去年秋天，一个做人形机器人本体的客户，带了一只大腿外壳的断件过来。
件是增韧 PA6 打的，撞在样机跌落测试的角铁上，裂了一道十厘米的口子。他把件搁在桌上，说："你们这料韧性不行，给我换个更韧的。"
我没接话，先问他三句：断的时候是常温还是低温？断口有缺口还是齐的？每一模都断还是偶尔断？
他愣了一下，说实验室常温没问题，是冬天样棚里跌的，零下十几度。
这料不是"不够韧"，是低温下那套增韧体系自己先脆了。 把"要韧性好"当成一句需求，是外壳选型最常见的坑。
人形机器人外壳这件改性尼龙，最该算的不是"韧不韧"，是两笔账。
一、人形机器人外壳，两笔账要分开算
外壳在机器人身上，是离"人"最近的那层皮。它不传力，但要同时扛两件事：
第一笔账，轻量化。 整机每减一公斤自重，末端能多带走接近一公斤的负载——这是行业里反复被引用的经验数。外壳占了整机不小的重量比例，换成塑料是减重的大头。
第二笔账，接触安全。 人和机器人同空间作业，外壳碰人、撞物、跌地，都不能出尖角、不能碎成片、不能闷出异味。这一笔，金属反而麻烦（棱角、冰冷、重）。
这两笔账方向相反：减重要薄，安全要厚；轻要低刚性，安全要抗冲击。 外壳选料，就是在这两头找交点。
这两笔账还牵出第三件要算的：装配与维修成本。
外壳要能拆、能换、能复位，卡扣和螺纹柱的寿命都得跟着壳子一起算。
一个装一次就松的壳，减重减得再多也留不住客户。
二、壳体被什么拽着：六个面
壳子的工况，比一般外观件多一层"与人共处"的约束。
温度。 体内电机、电池贴着壳内壁，局部能到 70–90℃。壳子要扛长期热，又不能闷死热。
冲击。 跌落、碰撞、搬运磕碰，都要算。协作场景里这一项比工业臂重得多。
介质。 手部油脂、清洁剂、汗液，都对表面有要求。
气味与析出。 和人同空间，低气味、低挥发是硬要求，涉及食品医疗周边要回 GB 4806.7 / FDA 语境。
寿命。 启停与磕碰以多年计，判据是"到寿命时外观和安全还行不行"。
外观。 露脸的壳，浮纤、色差、缩痕都要控，编织纹理这类处理还要好看。
六个面里，冲击和气味是硬门槛，剩下四个决定方案能不能收口。
三、轻量化与接触安全的两套材料
从金属换塑料，外壳的逻辑和臂杆、关节壳都不同——它不太要刚性，要的是轻、韧、安全、好看。
| 路线 | 组成 | 给什么 | 代价 |
|---|
| 增韧 PA6 | 弹性体增韧 | 轻、抗冲击、低气味、好喷涂 | 刚性低、耐温一般 |
| PA6-GF30 | 玻纤 + 增韧 | 尺寸稳、略刚、抗蠕变 | 表面浮纤、要控模温 |
| PA/ABS | 合金 | 外观好、易着色、手感佳 | 耐温偏低、改性复杂 |
三条没有"谁更好"，只有"这个壳子离人多近"。
增韧 PA6 这一行，是露脸壳的主力。轻、抗冲击、气味低，做协作机器人外壳比 PA66 好看也安全。代价是刚性和耐温一般，内部支撑要用得谨慎。
PA6-GF30 这一行，尺寸更稳、略刚，适合既要包住内部件又要一点支撑的壳。代价是表面浮纤，模温工艺要管住。
PA/ABS 这一行，外观和手感最顺，但耐温偏低，靠近电机的壳要避开。
一句话：外壳选料，先问"离人多近、撞什么"，再问牌号。离人近看安全，贴电机看耐温。
四、外壳选材判据表
把上面的约束落成可核对的指标。下表门限是方向性建议，不是验收标准——实际数值由具体项目、工况和实测定。
| 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 |
|---|
| 密度 / 减重 | 相对铝约减半 | 称重 + 整机核算 | 减重不达标 | 低玻纤或纯韧体系 | 助剂不担减重 |
| 落球冲击（常温） | 按工况定门限 | ISO 6603 | 凹陷、开裂 | 弹性体增韧 | 增韧剂（界面相容） |
| 低温冲击（-20℃） | 按工况定门限 | ISO 179（-20℃） | 低温脆断 | 核壳结构增韧 | 增韧剂（低温型） |
| 总挥发 / 气味 | 低气味等级 | VDA 270 / 内部法 | 闷味、投诉 | 低挥发基材 | 助剂迁移需评估 |
| 表面浮纤 | 目视 / 粗糙度达标 | 客户外观标准 | 喷漆挂不住 | 提模温 + 多级射胶 | 润滑剂（改善包覆） |
