齿轮箱壳体既要刚性，又躲不开嵌件和熔接线。这篇讲清六维工况怎么落数字、三条材料路线各差在哪、判据表怎么读，以及哪几种结构不该走玻纤尼龙这条路。
前天开完会回到办公室，桌上压着一张图纸。
是一家做电动工具齿轮箱的客户传过来的，标的是一处嵌件位，红色的圈里写着四个字："这里常裂。"
图纸上还留了行小字：M4 铜嵌件，滚花加深一档试试。
我按着这个思路往下想了一步，然后拿起电话打了过去。
他接起来先说的是："同一套模具、同一批嵌件，上一批没事、这一批就裂，到底是哪边的问题？"
我追问了三句：电机附近的实际温度多少、堵转温度算不算进去？嵌件是模内植入还是热压？装配扭矩是多少？
他答：连续工作 80℃ 上下，堵转能到 150℃；模内植入；扭矩定在 2.5 N·m。
三样答完，方向基本就定了。
这批壳体的经过，是一条批间波动的线。
起点是模具验收那一批全部合格，装配线没提过问题；潜伏是隔一阵冒出一两件嵌件位开裂，被当成个案处理；爆发是某个月集中裂了十几件，装配线停线；结算是把当月的模温记录调出来比对，最低的那两天正好对上了开裂的批号。
这篇把齿轮箱壳体材料这笔账讲清：六维工况怎么落数字，三条路线各差在哪，判据表怎么读，以及哪些结构不该走这条路。
一、工况六维：齿轮箱壳体被什么约束
电动工具的齿轮箱壳体，通常分前后两件，包着齿轮组和电机输出端。
它的工况跟一般外壳最大的不同是：热、力、扭矩三样同时在，而且都在很小的空间里。
温度这一维要按两个档位看。
连续工作时，靠近电机的那一端在 80℃ 上下；堵转或者卡钻的瞬间，局部能冲到 150℃。
这两档必须分开写进协议——连续温度决定长期老化，堵转温度决定短时刚性。
载荷这一维主要是啮合反力。
齿轮咬合时会把径向力和轴向力压到壳体上，这个力不大，但方向一直在变，属于循环载荷。
所以壳体要有刚性，不是为了扛住一次冲击，是为了让齿轮的中心距在长期运行里不走位。
介质这一维容易被忽略。
齿轮箱里有润滑脂，脂里的某些成分会让尼龙表面析出、发黏，也会影响长期性能；外部的切削液、粉尘、潮气也会进来。
寿命这一维按整机标准算。
专业级电动工具按几百小时累计运行算，家用级按几十小时算；次数不是重点，耐温老化才是。
外观这一维看表面：浮纤、色差、熔接线痕迹，工具是外观件，用户一眼能看到。
合规这一维主要是耐温等级、阻燃（部分场景）、以及出口市场的电气安全要求。
| 维度 | 电机侧 | 输出侧 | 漏了会怎样 |
|---|
| 温度 | 连续 80℃，堵转 150℃ | 相对低，齿轮啮合生热 | 长期老化算小 |
| 载荷 | 电机反扭矩 | 齿轮啮合循环反力 | 中心距走位 |
| 介质 | 润滑脂、粉尘 | 切削液、粉尘、潮气 | 表面析出发黏 |
| 寿命 | 数百小时累计 | 同左 | 只比单价 |
| 外观 | 浮纤、色差、熔接线痕 | 同左 | 客诉 |
| 合规 | 耐温、阻燃、电气安全 | 同左 | 出口卡住 |
把六维摆在一起，会看到一个结论：齿轮箱壳体是一个"三件事都不允许松"的件——刚性要够，熔接线要强，嵌件周围要稳。
这三件事在同一件上经常互相打架，这就是它的难点。
二、三条路线，并列摆开
先搞清玻纤在尼龙里做了什么。
玻纤的作用是把载荷从树脂基体接过去，所以它带来的是模量——也就是刚性。
代价有两个：一是熔接线，二是缺口敏感。
熔接线的道理不复杂：料流分成两股再汇合的时候，玻纤是顺着流动方向排的，汇合面上两条流向正好对上，玻纤不能跨过去交缠。
所以熔接线上的强度，靠的是树脂本身，玻纤帮不上忙。
| 路线 | 弯曲模量量级 | 熔接线强度 | 耐温与老化 | 适合哪种位置 |
|---|
| PA6-GF30 | 约 8–9 GPa | 保留率偏低，靠工艺补 | 连续 80℃ 量级可用 | 输出侧壳体、通用件 |
| PA66-GF30 | 约 9–10 GPa | 同上，窗口更高 | 耐温更稳，适合电机侧 | 电机侧、高温件 |
| PA6-GF30 + 矿物或玻微珠混填 | 略降一档 | 相对更好，翘曲更小 | 同 PA6 体系 | 尺寸严、翘曲敏感的壳体 |
三条路线没有谁更好，只有哪一条跟你的温度、结构和装配方式兜得住。
