entire piece.
Before finalizing the route, first write down on a page 'what was originally used and why it is being changed.' Is it because of delivery time, cost, or because the original material was discontinued? Different reasons lead to different routes, and the focus of subsequent verification will also differ.
3. Material Change Criteria Table: This table determines which items you will re-inspect
Turn the previous constraints into measurable indicators. The thresholds in the table below are directional suggestions, not acceptance criteria; the actual values must be determined by your parts, your furnace, and your actual measurements.
| Indicator | Directional threshold | Verification Method / Standard | Common failures after material change | Common solution | Corresponding auxiliary system |
|---|
| Melting point and peak in-furnace margin | Melting point higher than measured peak ≥40℃ | ISO 11357 / Measured Furnace Temperature Curve | Body softening, glue overflow, bent pins | Switch to a high-temperature setting | Antioxidant (Maximum Temperature Limit) |
| Degree of uniformity after passing through the furnace | Drift ≤ half of the assembly tolerance | Pass through the furnace according to the actual curve Imaging instrument or three-coordinate | Bent terminals, cold solder joints | Reduce orientation difference, reset mold temperature | — |
| Wet-state bending retention | After moisture adjustment ≥ 70% | ISO 178 Moisture Conditioning | Terminal block is whitening and brittle | Low water absorption substrate Humidity adjustment | Coupling agent (glass fiber interface) |
| Weld line strength | Should be determined according to the stress at the root of the skeleton | Short-shot sampling ISO 527 | Cracking at the base of the skeleton | Increase mold temperature and modify the gate | Lubricant (affects weld lines) |
| 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 | — |
| Moisture content after drying | PA6T / PA9T ≤0.05% | Moisture meter or dew point data | Silver lines, air burst | Dehumidification and drying, turnover closure | — |
How to read this table: first look at the first row and the second row.
Insufficient temperature margin is a material selection issue; coplanarity deviation is an orientation and process issue. The solutions for the two problems are completely different, so don't try to fix them together.
The third column is reserved for procurement: the verification method should be written into the specifications along with the standard number; if it's not clear, there will be disputes later.
Also, don't immediately increase the fiberglass content. When the content goes up, the rigidity of the main body increases gradually, but the weld line strength drops sharply—the weld lines of the frame parts often fall in the stress-bearing areas, so this needs to be sampled separately.
4. Four types of failures after material replacement, and their real causes
Failure 1: Excessive coplanarity after reflow.
The first reaction is often 'the temperature resistance is not enough, switch to a higher-grade material.' But after changing the material, both the fiberglass content and orientation are different. The root cause is often in the mold temperature and filling speed, not in the grade. First, reset the mold temperature according to the new material and reduce the filling speed to observe the orientation, then discuss whether the material needs to be changed. This is one inertia I most want to change in this article.
Failure 2: Silver streaks appear at the base of the terminal, which enlarge into microcracks after passing through the furnace.
These cases are mostly due to insufficient drying—the moisture absorption rate of these two types of high-temperature materials is faster than that of general-purpose materials. After opening the package and leaving it exposed for a few hours, the moisture content returns. If the moisture content exceeds the limit, the moisture in the barrel will cut the molecular chains. The parts may look fine, but going through the oven is when the problem is actually revealed.
The drying issue became apparent quite late. The piece was fine on the day it was made, but it only became brittle after being mounted on the board and run for a while; upon tracing back, the materials hadn’t changed, the mold hadn’t changed, what had changed was the humidity in the workshop during those few days.
Failure 3: The same batch of items has inconsistent yellowing.
This is not 'unstable material.' Common reasons are uneven dispersion of antioxidants, or the material temperature exceeding the thermal limit of the stable system. If you see local yellowing, first check the mixing and the material temperature, don’t rush to change the base material — this is an account on the additive side: the material was not chosen incorrectly, it’s just that the stabilization system was not matched to the process conditions.
Failure 4: Cracks along the weld line at the base of the frame.
