去年七月，一个做园林工具的客户寄了两个件过来。
黑色的玻纤增强件，手掌大小，是手持式割草机齿轮箱的上盖。他寄过来的时候用了个牛皮纸信封，没有泡沫垫，两个断件在里面互相撞出声响。
电话里他说得很直接："你们这料不行吧，用了半年就裂。"
我把两个件放在桌上摆开，先看的不是裂纹，是断口。
断口很干净，平齐、发白、边缘没有任何拉丝和韧窝。这种断口在行业里有个朴素的说法：叫脆断，不叫拉断。拉断是有先兆的——变形、发白、慢慢撕开；脆断是一瞬间的事，昨天还好好的，今天一碰就掉一块。
然后我问了他三句话。
第一句：断在常温还是低温？
他说常温，七月的广东，车间里三十七八度。
第二句：断口是齐的还是不齐的？
我让他摸一摸——齐的，像刀切。
第三句：是每一模都断，还是偶尔断？
他愣了一下，说："用的时候不都一样吗？怎么会偶尔断……"
我说不一样。这句话先放下，我们到第六节再回来。
后来弄清楚的事情是这样：料没问题，牌号也对，RTI 兑现了，GF30 的力学也兑现了。是工况超了。
那个齿轮箱盖的位置紧贴缸体后端，实际长期工作温度在 130℃ 上下。他很早就说过这个位置"挺烫的"，但没人把"挺烫的"换算成一个具体的数字——而 PA6 的长期连续使用温度，一般写在 100-120℃。
差的那十度，不是差一点，是差一整个材料体系。
这篇文章想讲的就是这件事：PA6 的性能边界在哪里，怎么判断你有没有踩过去，踩过去以后往哪儿走。
一、先看清 PA6 的真实底细
PA6 的基本盘，六个数字记住就够了：
| 参数 | 典型值 | 说明 |
|---|
| 熔点 | 220℃ | 比 PA66 低 45℃ |
| 平衡吸水率 | 8-10% | 全尼龙家族里偏高 |
| 纯树脂拉伸强度 | 70-80MPa | 干态 |
| GF30 后拉伸强度 | 160-180MPa | 工装结构件的主力水平 |
| 长期连续使用温度 | 100-120℃ | 加耐热体系可到 140-150℃ |
| 收缩率 | 1.5-2.0%（纯）／0.3-0.7%（GF30） | 玻纤含量影响极大 |
数字列完，接下来才是重要的部分：这些数字在你的车间里意味着什么。
为什么 PA6 最便宜
原因不在改性厂，在上游单体。
PA6 的单体是己内酰胺，PA66 的单体是己二胺加己二酸。己内酰胺这条产业链在国内已经跑了三十多年，产能集中、工艺成熟、单套装置的规模可以做得非常大。规模一大，单位成本就下来了，这是化工行业最朴素的规律。
所以你问"PA6 为什么便宜"，答案不太浪漫：是因为它做得早、做得大、竞争者多。
8-10% 的吸水率到底是什么概念
这是最容易被读成"一个属性"的数字。我们把它换算一下。
一件 200 克的 PA6 制品，扔在南方的梅雨天里吸到平衡，它能吸进去 16 到 20 克的水——差不多是一汤匙。
这一汤匙水进去，会发生三件事：
第一，尺寸会涨。 PA6 吸水后体积膨胀，量级在千分之几。听起来不大，但一个 100 毫米长的配合孔，从 100.00 涨到 100.30，在过盈配合上就是从"刚好"变成"卡死"。
第二，强度会掉。 水分子进入分子链之间，起到了增塑的作用——通俗说就是把本来紧密抱在一起的分子链"泡松了"。干态 80MPa 的纯 PA6，吸到饱和能掉到 45-55MPa，这不是理论值，是把试样泡在水里几天实测出来的量级。
第三，它还会回来。 到了冬天干燥起来，水又走了，件又缩回去。这一涨一缩， 就在配合面上来回蹭，是有次数上限的。
所以南方客户尤其要注意：你在塑件图纸上标的公差，是在什么含水率下标出来的？大部分人说不出来。