PA6-GF30 是这一行的主流：刚性够、成本可控、工艺成熟，代价是吸水偏高、熔接线偏弱。
PA66-GF30 的优势在耐温，适合靠近电机的那一端；代价是加工窗口更高，模温料温都要往上走。
混填那条的取舍是拿一部分刚性换尺寸稳定：矿物或玻微珠让收缩更接近各向同性，翘曲小，熔接线也相对好一点。
代价是模量下来一档、韧性也下来一点，用在强度要求高的位置要重新核。
一个常见的误判是：熔接线强度不够，那就把料温再往上提。
料温上提能改善愈合，但这条线很快撞到上限——温度过高，材料开始降解，强度和外观一起变差。
真正有效的是三件事：把熔接线推离受力位置、提高模温、把浇口位置排好。
三、选型判据表：这张表决定你验哪几项
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，实际数值要由你的件、你的结构和实测确定。
| 指标 | 方向性门限 | 验证方法 / 标准 | 常见失效 | 通行解法 | 对应助剂体系 |
|---|
| 弯曲模量 | 按结构核算，常见 8–9 GPa 量级 | ISO 178 | 中心距走位、异响 | 玻纤增强 + 结构补强 | 偶联剂（硅烷类） |
| 熔接线强度 | 按项目档定，通常会单独取件 | 熔接线位置取样拉伸 | 熔接线处开裂 | 浇口排布 + 提模温 | 润滑剂（内外平衡，过量致愈合差） |
| 嵌件扭转与拔出 | 按装配扭矩留余量，常见 1.5 倍 | 扭矩试验 / 拉脱试验 | 嵌件转动、拉脱 | 壁厚均匀 + 料温模温到位 | —（属结构与工艺） |
| 嵌件位热循环 | 循环后界面无裂纹 | 温度循环 + 切片 | 嵌件位开裂 | 减小膨胀差 + 结构圆角 | —（属结构与工艺） |
| 长期热氧保留率 | 按连续温度做数百小时档 | ISO 188 / ISO 527 | 件发黄、发脆 | 稳定化体系 | 抗氧剂（含铜害抑制） |
| 润滑脂相容性 | 浸脂后尺寸与外观无异常 | 脂样浸泡实测 | 表面析出、发黏 | 提前确认脂体系 | 润滑剂（迁移性一并评估） |
| 翘曲与关键尺寸 | 关键装配尺寸按调湿态定 | ISO 1110 调湿 + 量测 | 装配间隙、异响 | 调湿态出图与验收 | —（属状态管理） |
怎么读这张表，先看前两行。
弯曲模量和熔接线强度是一对，考的是同一件上的两个相反方向：本体要硬，接缝要韧。
第三、第四行是嵌件位，也是这一篇最该逐条核的两项。
嵌件的问题很少是嵌件本身，多数是"塑料这一侧给不给得住"——包紧力够不够，壁厚匀不匀，界面有没有应力集中。
第五行是这两年暴露比较多的一项，尤其是嵌件是铜件的时候。
聚酰胺遇到铜离子，热氧老化的速度会明显加快，件表面会先发黄、再变脆；这一条要在选材时就提出来，不能等到客户端发现问题。
四、四种失效，和它们真正的根因
失效一：嵌件位开裂，裂纹从嵌件向外放射。
这一类的根因通常是三件事叠在一起：金属与塑料的膨胀差、嵌件周围壁厚不均、以及注塑内应力。
膨胀差可以算一笔账：钢的线膨胀系数约 11×10⁻⁶，玻纤增强尼龙约 30×10⁻⁶ 上下，差出近两倍。
整机从冷态 -20℃ 到运行 100℃，温差 120℃；按直径 10 毫米的嵌件位算，相对变形量在千分之二上下。
千分之二听起来很小，但乘上 8 GPa 的模量，环向应力就接近 20 MPa——这个量级已经足够在有应力集中的地方把件撕开。
所以这一类问题的功夫要花在结构上：把嵌件周围的壁厚做匀、把圆角做出来、把装配扭矩留住余量。
失效二：一批好、一批裂，模具和嵌件都没换。
这是本篇最想改掉的一个惯性。
很多人的思路是把嵌件滚花加深一档，觉得咬得更牢就不会裂。
方向反了。滚花越深，嵌件周围的应力集中越厉害，裂纹往往起得更早。
同一套模具、同一批嵌件出现批间差异，先查的是模温和料温的稳定性、干燥是否到位、以及嵌件预热温度是否一致。
这些变量在同一个车间里，一天之内就能差出好几度。
失效三：熔接线位置开裂，裂纹顺着接缝走。
熔接线上的强度，靠的是树脂愈合，玻纤帮不上忙。
所以多孔件、多浇口件的熔接线位置，要看它落在哪里——落在受力面或者嵌件旁边，就是隐患。