| 阻燃（如需） | 按场景定等级 | UL94 / GB 标准 | 过不了安规 | 无卤阻燃体系 | 阻燃剂（协效） |
| 表面硬度 / 耐刮 | 按使用场景定 | 铅笔硬度 / 刮擦 | 划伤、发亮 | 表面处理或低玻纤 | 助剂不担此责 |
| 装配刚性 | 按壳体支撑要求定 | 装配后测形变 | 止口松、异响 | 局部加强筋 | 助剂不担此责 |
怎么用这张表：别逐行打分，先看落球冲击和低温冲击两行。常温韧、低温脆的料，冬天样棚一跌就露馅。
一个提醒：表里"低温冲击"这一项，很多项目只测常温。和人共处的外壳，低温跌落是真实工况，不测就是埋雷。
五、外壳的五个典型失效
失效一：常温不裂，低温一跌就断。
根因是增韧体系类型选错。普通弹性体增韧在低温会失效，要核壳结构才管用。把"要韧性好"当成一句需求，得到的答案大概率是错的。 断在常温还是低温，指向两套完全不同的方案。
失效二：跌落碎成片，尖角伤人。
根因是料太脆或壁太薄。外壳要的是"裂而不碎、碎不露尖"，靠增韧和壁厚一起保。薄到一定程度，再韧的料也挡不住尖角。
失效三：表面发白喷不上漆。
根因多半是模温低冻住玻纤（同关节壳）。不是换低玻纤料，是提模温、改多级射胶。
失效四：同一批件黄得深浅不一。
不是"料不稳定"，是抗氧剂分散不均。先查混料与母粒化，别急着换料。
韧性问错的那一句：客户说"要韧性好"，九成是因为他断过件。但追问一句就能分开：断在常温还是低温？断口有没有缺口？这两个答案指向完全不同的两套方案。低温脆断的件，用普通弹性体增韧等于没改——低温下它自己先脆了。
六、壁厚、纹理与跌落穿刺怎么验
壁厚。 外壳壁厚定均匀，薄壁（1.5–2.5 mm）要料流动性好、模温足；太薄扛不住跌落，太厚缩痕翘曲。壁厚是减重和安全的真正交点。
编织纹理。 协作机器人外壳常做编织尼龙纹理，既好看又藏浮纤。纹理是模具蚀纹做出来的，和材料无关，但纹理深处的排气要足，否则困气发白。
跌落与穿刺。 样机阶段要真跌、真撞，不是算。角跌落、棱跌落、面跌落分开测，低温工况单独测。穿刺（尖物顶）比平面跌落更狠，靠近关节的壳要重点验。
清洁与消毒。 和人同空间的外壳，会被反复擦拭。
酒精、含氯清洁剂、含表面活性剂的湿巾，都可能让壳面发白、发黏。
这一条要在选料时先问清客户用什么擦，不能等量产再补测。
尺寸与孔位。 壳子装在关节模组外面，孔位要跟内部件对得上。
调湿前后各测一次，这两次的差比绝对值更能说明问题。
差值偏大的件，装配环境的湿度要写进技术协议。
维修与更换。 外壳是整机上最常被拆下来的一块，螺丝柱要经得起反复拆装。
设计时把拆装次数写进要求，按次数做循环验证。
很多外壳不是用坏的，是修坏的。
验证顺序。 建议这样排：
1. 尺寸与外观（调湿后测）
2. 落球冲击（常温 + 低温）
3. 跌落 / 穿刺（角、棱、面 + 低温）
4. 气味与挥发
5. 环境叠加（温湿循环 + 装配）
顺序不能换。 前一项不通过就往下走，后面数据没有解释意义。
打样实录：有客户只测了常温落球就量产，冬天样棚一跌裂十厘米。我们把低温冲击加进打样确认单：先让客户把"跌在常温还是低温"说清，再定增韧体系。 普通增韧换核壳结构，同一壁厚，低温跌落从裂变成凹。
七、边界：外壳换塑料的红线
这一段可能比前面六段更值钱。
以下四种情况，人形机器人外壳走尼龙这条路要谨慎或暂缓：
其一，长期贴电机且内壁超过 130℃。 PA6 体系在这个区间长期保持数据支撑不足，要回 PA6T / PA9T 或金属。
其二，要求气密或液密的结构壳。 塑料壳焊接和密封比金属难，这类件先验证密封工艺。
其三，高刚性承力外骨架。 外壳不是骨架，要刚要承力的位置，回到臂杆或金属，别让皮扛梁的活。
其四，壁厚被压到极低又要过穿刺。 太薄再韧也挡不住尖角，这类件要先做工艺可行性，不能只靠换料。
把这四条写在前头，不是劝退，是省时间。 样品阶段一路顺、最后卡在低温跌落或密封上、整个方案回退的项目，我见过不止一个。
换料风险清单（从金属换到人形机器人外壳，要动的东西）
| 环节 | 要动什么 | 容易漏的点 |
|---|
| 模具 | 壁厚按减重与安全重定，蚀纹排气重做 | 编织纹理深处困气 |