Why is the weld line so brittle? Because when the two material flows meet in the cavity, the glass fibers are pushed aside and do not entangle with each other, making the interface the weakest line under all conditions. The higher the glass fiber content, the weaker this line is. It requires the structure, gate, and mold temperature to work together; simply changing the material grade cannot solve it.
5. Processing and Verification: Mold temperature and drying are the first two actions.
Mold temperature is more valuable than material temperature on the frame components.
The mold temperature for high-temperature materials usually needs to be raised by one level; after raising it, the material flow cools more slowly, the fiber orientation difference decreases, and the coplanarity can be controlled. The specific level to raise it to should be tested based on your part and mold, and you cannot directly copy the setting from a previous material.
For drying, the equipment must be replaced first, and then the parameters can be discussed. These two types of materials should use dehumidifying drying; the temperature is determined according to the system, and the time is determined according to the initial moisture content. Before feeding into the machine, confirm with a moisture meter or dew point data, not by hand feeling. In the southern plum rain season, the moisture content of unpacked materials can rise after being left in the workshop for a few hours; no matter how well the drying is done, an open transfer stage is still futile.
Filling speed is the third item that is prone to defects. When the speed is fast, the fiberglass aligns more neatly along the flow direction, resulting in greater directional differences; when the speed is reduced, the orientation becomes more uniform. This factor needs to be adjusted together with mold temperature and cannot be adjusted independently.
It is recommended to arrange the verification sequence like this, do not change the order:
1. Material level: moisture content, dry and wet strength retention
2. Process window: Change mold temperature, change injection filling, make comparison parts to check orientation
3. Component Level: Coplanarity after soldering, terminal insertion, and retention force
4. Second Pass Through the Oven: Simulating Double-Sided Mounting
5. Post-aging furnace pass: Run the heat-aged sample through the furnace once more
Why can't the order be changed? Because coplanarity depends on both moisture content and orientation. If these two factors aren't locked in and you adjust the mold temperature, the resulting window will only be effective for that specific mold, and when producing in bulk, it will drift again.
There is one more thing to check when changing materials: the gate location. The gate determines the direction of the material flow, which in turn determines the orientation of the glass fibers and thus the direction of shrinkage differences. If, after changing materials, the weld line happens to fall in a high-stress area, it is usually easier to first change the gate location before adjusting the formulation.
6. Boundaries: For these types of framework components, stop changing materials first
This section may be more valuable than the previous few sections because it helps you cut losses before starting work.
First, parts that have already been through the furnace more than three times. For double-sided mounting with rework, the materials need to survive according to the three-time peak value, and the remaining high-temperature materials will be tight. At this time, one should look at the LCP or PPS route.
Second, the long thin frame with extremely tight coplanar dimensions. The longer and thinner the piece, the more severely any misalignment is amplified; when the structure cannot be modified at the gate, the room for improvement by changing the material is very limited, so the structure is adjusted first.
Third, large-volume items whose unit prices have already been pushed very low. For these items, the accounts should be calculated in total, and the part where the material price is higher may not be recoverable, so it is more stable to keep them in the original system.
Fourth, positions above 150°C for long periods, or with strong electric arcs nearby. For such positions, one should consider systems with higher temperature resistance, or thermosetting materials; ordinary replacement materials cannot fill this gap.
There is another type that should not be replaced hastily: parts whose failure points have not yet been pinpointed. If a part is cracked or warped, first determine whether it is due to melting, warping, or a gas explosion, as the solutions for these three scenarios are completely different. Replacing the material before completing the fault localization is just a trial, no matter which one you replace.
Writing these four points at the beginning is not meant to discourage, it's to save time. The learning costs for smooth samples, blocked batches, and cases being returned are much higher than changing at the beginning.