处理顺序是：先排浇口把熔接线推走，再提模温改善愈合，最后才谈换料。
失效四：件表面发黄、发脆，用久了还起粉。
这一类要往两个方向查。
一个方向是长期热氧老化：靠近电机的件长期在 80℃ 上下，稳定化体系不够就会先黄后脆。
另一个方向是铜害：铜嵌件在湿气加温度的条件下会析出铜离子，催化聚酰胺的降解，这种情况下的发黄往往集中在嵌件周围一圈。
这两个方向的解法不同，所以要先看黄的位置——是整件还是局部。
五、加工与验证：干燥、模温、嵌件
干燥在尼龙上永远是排在前面的动作。
上机前要做到露点 -40℃ 以下，含水率压到 0.15% 以内；水分进料筒之后会在高温下把分子链切短，韧性和熔接线强度一起受影响。
模温在齿轮箱壳体上有两个作用。
一是让表面致密、熔接线愈合得好；二是让收缩更均匀，嵌件周围的包紧力更一致。
模温低，这两件事同时变差，而车间里最常见的波动恰好就是模温。
嵌件的处理有三个动作要做对。
一是预热：嵌件冷着植入，周围的塑料被局部急冷，收缩不均匀，应力就留在那里了。
二是位置：嵌件的轴向与径向位置要在模具里定住，不能靠塑料本身去"卡"。
三是圆角：塑料侧的圆角比嵌件侧的滚花更有用，这一点跟直觉是反的。
调湿这一道按件的精度要求定。
嵌件位、轴承位这类配合尺寸，建议按调湿态出图与验收；纯外观的配合面可以放宽，但状态要写进协议。
验证顺序建议这样排，不要换：
1. 材料级：弯曲模量、熔接线取样强度、连续温度下的热氧保留率
2. 工艺窗口：变模温与料温，看熔接线外观与嵌件位收缩
3. 件级：嵌件扭转与拉脱、调湿后关键尺寸、外观
4. 循环级：温度循环后切片看嵌件界面
5. 装机级：按实际工况跑，包含堵转工况
顺序为什么不能换？因为嵌件位的开裂既依赖温度，也依赖装配扭矩，状态没定，前面测出来的数只对那一批有效。
六、边界：这几种结构，先别走玻纤尼龙这条路
这一段可能比前面几段更值钱，因为它帮你在开模之前止损。
其一，堵转温度长期超过 150℃ 的位置。这个量级下长期运行的性能保留数据支撑不足，该往高温尼龙或者金属件看。
其二，嵌件周围的壁厚做不到均匀的薄壁结构。塑料侧的包紧力是结构给的，结构给不了，换料也补不上。
其三，装配扭矩要求极高、且需要反复拆装的位置。这类位置该用金属螺纹套或者回到金属件。
其四，长期泡在强溶剂或者强碱性介质里的位置。尼龙在这类介质里的长期表现支撑不足。
把这四条写前面不是劝退，是省时间——齿轮箱壳体开模不便宜，验证也要整机配合，选错一次从头来的代价很大。
换料风险清单（从原体系换到玻纤增强路线，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，尺寸链要重排 | 只换料不核模 |
| 干燥 | 换除湿干燥机，按实测含水率定窗口 | 潮湿季节热风干燥基本无效 |
| 料温 / 模温 | 提模温改善熔接线愈合，不照抄上一支 | 模温低导致接缝弱 |
| 浇口与熔接线 | 重排浇口，把熔接线推离受力与嵌件位 | 接缝落在嵌件旁 |
| 嵌件 | 预热到位、位置定住、塑料侧做圆角 | 靠加深滚花求牢 |
| 调湿 | 配合尺寸按调湿态出图与验收 | 按干态尺寸放行 |
| 验证顺序 | 材料级 → 工艺窗口 → 件级 → 循环级 → 装机级 | 前一项没过就往下走 |
一页纸汇报表（给要向上汇报的人）
| 项 | 一句话结论 |
|---|
| 换什么 | 电机侧看耐温路线，输出侧与通用件可按成本路线走 |
| 动什么 | 浇口重排、模温料温重定、干燥换除湿、嵌件预热与圆角 |
| 验什么 | 弯曲模量、熔接线取样强度、嵌件扭转与拉脱、热循环后切片 |
| 什么时候能放量 | 熔接线不裂、嵌件位无裂纹、堵转工况跑通 |
读者常问的三句
问：嵌件老裂，是不是嵌件选小了？
先看三件事：嵌件周围壁厚匀不匀、模温稳不稳、装配扭矩留没留余量。这三件占了大头，换更大的嵌件常常解决不了问题，还可能把壁厚挤得更薄。
问：熔接线强度能靠加玻纤补吗？
补不了。玻纤顺着流动方向排，两条流向在接缝上对不上，跨不过去交缠；接缝的强度靠树脂愈合。所以熔接线的功夫在浇口、模温和结构，不在配方。
问：铜嵌件是不是比钢嵌件更容易出问题？