| 干燥 | 按实测含水率定窗口 | 回用料掺入带入水分 |
| 料温 / 模温 | 按表面质量定模温（常 110℃+） | 只按牌号推荐值给 |
| 保压 / 脱模 | 薄壁壳保压曲线重定 | 缩痕、变形 |
| 调湿 | 强制调湿 + 称重 + 复测 | 按平均壁厚估时间 |
| 色差 | 外观件色板提前确认 | 高玻纤件本身发暗 |
| 验证顺序 | 尺寸 → 冲击 → 跌落 → 气味 → 环境 | 前一项未过就往下走 |
打样试模排程
| 轮次 | 上机内容 | 每轮验什么 | 留样 |
|---|
| 第一轮 | 螺杆试小样，短射看填充 | 薄壁是否打满、浮纤 | 留 3 件，3 个月 |
| 第二轮 | 正式试模，调模温保压 | 落球冲击（常温） | 留 5 件，6 个月 |
| 第三轮 | 低温冲击 + 跌落穿刺 | 落球（低温）、穿刺 | 留 8 件，12 个月 |
一页纸汇报表（给要向上汇报的人）
`
项目：人形机器人外壳 · 尼龙壳体路线评估
结论方向：可作为候选路线，能否落地取决于三项前置条件
一、必须守住的三条
1. 常温与低温冲击都测，不只测常温
2. 壁厚按减重与安全交点定，不盲目压薄
3. 气味 / 挥发按与人共处标准验
二、前置条件（任一不满足则建议暂缓）
· 内壁长期温度 ≤ 130℃ 量级
· 不要求气密 / 液密结构
· 非高刚性承力外骨架
· 壁厚工艺可行
三、下一步动作
1. 取金属壳，测壁厚与止口
2. 定增韧体系（常温 / 低温）
3. 调湿前后 + 低温跌落复测
风险提示：本路线主要不确定性在低温冲击与气味，不在初始强度。
`
读者常问的三句
问：和进口料差在哪？
只讲两件能对照的事：同一指标，看它标没标测试条件；同一件上，看它给没给长期数据。外壳类指标对低温和状态敏感，条件不明的数字不宜直接比。有些件上走国产路线已经比较成熟，有些件目前仍不建议替——具体到你的壳，要看离人多近、撞什么两样。
问：能不能一体成型省装配？
能，但要先付模具和工艺代价。一体成型的壳，浇口、排气、蚀纹都比分件难，验证要重做。先算装配省下的钱够不够覆盖模具增量，再决定。
问：外壳减重到底能减多少？
纵向比，相对铝件常见是减一半的量级；具体到这个壳，要看壁厚、加强筋和玻纤含量。
横向比，别拿别家报的壳体克重直接对，件形不一样，数没法比。
减重是整机算出来的，不是单件称出来的。
结语
人形机器人外壳的塑化，说到底是一道轻与安全的平衡题，不是强度题。
判断链只有三条：
温度定基材 → 壁厚定安全 → 低温定增韧。
三条都定完，"能不能用塑料"这个问题自然就有答案了。
如果你手上正有一个机器人外壳要定料，把三样东西发过来就能给方向：内壁温度、是否与人共处、跌在常温还是低温。
补一句给做方案的人：外壳是最容易被当成"配角"的件。
但它同时压着减重、安全、气味三件事，哪一件没算清，量产都会回头找你。
还有一句：外壳的验证别只看单件。
单件过不了的件，装配后更过不了；单件过了的件，也建议整机再复测一次。
很多问题，只有装上去才看得见。
改性尼龙的粒子，出厂时只是一颗粒子。
它变成外壳、关节壳、支架，中间隔着一整套方案——增强多少、耐温做到哪一档、表面怎么处理、尺寸稳不稳。
我们做的事很具体：把 PA6、PA66、PA46、PA11、PA12、PA6T、PA9T 和尼龙合金这些树脂，改成某个件真正能用的样子；顺带做改性 PPO、PPS 和热塑性弹性体；也经营各大化工巨头的尼龙树脂、副牌料和大包料，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。
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。
Last fall, a client who makes humanoid robot bodies brought over a broken outer thigh shell.