7. Material Change Risk Checklist (the route from raw materials to this side, items to be moved)
| Steps | What needs to be moved | Points prone to leakage |
|---|
| Mold | Shrinkage rate changes with substrate, positioning holes may need repair | Only material replacement, no mold repair, coplanarity floats first |
| Drying | Replace the dehumidifier dryer, set the window based on actual measured moisture content | Hot air drying is basically ineffective for these two types of materials |
| Humidity adjustment | Forced humidity adjustment + weighing determination + re-measurement of dimensions | Estimate time based on wall thickness, thick walls not fully absorbed |
| Material temperature / mold temperature | Mold temperature is the first barrier for orientation difference; reset | copy the previous material's setting |
| holding pressure / demolding | Terminal block and base of the frame must be re-held at holding pressure | When glass fiber is high, the fusion line becomes more brittle |
| Color difference | Match the color plates for natural and black parts separately | Differences in base color between batches |
| Verification sequence | Water-containing → process→ Component-level → secondary furnace → aging | If the previous item fails, proceed to the next step |
8. Sample mold trial scheduling (how many rounds of machine loading, what to test in each round, how long to keep samples)
We usually schedule mold trials for these types of skeleton parts, usually in three rounds, with no skipping between rounds.
First round · Small sample comparison: Use your original mold to make 3–5 molds, only checking moisture content, appearance, and short shot welding line position. For this round, first confirm whether the material can fill the fine ribs of the frame. Retain two samples, mark batch numbers and drying parameters, and save them at least until the end of the second round.
Second round · Process window: Fix the material, change mold temperature and filling speed, and make two sets of comparison pieces. After passing the furnace, check the coplanarity, wet strength retention, terminal pressing, and holding force. This round of quantitative mass production parameters. Retain the samples by batch and seal them at least until three months after mass production stabilizes.
Third round · Pass-through test: run the actual furnace temperature curve twice, then run the heat-aged sample once. Retest the coplanarity and terminals midway. Only after this round is it recommended to increase volume.
Why can't you skip between three rounds? Because each round of inspection is a prerequisite for the next round: if the filling isn't confirmed, the process window can't be discussed; If the process window is uncertain, the coplanarity data only applies to that mold.
Sample retention must be recorded together with drying parameters, mold temperature, and filling speed; you can't just write the batch number. If a problem really arises and you trace back, the parameter records are missing, and the sample is just a piece of plastic.
The additive system in the formula is matched according to the working condition of the piece—regular additives are always in stock, special models are matched as needed; You report the operating conditions and grade, and the materials and additives are all prepared at once.
Three Questions Readers Often Ask
Question: If the parts get crooked after material change, is it due to insufficient temperature resistance, or has the orientation changed? First, measure the coplanarity curve before and after the furnace, then decide whether to change the material. The solutions for these two matters are different.
Question: The original material just has a long lead time, not a bad one. Can the formula be copied directly? If the material properties are copied, mold temperature, intake, and drying follow the equipment and mold, so copying must be done.
Question: Will raising the mold temperature extend the cycle? Yes, the longer cycle is the cost of raising the mold temperature. This account should be calculated together with the income from coplanarity; don't just look at one side.
9. One-Page Report Form (for reporting to procurement officials)
If you want to use this to report, you can collect four lines.
| Item | One-sentence conclusion |
|---|
| What to replace | Switching from original high-temperature material to PA6T-GF route, first confirm the temperature resistance margin |
| Move what | Mold temperature reset, flush and fill speed reduction, drying and dehumidification, all three moving together |
| Inspect what | After passing the furnace, coplanarity, wet strength, welding lines, the three divisions of labor |
| When will the volume ramp up | All three rounds of mold trials, after aging and furnace retesting, pass |
Write these four lines clearly, and there will be few turning back at the review meeting.
Material is sold by someone, but judgment doesn't always mean someone will give it.
Those terminals installed on the board, the fractional millimeter drift after passing through the furnace are all margins that must be left when selecting materials. Back to the three questions at the beginning: asking about parts, failure patterns, and molding windows—clarifying these three makes material change not just a matter of luck