从长期老化看，铜离子会催化聚酰胺的热氧降解，这一点要提前考虑。做法通常是提高稳定化体系的档位，同时把防潮和封边一起做。
结语
回到开头那三句追问：电机附近的温度、嵌件的植入方式、装配扭矩。
这三样答全了，齿轮箱壳体材料往哪条路线走基本就定了。
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。
前两天接了个电话，问的就是那张图纸上"这里常裂"的四个字。
嵌件位这件事，难的不是嵌件，是它周围那一圈塑料——壁厚、模温、圆角，这三样定了，裂不裂基本就定了。
我们做改性尼龙（PA6 / PA66 / PA46 / PA11 / PA12 / PA6T / PA9T 及尼龙合金），也做改性 PPO / PPS 与热塑性弹性体；另长期收尼龙原料、水口回料与各类尼龙废料，有正规处置渠道。
The gearbox housing needs to be rigid, yet it can't avoid inserts and weld lines. This article explains how to assign numbers in six-dimensional working conditions, the differences among the three material routes, how to read the criteria table, and which structures shouldn't follow the glass fiber nylon route.
The day before yesterday, after finishing the meeting and returning to the office, there was a blueprint on the desk.
It was sent over by a customer who makes electric tool gearboxes. The mark points to an insert position, and inside the red circle are four characters: '常裂 here.'
There were also small notes left on the drawing: M4 brass insert, try deepening the knurling by one level.
I followed this line of thought one step further, then picked up the phone and called.
He picked up the call and first said: 'Using the same set of molds and the same batch of inserts, the last batch was fine, but this batch is cracking. Where exactly is the problem?'
I asked three follow-up questions: What is the actual temperature near the motor? Does the stall temperature count? Are the inserts molded in or hot-pressed? What is the assembly torque?
He replied: continuous operation around 80°C, can reach 150°C when stalled; insert molded; torque set at 2.5 N·m.
After answering the three questions, the direction is basically set.
The process of this batch of shells is a line of fluctuations between batches.