"PA6 韧性最好"这句话要打个折
教科书的说法是：PA6 韧性优于 PA66，这是因为它分子链上的酰胺基密度低一些（每隔 6 个碳一个），分子间氢键少了些，链段活动余地大。
这句话本身没错，但它在三种情况下会失效：
加了玻纤之后。 GF30 的 PA6 和 GF30 的 PA66，韧性差距会被大幅抹平——真正决定韧性的是玻纤含量和增韧体系，不是基材
在低温。 常温下的"韧"，到 -20℃ 可能完全不存在
在有缺口的地方。 尖角、壁厚突变、熔接线，这些地方 PA6 一样脆，而且毫无预警
一句话：如果工况允许，PA6 就是性价比答案。问题在于——很多人判断不出"工况是否允许"。
二、PA6 还是 PA66？一张表定
这是选 PA6 时真正要回答的问题。看这张对比：
| 维度 | PA6 | PA66 | 谁更优 |
|---|
| 熔点 | 220℃ | 265℃ | PA66 |
| 长期连续耐温 | 100-120℃ | 120-140℃ | PA66 |
| 短期峰值耐受 | 约 150℃ | 约 180-200℃ | PA66 |
| 平衡吸水率 | 8-10% | 8-9% | PA66 略优 |
| 韧性（缺口冲击） | 更高 | 略低 | PA6 |
| 刚性和强度 | 略低 | 更高 | PA66 |
| 耐油、耐燃油 | 一般 | 更好 | PA66 |
| 表面质量（浮纤倾向） | 较好 | 较差 | PA6 |
| 加工窗口 | 宽，好加工 | 窄，料温高 | PA6 |
| 相对价格 | ★ | ★★ | PA6 |
那个 45℃ 的熔点差是从哪儿来的
表上第一行，220 对 265，差了 45℃。这个差值不是调配出来的，是分子结构决定的，改不动。
PA6 的分子链上，每隔 5 个亚甲基出现一个酰胺基；PA66 是每隔 4 个（己二胺段）和 4 个（己二酸段）交替。酰胺基是形成分子间氢键的地方，氢键密度越高，链与链之间锁得越紧，要把它拉开就需要更多能量——宏观表现就是熔点更高、耐热更好。
PA66 的酰胺基密度比 PA6 高，所以它的氢键网络更密。这不是"加个耐热剂能补回来"的事，耐热剂的作用是延缓老化，不是抬高天花板。
这就是为什么开篇那个客户的件会断。他选的是 PA6，位置在 130℃。PA6 加耐热体系能推到的上限大概 140-150℃，但那是"能撑住不立刻坏"，不是"能长期稳定"。
长期的账，是要按年算的。
必须换 PA66 的五种情况
出现下面任意一条，不要用 PA6 硬撑：
① 长期连续使用温度 >120℃
发动机舱件、灯座、靠近热源的件，直接用 PA66。这条线没有例外，也不建议靠耐热母粒去"试试看"。
② 需要更高的刚性和强度
同样 GF30，PA66 的强度比 PA6 高约 10-15%。如果结构已经到极限、不能再加玻纤，换 PA66 是一个不动模具就能提性能的办法。
③ 长期接触机油、燃油
PA66 的耐油性明显优于 PA6。发动机油底壳、油路附近的件，优先 PA66。
④ 尺寸精度要求高
PA66 结晶度高、吸水率略低，尺寸稳定性优于 PA6。公差 ±0.05mm 级的件要慎重。
⑤ 有电气安规要求
断路器、接触器、高压件，PA66 的耐热和电性能基线更好，更容易做到高 GWIT 和 CTI。
继续用 PA6 的四种情况
① 长期 ≤120℃ 的通用结构件——这是 PA6 的主场，换 PA66 是浪费。
② 需要高韧性的抗冲击件——PA6 韧性天然更好，做护罩、外壳、卡扣更合适。