The part was made of toughened PA6. It hit an angle iron during the prototype drop test and cracked a ten-centimeter-long gap. He put the part on the table and said, 'The toughness of this material isn't good enough; give me a tougher one.'
I didn't respond and first asked him three questions: Was it broken at room temperature or at low temperature? Did the fracture have a notch or was it smooth? Does it break every time in each mold or only occasionally?
He was stunned for a moment and said that the laboratory's normal temperature was fine; it fell in the sample shed in winter, at minus ten-something degrees.
This material is not 'not tough enough'; it's that the toughening system itself becomes brittle at low temperatures. Treating 'needs to be tough' as a requirement is the most common pitfall in selecting an enclosure material.
For the humanoid robot shell, this modified nylon should not be judged on 'toughness' but rather on two calculations.
1. Humanoid robot casing, the two accounts need to be calculated separately
The outer shell on a robot is the layer of 'skin' closest to a human. It does not transmit force, but it has to handle two things at the same time:
The first account: light weighting. For every kilogram reduced from the total weight of the machine, the end can carry nearly one more kilogram of load — this is an experience repeatedly cited in the industry. The casing accounts for a significant proportion of the total weight of the machine, and replacing it with plastic is the main source of weight reduction.
The second point is contact safety. When humans and robots work in the same space, the robot's exterior must not have sharp corners, break into pieces, or emit odors when it touches people, hits objects, or falls to the ground. In this regard, metal is actually troublesome (edges, coldness, weight).