The starting point was that the batch of molds passed inspection with all items qualified, and the assembly line did not report any issues; the lurking stage was when one or two insert cracks appeared intermittently and were treated as isolated cases; the outbreak was when a dozen or so items cracked in a particular month, causing the assembly line to stop; the settlement was to pull out the mold temperature records for that month for comparison, and the lowest two days coincided exactly with the batch numbers that had cracks.
This article clarifies the accounting of gearbox housing materials: how to derive numbers under six-dimensional working conditions, where the three routes differ, how to read the criteria table, and which structures should not follow this path.
1. Six-dimensional working condition: What constraints are applied to the gearbox housing
The gearbox housing of the power tool is usually divided into two parts, front and rear, enclosing the gear set and the motor output end.
The biggest difference between its working conditions and those of a typical casing is that heat, force, and torque all occur simultaneously, and all within a very small space.
The temperature dimension should be viewed according to two levels.
During continuous operation, the end near the motor is around 80°C; at the moment of stall or bit jamming, the local temperature can spike to 150°C.
These two items must be written separately in the agreement—the continuous temperature determines long-term aging, and the stall temperature determines short-term rigidity.
This dimension of load is mainly the meshing reaction force.
When the gears mesh, they press the radial and axial forces onto the housing. This force is not large, but its direction keeps changing, which belongs to cyclic loading.
So the housing needs to be rigid, not to withstand a single impact, but to keep the center distance of the gears from shifting during long-term operation.
This dimension of the medium is easy to overlook.
There is grease in the gearbox, and some components in the grease can cause the nylon surface to exude and become sticky, which will also affect long-term performance; external cutting fluids, dust, and moisture can also get in.
The lifespan dimension is calculated according to the complete machine standard.
Professional-grade power tools are measured by hundreds of hours of cumulative operation, while household-grade ones are measured by dozens of hours; the number of uses is not the key point, heat resistance and aging are.
From the perspective of appearance, look at the surface: floating fibers, color differences, weld marks. The tool is the external part, which the user can see at a glance.
The compliance dimension mainly involves temperature rating, flame retardancy (in certain scenarios), and electrical safety requirements for export markets.
| Dimension | Motor side | Output side | What will happen if it leaks? |
|---|
| Temperature | Continuous 80°C, stall 150°C | Relatively low, gear meshing generates heat | Long-term aging is considered small |
| Load | Motor counter torque | Gear Meshing Cyclic Reaction Force | Center distance positioning |
| Medium | Grease, dust | Cutting fluid, dust, moisture | Surface precipitation makes it sticky |
| Lifespan | Hundreds of hours accumulated | Same as the left | Only compare unit price |
| Appearance | Floating fibers, color difference, weld marks | Same as the left | Customer complaint |
| Compliance | Temperature resistance, flame retardancy, electrical safety | Same as the left | The export is stuck |
Putting the six dimensions together, you will see a conclusion: the gearbox housing is a component where 'nothing is allowed to be loose' — the rigidity must be sufficient, the weld lines must be strong, and the areas around the inserts must be stable.
These three things often conflict with each other on the same matter, and that is its difficulty.
Two or three routes, placed side by side
First figure out what the fiberglass is doing in the nylon.
The role of fiberglass is to transfer the load from the resin matrix, so what it brings is modulus—that is, stiffness.
There are two costs: one is the weld line, and the other is notch sensitivity.
The principle of weld lines is not complicated: when the material flow splits into two streams and then merges, the glass fibers are aligned in the direction of the flow. On the merging surface, the two flow directions exactly meet, so the glass fibers cannot cross over and entangle.
So the strength on the weld line depends on the resin itself; the glass fiber can't help.
| Route | Order of magnitude of bending modulus | Weld line strength | Temperature Resistance and Aging | Suitable for which position |
|---|
| PA6-GF30 | Approximately 8–9 GPa | The retention rate is relatively low, compensated by the process. | Continuously usable at around 80°C | Output side housing, general parts |
| PA66-GF30 | Approximately 9–10 GPa | Same as above, the window is higher | More temperature-resistant and stable, suitable for the motor side | Motor side, high-temperature components |
| PA6-GF30 Mineral or glass microsphere filled | Drop slightly by one level | Relatively better, less warping | Same as PA6 system | Tight-tolerance, warp-sensitive casing |
None of the three routes is better; it's only about which one matches your temperature, structure, and assembly method.