③ 表面质量要求高的件——PA6 浮纤倾向比 PA66 小，外观件更友好。
④ 薄壁、复杂流道、加工难度大的件——PA6 流动性好、窗口宽，成品率更高。
那笔"省了八毛"的账
有家做电动工具外壳的厂，把 PA66-GF30 换成了 PA6-GF30。理由很实在：同样玻纤含量，PA6 便宜，一公斤省七八毛。
一个外壳大概用 180 克料。年用量算下来，一年省了六万多。财务那边很满意。
问题出在第十七个月。
那批机器卖到了西北。夏天中午户外作业，外壳暴晒之后壳内温度能到 90℃ 以上，PA6 在这个温度下长期跑，老化速度肉眼可见——先是表面失去光泽、泛白，然后卡扣位置开始出现细小的银纹，最后是在跌落测试里一次性碎了三个。
他们统计的返修率是 3.2%。听起来不多，但那是几千台的规模，加上物流、人工、换新机的成本，以及渠道那边的不满——这笔账最后是六万的二十几倍。
一条判断捷径：温度是分界线，其他都是微调。 先看长期温度是否超过 120℃，这一条能筛掉八成犹豫。
而且反过来也成立：温度没超 120℃，你就没有理由多花那份钱。省下来的是真金白银。
三、PA6 的四条改性路线
路线一：增强（用量最大）
| 玻纤含量 | 拉伸强度 | 弯曲模量 | 缺口冲击 | 热变形温度 | 收缩率 |
|---|
| 纯 PA6 | 70-80MPa | 基准 | 最高 | 基准 | 1.5-2.0% |
| GF15 | 约 100-120MPa | 约 2 倍 | 下降约 40% | +40℃ | 0.8-1.2% |
| GF30 | 约 160-180MPa | 约 3 倍 | 下降约 50% | +70℃ | 0.3-0.7% |
| GF50 | 约 190-210MPa | 约 4 倍 | 下降约 60% | +90℃ | 0.2-0.4% |
（典型趋势值，具体以牌号 TDS 为准）
怎么选含量：
GF15：轻增强，主要提升刚性和尺寸稳定性，保持较好韧性。适合壳类、盖类
GF30：主力区间。绝大多数结构件的答案，刚性够、韧性还能接受
GF50：高刚性高耐蠕变，但缺口冲击掉到 GF30 的一半左右，浮纤和翘曲明显
走 GF50 之前先问一句：是刚性真不够，还是结构设计不够？很多时候加强筋比加玻纤便宜。
玻纤这件小事：长度比含量更容易被忽略
同样标称 GF30，出来的件性能可能差两成。原因是玻纤的保留长度。
玻纤在螺杆里会被剪断。剪切越强、背压越高、浇口越窄，断得越厉害。粒子里的玻纤如果是 3 毫米，打进制品里可能只剩 0.2 到 0.4 毫米——而这 0.2 和 0.4 之间，冲击强度能差出一截。
这不是材料厂能完全控制的事，它一半在配方，一半在你的机器上。
所以有个很土但很有效的判断办法：把打出来的件砸断，看断面的玻纤。抽出来的长、拉丝明显，说明保留长度还行；断得跟粉末一样、纤维很短，就要回头查背压和浇口了。
我在注塑车间见过老师傅就这么干——随手从废料箱里捡个浇口，往地上摔一下，听了声就知道大概。这套判断不写在任何标准里，但它十次能对七八次。
路线二：增韧
PA6 本身韧性不错，但低温和缺口敏感场景仍会脆断。
超韧 PA6：常温缺口冲击可做到 50-70 kJ/m²，-40℃ 仍不脆裂
常温和低温要分清：普通弹性体增韧在低温会失效，-30℃ 以下场景要用核壳结构增韧
代价：刚性、耐热、耐候同步下降。增韧剂加到 15% 以上，件会软到影响尺寸稳定性
判断低温韧性看一个数：-40℃ 缺口冲击强度（ISO 179）。常温 60 kJ/m² 但 -40℃ 掉到 5 的料，在寒区就是脆断源。
这里有个真实的地理分界线。