These two accounts go in opposite directions: reduce weight but make it thin, safety should be thick; being light requires low rigidity, safety needs impact resistance. Selecting materials for the casing is about finding the intersection between these two ends.
These two accounts also lead to a third item that needs to be calculated: assembly and maintenance costs.
The casing must be detachable, replaceable, and resettable, and the lifespan of the clips and threaded posts must be counted along with the casing.
A case that becomes loose after just one installation, no matter how much weight is reduced, cannot retain customers.
2. What is the casing being pulled by: six faces
The operating conditions of the shell have one more layer of 'being with others' constraints than ordinary exterior parts.
Temperature. The internal motor and battery are close to the inner wall of the case, and locally can reach 70–90°C. The case must withstand prolonged heat, but it also can't trap all the heat inside.
Impact. Falls, collisions, and bumps during handling all count. In collaborative scenarios, this item is much heavier than industrial arms.
Medium. Hand oils, cleaning agents, and sweat all have requirements for the surface.
Odor and emission. When sharing space with people, low odor and low volatility are strict requirements, relating to food and medical contexts that need to comply with GB 4806.7 / FDA regulations.
Service life. Start-stop operations and bumps are counted over many years, and the criterion is whether the appearance and safety are still acceptable when it reaches the end of its service life.
Appearance. For exposed cases, issues like floating fibers, color differences, and shrink marks must be controlled, and woven textures and similar treatments also need to look good.
Among the six aspects, impact and smell are hard thresholds, while the remaining four determine whether the plan can be finalized.
3. Two sets of materials for lightweighting and contact safety
Switching from metal to plastic, the logic for the casing is different from that of the arm and joint housings—it doesn't need much rigidity, what it requires is lightness, toughness, safety, and a good appearance.
| Route | compose | Give what | Cost |
|---|
| Toughened PA6 | Elastomer toughening | Lightweight, impact-resistant, low odor, easy to spray | Low rigidity, average temperature resistance |
| PA6-GF30 | Glass fiber Toughening | Stable in size, slightly stiff, creep-resistant | Surface floating fibers, need to control mold temperature |
| PA/ABS | Alloy | Good appearance, easy to color, pleasant to touch | Low temperature resistance, complex modification |
None of the three asks 'who is better,' only 'how close this shell is to people.'
The toughened PA6 line is the main choice for show casing housings. It is light, impact-resistant, and low in odor, making it better looking and safer than PA66 for collaborative robot shells. The downside is that its rigidity and heat resistance are average, so internal supports need to be used carefully.
For the PA6-GF30 line, the dimensions are more stable and slightly stiffer, suitable for enclosures that need to cover internal components while providing some support. The trade-off is surface fiber bloom, so the mold temperature process needs to be controlled.
In the PA/ABS line, the appearance and feel are the smoothest, but the temperature resistance is relatively low, so the housings near the motor should be avoided.
In one sentence: When choosing the material for the shell, first ask 'How close will it be to people, and what will it collide with?' then ask about the grade. If it's close to people, check safety; if it's next to a motor, check temperature resistance.
4. Criteria Table for Shell Material Selection
Turn the above constraints into verifiable indicators. The thresholds in the table below are directional recommendations, not acceptance standards — the actual values are determined by specific projects, operating conditions, and actual measurements.