PA6-GF30 is the mainstream in this line: it has enough rigidity, controllable cost, and mature processing techniques, with the trade-off being higher water absorption and weaker weld lines.
The advantage of PA66-GF30 lies in its heat resistance, making it suitable for the end near the motor; the cost is a higher processing window, with both mold temperature and material temperature needing to go up.
The trade-off of using hybrid filler is exchanging some rigidity for dimensional stability: minerals or glass microspheres make shrinkage closer to isotropic, reduce warping, and also result in relatively better weld lines.
The cost is that the modulus drops by one level, and the toughness also decreases slightly. When used in positions with high strength requirements, it needs to be re-evaluated.
A common misjudgment is: if the weld line strength is insufficient, just raise the material temperature further.
Raising the material temperature can improve curing, but this line quickly hits the upper limit—if the temperature is too high, the material begins to degrade, and both strength and appearance deteriorate.
The three things that are truly effective are: pushing the weld line away from the stress position, raising the mold temperature, and arranging the gate position properly.
3. Selection Criteria Table: This table determines which items you will inspect
Translate the previous constraints into verifiable metrics. The thresholds in the table below are directional suggestions, not acceptance criteria; the actual values need to be determined by your components, your structure, and actual measurements.
| Indicator | Directional threshold | Verification Method / Standard | Common Failures | Common solution | Corresponding auxiliary agent system |
|---|
| Flexural Modulus | According to structural calculations, the common magnitude is 8–9 GPa | ISO 178 | Center distance movement, abnormal noise | Glass fiber reinforced Structural reinforcement | Coupling agent (silane type) |
| Weld line strength | According to the project file, items are usually picked up separately. | Splice line location sampling tensile test | Cracking at the weld line | Gate layout Mold releasing temperature | Lubricant (internal and external balance, excessive amounts lead to poor healing) |
| Insert twisting and pullout | Leave a margin according to the assembly torque, commonly 1.5 times | Torque Test / Pull-Out Test | Insert rotation and pull-out | Wall thickness is uniform, material temperature and mold temperature are in place | —(Belongs to Structure and Technology) |
| Insert Position Thermal Cycling | No cracks on the interface after cycling | Temperature cycling Slices | Insert cracking | Reduce expansion difference Structural fillet | —(Structural and Process Attributes) |
| Long-term thermal oxygen retention rate | Set to continuous temperature for hundreds of hours | ISO 188 / ISO 527 | Yellowed and brittle | Stabilization system | Antioxidant (with copper tarnish inhibition) |
| Lubricating grease compatibility | No abnormalities in size or appearance after resin infiltration | Actual measurement of lipid soaking | Surface precipitation, stickiness | Confirm the lipid system in advance | Lubricant (migratory assessment included) |
| Warping and Key Dimensions | Key assembly dimensions are determined according to the conditioned state | ISO 1110 Humidity Control Measurement | Assembly clearance, abnormal noise | Drawing and Acceptance in Conditioned Humidity | - (Belongs to state management) |
How to read this table, first look at the first two rows.
Bending modulus and weld line strength are a pair, testing two opposite aspects of the same item: the body needs to be hard, and the seam needs to be tough.
The third and fourth lines are the positions of the inserts, and they are the two items in this article that should be checked item by item the most.
The problem with inserts is rarely the insert itself; most of the time it's whether 'the plastic side can hold up' — whether the clamping force is sufficient, whether the wall thickness is uniform, and whether there is stress concentration at the interface.
The fifth line has been exposed relatively frequently in the past two years, especially when the insert is a copper part.
When polyamide comes into contact with copper ions, the rate of thermal-oxidative aging will significantly accelerate, and the surface of the part will first turn yellow and then become brittle; this point should be raised during material selection and cannot wait until the client discovers the problem.
4. Four types of failures and their true causes
Failure 1: Cracking at the insert position, with cracks radiating outward from the insert.
The root cause of this type usually involves three things together: the difference in expansion between metal and plastic, uneven wall thickness around the insert, and injection molding internal stress.