同一个件，在广东湛江卖三年没出过一个投诉，到了黑龙江佳木斯，第一批就断了一片。厂家第一反应是"这批料批次有问题"，查来查去，批次很稳——是温度变了，而当初的选型是按常温做的。
增韧剂本身也是材料，它也有自己的玻璃化温度。普通的弹性体类增韧剂，到了低温自己先变硬变脆了，它在基体里就不起"缓冲"作用了，等于白加。
所以寒区件、冷藏物流件、户外北方设备，不要只问"增韧了没有"，要问"用的是什么类型的增韧体系"，然后直接要 -40℃ 的那一行数据。
路线三：阻燃
两个体系的选择逻辑：
| 有卤（溴系+锑） | 无卤（磷系/次膦酸盐/MCA） |
|---|
| 效率 | 高，加量少 | 低，加量大 |
| 对力学影响 | 小 | 韧性、流动下降明显 |
| 成本 | 低 | 高 20-40% |
| 合规 | 部分法规受限 | 环保合规 |
| 适用 | 一般工业件 | 家电、出口、汽车内饰 |
PA6 阻燃要注意一个额外代价：PA6 本身韧性好，但加阻燃剂后韧性掉得比 PA66 明显。做阻燃 PA6 时，往往需要同时补一点增韧体系，这是 PA6 阻燃方案比 PA66 更复杂的地方。
选阻燃 PA6 必须看三项，不是一项：
UL94：考自熄性（入门券）
GWIT：考灼热丝起燃（家电、断路器安规真正卡的）
CTI：考漏电起痕（高压件、光伏件的分水岭）
这三项分开是对的，因为它们考的根本不是一件事。
UL94 是把试样点着、拿开火源、看多久自熄，它考的是"着火以后会怎样"。
GWIT 是把一根灼热丝压在样片上，看会不会起燃，它模拟的是"元件发热会不会引燃外壳"——这是家电和低压电器安规真正卡的那一项。
CTI 是在样片表面滴电解液、加电压，看会不会形成导电通路，它考的是"脏污潮湿的表面会不会爬电"——高压件和光伏件的分水岭。
一个料可以 UL94 V-0 做得漂漂亮亮，GWIT 只有 700℃，装在断路器里照样过不了认证。这不是厂家偷工减料，是三个测试考的压根不是同一道题。
路线四：填充与功能
矿物填充：低翘曲、尺寸稳定、表面好，代价是增重、强度提升有限
玻微珠：改善各向异性，适合精密件
耐磨改性（PTFE / MoS₂）：做齿轮、滑块、衬套。注意对偶件匹配
导热 / 导电：功能性需求，用量小但单价高
四、典型应用与材料对应
| 应用 | 常用牌号方向 | 关键要求 |
|---|
| 电动工具外壳 | PA6-GF30（+增韧） | 抗冲击、外观、刚性 |
| 家电结构件 | PA6-GF15/GF30 | 成本、刚性、耐热 80-100℃ |
| 汽车内饰件 | PA6+增韧 / PA/ABS | 韧性、低气味、可喷涂 |
| 齿轮、滑块 | PA6+GF+MoS₂ 或 PA6 增韧 | 耐磨、疲劳强度 |
| 电子外壳 | PA6-GF15 阻燃 | UL94 V0、GWIT |
| 家居五金 | PA6-GF30 / 纯 PA6 | 成本、尺寸稳定 |
| 水暖管件 | PA6-GF30 | 耐水解（需专门体系） |
| 运动器材 | PA6 增韧、PA6-GF | 抗冲击、韧性 |
这张表真正有用的用法是倒着看：先确定你最在意的那一项，再往上找对应的料，而不是先看应用。
因为同一个"电动工具外壳"，手持式的和台式的、有跌落要求的和没有的、价格敏感的和高端的，选料方向完全不同。
五、PA6 加工的四个要点
要点一：干燥比什么都重要
PA6 吸水率高，干燥不充分是 PA6 件出问题的头号原因。
条件：80-100℃ × 4 小时（视设备和环境湿度调整）