| Indicator | Directional threshold | Verification Method / Standard | Common Failures | Common solution | Corresponding auxiliary agent system |
|---|
| Density / Weight Reduction | Reduced by about half compared to aluminum | Weighing Whole machine accounting | Weight loss target not met | Low glass fiber or pure tough system | The additive does not account for weight reduction |
| Ball Impact (Room Temperature) | Set threshold according to operating conditions | ISO 6603 | Depression, cracking | Elastomer toughening | Toughening agent (interface compatible) |
| Low-temperature shock (-20°C) | Set threshold according to operating conditions | ISO 179 (-20°C) | Brittle fracture at low temperature | Core-shell structure toughening | Toughening Agent (Low Temperature Type) |
| Total Volatile / Odor | Low odor level | VDA 270 / Internal Method | Stuffy smell, complaint | Low-volatility substrate | The migration of additives needs to be assessed |
| Surface floating fibers | Visual / Roughness Standard Met | Customer appearance standards | The spray paint won't stick | Mold Temperature Multi-stage Injection Molding | Lubricant (improves coating) |
| Flame retardant (if needed) | Set the level according to the scenario | UL94 / GB Standard | Cannot pass safety regulations | Halogen-free flame retardant system | Flame Retardant (Synergist) |
| Surface Hardness / Scratch Resistance | Decided according to the usage scenario | Pencil Hardness / Scratch | Scratch, shiny | Surface treatment or low glass fiber | The additive does not bear this responsibility |
| Assembly rigidity | Determine according to the casing support requirements | Measure deformation after assembly | Loose closure, abnormal noise | Local reinforcement rib | The additive does not bear this responsibility |
How to use this table: Don't score row by row. First, look at the rows for drop impact and low-temperature impact. Materials that are tough at room temperature but brittle at low temperature will fail in winter when samples are dropped in the shed.
A reminder: For the 'low-temperature shock' item inside and outside, many projects only test at normal temperature. For cases that come into contact with people, low-temperature drops are a real working condition; not testing is like planting a landmine.
5. Five Typical Failures of the Casing
Failure 1: Does not crack at room temperature, but breaks with one fall at low temperature.
The root cause is choosing the wrong type of toughening system. Ordinary elastomer toughening will fail at low temperatures; only a core-shell structure works. Treating 'needs to be tough' as a requirement will most likely lead to the wrong answer. Fracturing at room temperature or low temperature points to two completely different solutions.
Failure 2: Breaks into pieces when dropped, sharp edges can cause injury.
The root cause is that the material is too brittle or the wall is too thin. The shell requires 'cracks without breaking, breaking without exposing sharp points,' which relies on both toughening and wall thickness for protection. Once it is thin to a certain extent, no matter how tough the material is, it cannot prevent sharp corners.
Failure 3: The surface turns white and paint cannot be applied.
The root cause is mostly that the mold temperature is too low, causing the fiberglass (same joint shell) to freeze. It's not about switching to lower fiberglass material, but raising the mold temperature and modifying to multi-stage injection.
Failure 4: The same batch of items has inconsistent yellowing depth.
It's not that the material is unstable, it's that the antioxidant is unevenly dispersed. First, check the mixing and masterbatching; don't rush to change the material.
The sentence often asked wrong about toughness: When a customer says 'needs good toughness,' nine times out of ten it's because they have had broken parts. But just asking one follow-up question can clarify: Did they break at room temperature or low temperature? Are there any notches at the fracture? These two answers point to completely different sets of solutions. For parts that are brittle at low temperature, toughening with ordinary elastomers is equivalent to no modification—it will break itself first at low temperature.
6. How to inspect wall thickness, texture, and drop puncture
Wall thickness. The outer shell wall thickness should be uniform. Thin walls (1.5–2.5 mm) require good material flow and adequate mold temperature; if too thin, they cannot withstand drops, and if too thick, they cause sink marks and warping. Wall thickness is the real intersection of weight reduction and safety.
Woven texture. Collaborative robot housings often feature woven nylon textures, which are both attractive and conceal floating fibers. The texture is created by etching the mold and has nothing to do with the material, but there needs to be sufficient venting in the deep parts of the texture, otherwise trapped air can cause whitening.
Drop and puncture. In the prototype stage, drops and impacts need to be real, not calculated. Angle drops, edge drops, and flat drops should be tested separately, and low-temperature conditions should be tested separately. Puncture (pointed object impact) is more severe than flat drops, and shells near the joints should be a key focus for testing.
Cleaning and disinfection. The outer shell that shares space with people will be wiped repeatedly.
Alcohol, chlorine-containing cleaners, and wipes with surfactants may cause the shell surface to turn white and sticky.