The expansion difference can be calculated: the linear expansion coefficient of steel is about 11×10⁻⁶, while glass fiber reinforced nylon is about 30×10⁻⁶, nearly twice the difference.
The entire machine goes from a cold state of -20℃ to an operating temperature of 100℃, with a temperature difference of 120℃; based on a 10 millimeter diameter insert, the relative deformation is around 0.2%.
Two thousandths sounds very small, but multiplied by a modulus of 8 GPa, the circumferential stress is close to 20 MPa — this magnitude is enough to tear the part in areas with stress concentration.
So the effort for this type of problem needs to be spent on the structure: making the wall thickness around the inserts uniform, creating the fillets, and leaving some margin for assembly torque.
Failure Two: Some came out fine, some cracked, and neither the mold nor the inserts were changed.
This is the habit I most want to change in this article.
Many people think that deepening the knurling on the insert by one level will grip more tightly and therefore not crack.
The direction is reversed. The deeper the knurling, the more severe the stress concentration around the insert, and cracks often initiate earlier.
When the same set of molds and the same batch of inserts show differences between batches, the first things to check are the stability of the mold temperature and material temperature, whether drying is adequate, and whether the preheating temperature of the inserts is consistent.
These variables in the same workshop can differ by several degrees within a single day.
Failure 3: The weld line location is cracked, and the cracks follow the seam.
The strength on the weld line relies on resin healing; the fiberglass doesn’t help.
So, the weld line positions of porous parts and multi-gate parts depend on where they fall—if they fall on a stress-bearing surface or next to an insert, it’s a hidden risk.
The processing order is: first, position the gates to push away the weld lines, then improve mold temperature to enhance healing, and only after that consider changing the material.
Failure 4: The surface of the part turns yellow and brittle, and after prolonged use, it starts to powder.
This type needs to be checked in two directions.
One aspect is long-term thermal-oxidative aging: components near the motor are subjected to around 80°C for a long time, and if the stabilizing system is insufficient, they will yellow first and then become brittle.
Another aspect is copper damage: copper inserts can release copper ions under conditions of moisture and elevated temperature, catalyzing the degradation of polyamide. In this case, yellowing often concentrates around a circle surrounding the insert.
The solutions in these two directions are different, so you first need to look at the position of the yellow — whether it is the whole piece or just a part.
5. Processing and Verification: Drying, Mold Temperature, Inserts
Drying on nylon is always the first step.
Before going online, the dew point must be below -40℃, and the moisture content must be reduced to within 0.15%; after the moisture enters the feeding barrel, it will shorten the molecular chains at high temperatures, affecting both toughness and weld line strength.
The mold temperature has two functions on the gearbox housing.
First, it makes the surface dense and the weld lines heal well; second, it makes the shrinkage more uniform and the clamping force around the insert more consistent.
When the mold temperature is low, these two things deteriorate simultaneously, and the most common fluctuation in the workshop happens to be the mold temperature.
There are three actions that need to be done correctly in handling the insert.
First is preheating: if the insert is implanted cold, the surrounding plastic is locally quenched, shrinking unevenly, and stress remains there.
Second is positioning: The axial and radial positions of inserts must be fixed in the mold, and the plastic cannot be 'stuck' by itself.
Third is the rounded corners: the rounded corners on the plastic side are more useful than the knurling on the insert side, which is counterintuitive.
The humidity adjustment is determined according to the precision requirements of each piece.
For fitting dimensions such as insert positions and bearing positions, it is recommended to produce drawings and carry out inspections according to the humidity-adjusted state; purely aesthetic mating surfaces can be relaxed, but the state must be specified in the agreement.
It is recommended to arrange the verification sequence like this, do not change it:
1. Material Level: Flexural modulus, weld line sampling strength, thermal-oxidative retention rate at continuous temperatures
2. Process window: change mold temperature and material temperature, observe weld line appearance and insert position shrinkage
3. Item Level: Insert twisting and pull-out, key dimensions after humidity adjustment, appearance
4. Cyclic level: observe the insert interface after temperature cycling by slicing
5. Installed machine level: run according to actual working conditions, including stall conditions
Why can't the order be changed? Because the cracking at the insert position depends on both temperature and assembly torque. Since the state is not fixed, the numbers measured earlier are only valid for that batch.