目标含水率：< 0.1%-0.2%
关键提醒：回料必须单独烘干，且回料吸水更快
含水率超标的后果是熔融水解、分子量断链——冲击强度断崖式下跌，而且外观上看不出来。
最后半句是最要命的。外观上完全看不出来。
见过一个厂，夜班图省事，把当天打下来的浇口和废件直接粉碎，倒在料斗里，和新料掺一起，第二天开机打。颜色正常，尺寸也合格，打出来堆在成品区看着漂漂亮亮。
出问题是在三个月后的客户端：一批件在装配线上卡扣一掰就断。追溯回料批次，发现是那一夜的货。
原因是在螺杆里的那几分钟。水遇到 250℃ 的熔体，会把 PA6 的分子链剪断——这就是水解。分子量一降，韧性跟着崩。这种损伤是化学层面的，打出来的件看不出，测尺寸也测不出，只有做冲击测试才会暴露。
所以行业里有句老话：PA6 的质量，一半在配方，一半在烘箱。
要点二：料温和模温
| 材料 | 料温 | 模温 |
|---|
| PA6 / PA6-GF | 240-260℃ | 80-100℃ |
| 增韧 PA6 | 230-250℃ | 60-80℃ |
| 阻燃 PA6 | 230-250℃ | 70-90℃ |
模温是 PA6 件表面质量的开关。 模温低于 60℃，浮纤、光泽差、熔接线强度不够会同时出现。做外观件，模温必须拉起来。
浮纤是什么样子：深色件的表面浮起一层细密的、发白的雾状纹路，像蒙了一层霜，手摸上去没有凹凸感，但光打上去会散。它不是缺陷那么简单——它意味着表面的树脂层太薄，玻纤直接顶到了皮下的位置。
解决办法最有效的往往就是一件事：把模温拉高。模温一高，熔体在型腔里流动的前锋温度高，玻纤不容易被"冻结"在表面，能被后续的熔体推回去。
代价是周期变长。这是一笔可以直接算的经济账：多出来的几秒钟，乘以每天的模次，乘以电费和产能。有的厂算了以后决定忍一忍，接受轻微的浮纤；有的厂做外观件，多花的这几秒是必须付的。
要点三：玻纤件的尺寸要等
PA6-GF 件测尺寸，注塑后 24 小时和 48 小时的数据能差 0.1-0.2%。刚下线就送检等于没测。
而且玻纤件纵横收缩率不同（GF30 可能到 1:2 甚至 1:3），长条件必然有翘曲倾向——这是材料特性，不是工艺没做好。
这个"1:2"值得解释一下。
玻纤在熔体里会顺着流动方向排列，一旦顺着这个方向排列，它就在这个方向上抑制收缩（纤维不收缩），而在垂直于它的方向上没有这个约束，收缩照旧。
结果是一个长条形的 PA6-GF30 件，沿着流动方向收缩可能只有 0.3%，垂直于流动方向却有 0.7%。两边缩得不一样多，件自然往一边弯。
这不是可以通过调机能根治的问题——它是各向异性，写在玻纤增强的物理本质里。能做的只有三件事：改浇口位置和数量让纤维排布更乱、加矿物填料（玻微珠）降低取向性、或者在设计阶段就把翘曲量算进去。
要点四：模具钢材和浇口
PA6 本身对模具磨损不大，但加玻纤后磨损明显加剧。GF30 以上的料，模具建议用硬化钢，浇口设计要避免高剪切（防止玻纤断裂）。
六、五个最常见的坑
坑 1：拿 PA6 顶 PA66 的耐热活
长期 >120℃ 的工况，加耐热体系的 PA6 也撑不住。短期看不出问题，一年后开始脆化。温度是硬边界，不能靠配方突破。（开篇那个齿轮箱盖就是这一条）
坑 2：只看干态强度选料
PA6 吸水后强度可能掉 30% 以上。数据表好看的是干态，实际在南方梅雨天里运行的是湿态。选受力件必须问湿态数据。
有个非常简单的现场验证：同一个件，夏天梅雨季装配顺畅，冬天干燥以后同样的配合开始松动——这是尺寸走了；反过来，冬天装配正常、夏天卡死的，多半是吸胀。两个方向，指向的是同一件事。