For this item, you need to ask the customer what they will use for wiping when selecting materials; you can't wait until mass production to do additional testing.
Size and hole positions. The casing is placed outside the joint module, and the hole positions need to align with the internal components.
Measure once before and after humidity adjustment; the difference between these two measurements is more indicative than the absolute values.
For parts with large deviations, the humidity of the assembly environment must be specified in the technical agreement.
Maintenance and replacement. The casing is the part of the whole machine that is most often removed, and the screw posts must withstand repeated disassembly and assembly.
When designing, include the number of assembly and disassembly times in the requirements, and perform cyclic verification according to the number of times.
Many shells are not damaged from use, they are damaged from repairs.
Verification order. It is recommended to arrange it like this:
1. Size and appearance (measured after humidity adjustment)
2. Ball Drop Impact (Normal Temperature Low Temperature)
3. Drop / Puncture (corners, edges, surfaces, low temperature)
4. Smell and Volatility
5. Environmental Stacking (Temperature and Humidity Cycling Assembly)
The order cannot be changed. If the previous item fails, just move on; the subsequent data has no explanatory meaning.
Sample Testing Record: Some customers only tested the room-temperature drop ball and went into mass production, and in winter the sample shelter would crack ten centimeters with one fall. We added low-temperature impact to the sample confirmation form: first, let the customer clarify whether the drop is at room temperature or low temperature, and then determine the toughening system. Ordinary toughening is replaced with a core-shell structure; with the same wall thickness, low-temperature drops change from cracking to denting.
7. Boundary: Red line for the shell replaced with plastic
This section might be more valuable than the previous six sections.
In the following four situations, humanoid robot shells should be cautious or delay taking the nylon route:
First, if the motor is in continuous contact and the inner wall exceeds 130℃. The PA6 system cannot maintain sufficient data support over this range for long-term use, so one should revert to PA6T / PA9T or metal.
Secondly, structures that require airtight or watertight enclosures. It is more difficult to weld and seal plastic housings than metal ones, so the sealing process for these parts should be verified first.
Thirdly, a high-rigidity load-bearing exoskeleton. The shell is not the framework; for places that need rigidity and load-bearing capacity, return to using arm rods or metal, and don't let the skin bear the beam's work.
Fourth, the wall thickness is pressed to an extremely low level and must withstand puncture. Even if it is tough, if it is too thin it cannot block sharp corners. For this type of part, process feasibility must be assessed first, and it cannot rely solely on changing the material.
Writing these four points at the beginning is not to discourage, but to save time. I have seen more than one project where everything went smoothly during the sample stage, only to get stuck at low-temperature drop tests or sealing issues, causing the entire plan to roll back.
Material Change Risk List (From metal to humanoid robot casing, things that need to be modified)
| link | What needs to be moved? | Points that are easy to overlook |
|---|
| Mold | Wall thickness is readjusted based on weight reduction and safety, and pitting venting is redone | Trapped energy deep within the woven texture |
| Dry | Determine the window based on the measured moisture content | Incorporating recycled materials introduces moisture |
| Material Temperature / Mold Temperature | Set mold temperature based on surface quality (usually 110℃) | Give only according to the recommended value by grade |
| Pressure Holding / Demolding | Thin-walled shell holding pressure curve redefinition | Dents and deformation |
| Humidity control | Forced humidification Weighing Retest | Estimate time based on average wall thickness |
| Color difference | Advance confirmation of exterior color samples | The high glass fiber parts themselves appear dark |
| Verification order | Size → Impact → Drop → Odor → Environment | If the previous item fails, just move on. |
Sample Printing and Trial Molding Scheduling
| Round | Computer-based content | What is checked each round? | sample retention |
|---|
| First round | Screw trial small sample, check filling for short shot | Whether the thin wall is fully filled, floating fibers | Keep 3 items for 3 months |