6. Boundaries: For these types of structures, don't go down the fiberglass nylon path yet
This section may be more valuable than the previous few sections because it helps you cut losses before opening the mold.
First, the location where the locked-rotor temperature consistently exceeds 150°C. There is insufficient data supporting long-term performance retention at this level, so attention should be given to high-temperature nylon or metal components.
Secondly, the wall thickness around the insert cannot achieve a uniform thin-wall structure. The clamping force on the plastic side is provided by the structure; if the structure cannot provide it, changing the material cannot make up for it either.
Third, positions that require extremely high assembly torque and need to be repeatedly disassembled and assembled. Such positions should use metal threaded inserts or return to metal parts.
Fourth, places that are long-term immersed in strong solvents or strongly alkaline media. Nylon's long-term performance in such media lacks sufficient support.
Listing these four points upfront is not to discourage, but to save time — the gearbox housing mold is not cheap, and verification requires the whole machine to work together. If you choose wrong once, the cost of starting over is very high.
Material Change Risk List (Things that need to be addressed when switching from the original system to the glass fiber reinforced route)
| 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 dimensional chain needs to be rearranged. | Only replace materials without checking the mold |
| Dry | Replace the dehumidifying dryer and set the window according to the measured moisture content. | In the humid season, hot air drying is basically ineffective |
| Material Temperature / Mold Temperature | Improving the mold release temperature helps the weld line heal, do not copy the previous one | Low mold temperature causes weak seams |
| Gate and Weld Line | Rearrange the gate, pushing the weld line away from the stress and the insert position | The seam falls next to the insert |
| Insert | Preheat in place, fix the position, make rounded corners on the plastic side | Rely on deepening the knurling for strength |
| Humidity control | Coordinate dimensions according to the damp state for drawing and acceptance | Release according to dry-state dimensions |
| Verification order | Material level → Process window → Part level → Cycle level → Installed unit level | If the previous item fails, just move on. |
One-page report sheet (for people who need to report upwards)
| item | In short, a conclusion |
|---|
| Change what | From the motor side, check the temperature-resistant route; the output side can follow the cost route with general parts. |
| Move what | Gate rearrangement, mold temperature and material temperature redefinition, drying and dehumidification, insert preheating and rounding |
| Test what | Bending modulus, weld line sampling strength, insert torsion and pull-out, slicing after thermal cycling |
| When can the volume increase? | The weld line does not crack, the insert position has no cracks, and the lock-up condition runs smoothly. |
Three questions readers often ask
Question: The inserts keep cracking; is it because the inserts were chosen too small?
First, look at three things: whether the wall thickness around the insert is uniform, whether the mold temperature is stable, and whether the assembly torque has any margin. These three account for the most significant part; replacing with a larger insert often does not solve the problem and may even make the wall thinner.
Question: Can the strength of the weld line be improved by adding fiberglass?
It can't be fixed. The fiberglass aligns with the direction of flow, and the two flow directions don't match at the seam, so they can't intertwine across it; the strength of the seam relies on the resin to heal. Therefore, the skill in welding lies in the gate, mold temperature, and structure, not in the formula.
Question: Are copper inserts more prone to problems than steel inserts?
From the perspective of long-term aging, copper ions can catalyze the thermo-oxidative degradation of polyamide, which should be considered in advance. The usual approach is to upgrade the stabilization system while addressing moisture protection and edge sealing at the same time.
Conclusion
Returning to the first three questions: the temperature near the motor, the method of inserting the insert, and the assembly torque.
If these three things are all answered, the material route for the gearbox housing is basically decided.
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.
A couple of days ago, I received a phone call asking about the four characters 'here often cracks' on that blueprint.
The tricky part about inserting the part isn't the insert itself, it's the surrounding ring of plastic—wall thickness, mold temperature, and fillet radius. Once these three are set, whether it cracks or not is basically determined.
We produce modified nylon (PA6 / PA66 / PA46 / PA11 / PA12 / PA6T / PA9T and nylon alloys), and also manufacture modified PPO / PPS and thermoplastic elastomers; additionally, we regularly purchase nylon raw materials, sprue regrind, and various nylon waste, with formal disposal channels.