坑 3：以为 PA6 韧性好，就不用增韧
PA6 常温韧性确实好，但低温和缺口敏感场景一样会脆断。而且加玻纤之后，PA6 的韧性优势会被大幅削弱——GF30 的 PA6 并不比 GF30 的 PA66 韧多少。
坑 4：阻燃 PA6 只看 UL94
家电类真正卡的是 GWIT，高压件卡的是 CTI。UL94 V0 过了但 GWIT 不达标，认证照样过不了。
坑 5：不干燥就上机
PA6 吸水高，这条比任何其他材料都严重。开机就投料、料斗不盖、回料不烘——这三个动作足以废掉一整批。
七、边界声明：这些活 PA6 干不了
| 工况 | 结论 | 替代方向 |
|---|
| 长期连续 >140℃ | PA6 体系有天花板 | PA66 + 热稳定，或 PA46 / PA6T |
| 长期接触高温水（>80℃） | 水解降解 | PA612 / PA1010，或专门耐水解体系 |
| 长期接触燃油机油 | 耐油性不足 | PA66 或长碳链 |
| 公差 ±0.05mm 长期稳定 | 吸水导致尺寸不可控 | PA66、PA9T |
| 高 CTI 600V + 薄壁 | PA6 体系难兼顾 | PPA 或 PA9T |
| 长期户外 20 年 + 承载 | 老化风险 | 完整耐候体系或换材 |
行业里的一条实感：客户说"要韧性好"，十次有九次说的其实是"别再断了"——但"韧性"这个词本身，是没法直接选料的。 我们这边 PA6 类询盘里最常见的误用，是拿 PA6 去顶 PA66 的耐热件。 理由通常很实在：手上有 PA6 的库存、PA6 便宜、"加个耐热改性应该就能行"。短期确实打得出来，但长期高温工况下，它会用变形和老化把这件事还回来——PA6 的耐热天花板不在助剂上，在分子结构上，改不动。
回到那三句话
还记得开篇我问客户的那三句吗？现在可以解释了。
第一句：断在常温还是低温？
常温脆断，多半是韧性不够或者已经老化降解；低温脆断，往往是增韧体系的类型选错了——普通弹性体在低温下自己先脆了，要换核壳结构才管用。
同一个"脆"字，两套药。问错一句，改半年也改不对。
第二句：断口是齐的还是不齐的？
齐口、发白、无拉丝，是典型的脆性断裂；有拉丝、有韧窝、断面发白区域呈纤维状，说明材料在断之前经历了明显的塑性变形——那通常是过载，不是材料问题。
看断口这件事，比看任何报告都快。五秒钟能给你一个方向，然后你再用数据去验证它。
第三句：是每一模都断，还是偶尔断？
这句话客户当时答不上来，后来他去翻了记录，发现是集中在某几个批次。
这就把问题从"材料整体不行"缩小到了"某批生产过程有变量"。后来查出来，是那几个批次用了回收比例偏高的回料——又绕回了干燥那一节。
你看，一个电话里的三句话，把排查范围从"换料"缩小到了"查批次"。
真正的答案最后很简单：那个位置长期 130℃，改 PA66-GF30，加耐热稳定体系。改完到现在一年多，没有再断过。
结语
PA6 是最值得用好的一个材料——它便宜、好加工、韧性好，用对了就是性价比之王。
但它有两个硬边界：
一个是温度（长期 120℃），一个是吸水（8-10%）。
越过温度边界，换 PA66；越过吸水边界，换长碳链。
这两条守住，PA6 能解决你八成的通用结构件需求；守不住，再好的配方也救不回来。
而剩下的那两成，往往不是配方的问题——是在选料的那十分钟里，有没有把"挺烫的"这三个字换算成一个具体的数字。
Last July, a client who makes garden tools sent two pieces.