| Second round | Official mold trial, adjust mold temperature and hold pressure | Drop Ball Impact (Room Temperature) | Keep 5 items, 6 months |
| Round Three | Low-temperature impact + drop puncture | Drop ball (low temperature), puncture | 8 pieces, 12 months |
One-page report form (for those reporting upward)
'
Project: Humanoid robot shell · nylon shell route evaluation
Conclusion direction: Can be used as a candidate route, Whether it can be implemented depends on three prerequisites
1. Three essential conditions to be maintained
1. Both normal and low-temperature impacts are tested, not just normal temperature
2. Wall thickness should be determined at the intersection of weight reduction and safety, not blindly thinned
3. Odor/evaporation tested according to cohabitation standards
2. Prerequisites (if either is not met, postponement is recommended)
· Long-term inner wall temperature ≤ 130°C Order
· No airtight / liquid-tight structure required
· Non-high-rigidity load-bearing external frame
· Wall thickness process feasible
III. Next steps
1. Take the metal shell and measure wall thickness and stop
2. Fixed toughening system (room temperature / low temperature)
3. Before and after humidity regulation + Low-temperature drop retest
Risk Warning: The main uncertainties of this route lie in low-temperature shock and odor, not in initial intensity.
'
Three Questions Readers Often Ask
Question: What's the difference from imported materials?
Let's talk about two things you can compare: For the same indicator, check if it marks the test conditions; For the same item, check if it provides long-term data. Shell indicators are sensitive to low temperatures and conditions; if the conditions are unclear, don't compare directly. Some parts have already adopted a domestic route, but some are still not recommended for replacement—specifically, it depends on how close you are to people and what you bump into.
Question: Can it be molded in one piece to save assembly?
Yes, but you have to pay for molds and craftsmanship first. For one-piece shells, gates, vents, and etching are all harder than part-splitting, so verification requires rework. First, calculate whether the cost saved from assembly is enough to cover the mold's increment, then decide.
Question: How much weight reduction can the shell actually be ?
Longitudinal comparison: compared to aluminum parts, it's usually reduced by about half; For this shell, you need to look at wall thickness, reinforcement ribs, and fiberglass content.
Horizontal comparison: don't directly compare the weight of other brands' shells—the shape is different, so you can't compare numbers.
Weight reduction is calculated by the whole machine, not by individual pieces.
Conclusion
The plasticization of humanoid robot shells is, at its core, a balancing issue between lightness and safety, not strength.
There are only three judgment chains:
Temperature determines substrate → wall thickness determines safety, → low-temperature flexible intensification.
Once all three are set, the question of "can plastic be used?" naturally comes to an answer.
If you have a robot shell to be fixed, send over three things and you can give directions: inner wall temperature, whether it coexists with people, and whether it falls at normal temperature or low temperature.
Add a note to the solution maker: the shell is the most easily treated as a "supporting role."
But it also weighs on weight reduction, safety, and odor. If you don't calculate any of them, mass production will come back to you.
There's another saying: don't just look at individual pieces when verifying the casing.
Parts that don't pass a single piece will fail even less after assembly; If a single piece fails, it's also recommended to retest the whole machine.
Many problems are only visible once installed.
Modified nylon particles are just a single particle at the factory.
It becomes the shell, joint shell, and bracket, with a whole set of solutions in between—how much to reinforce, what temperature resistance level, how to treat the surface, and whether the dimensions are stable.
What we do is very specific: converting resins like PA6, PA66, PA46, PA11, PA12, PA6T, PA9T, and nylon alloys into a truly usable form; We also produce modified PPO, PPS, and thermoplastic elastomers; We also handle nylon resins, sub-brand materials, and bulk materials from major chemical giants, and we also collect nylon raw materials, sprue returns, and various nylon waste materials for a long time, with official disposal channels.
The additive system in the formula is tailored to the working conditions of each piece—regular additives are always in stock, special models are matched as needed; You report the working conditions and grade, and the materials and additives are all prepared in one go