Black fiberglass reinforced parts, palm-sized, are the top cover of a handheld lawn mower gearbox. When he sent them, he used a kraft paper envelope without foam pads, and the two broken pieces made noise when they clashed inside.
He spoke directly over the phone: "Your material isn't good, right? It cracked after half a year." "
I put the two pieces on the table and laid them out. The first thing I noticed wasn't the cracks, but the fractures.
The breaks were very clean—even, white, with no strings or tough edges. There's a simple saying in the industry: 'brittle break', not 'pull break.' Tearing has warning signs—deformation, whitening, slowly tearing apart; Brittle break happens in an instant. It was fine yesterday, but today a piece falls off with just a touch.
Then I asked him three questions.
First sentence: Is the break at room temperature or at low temperature?
He said room temperature, July in Guangdong, the workshop is 37 or 8 degrees.
Second sentence: Is the fracture uniform or uneven?
I let him touch it—aligned, like a knife cut.
Third sentence: Does it break every mold, or does it break occasionally?
He was stunned for a moment and said, "Isn't it the same when you use it? Why do they break ...... occasionally?"
I said it's different. Let's put that aside for now; we'll come back in the sixth section.
What he later figured out was: the material was fine, the grade was correct, the RTI was delivered, and the GF30's mechanics were fulfilled. It was the operating condition was exceeded.
That gearbox cover was right next to the rear end of the cylinder block, and the actual long-term operating temperature was around 130°C. He had said long ago that this spot was "quite hot," but no one converted "quite hot" into a specific number—while the long-term continuous operating temperature of PA6 is generally written as 100-120°C.
The difference of ten degrees isn't just a little, it's a whole material system off.
This article is exactly what this article wants to talk about: where is the performance boundary of PA6, how to tell if you've stepped on it, and where to go after stepping on.
1. First, get a clear look at the true details of PA6
PA6 The basic set is just six numbers to remember:
| Parameters | Typical Value | Description |
|---|
| Melting Point | 220℃ | 45℃ lower than PA66 |
| Equilibrium Water Absorption | 8-10% | Relatively high within the Nylon family |
| Tensile Strength of Pure Resin | 70-80MPa | Dry State |
| Tensile Strength after GF30 Reinforcement | 160-180MPa | Main level for tooling structural parts |
| Long-term Continuous Use Temperature | 100-120℃ | Can reach 140-150℃ with heat-resistant system |
| Shrinkage Rate | 1.5-2.0% (pure) / 0.3-0.7% (GF30) | Greatly affected by glass fiber content |
After listing the numbers, the important part comes next: what these numbers mean in your workshop.
Why is PA6 the cheapest ?
The reason isn't in modification plants, but in the upstream monomer.
PA6 monomer is caprolactam, while PA66 monomer is hexamethylenediamine plus adipic acid. The caprolactam industry chain has been running domestically for over thirty years, with concentrated capacity, mature processes, and a single unit being very large. With large scale, unit costs drop—this is the simplest rule in the chemical industry.
So when you ask "Why is PA6 cheaper?", the answer isn't very romantic: it's because it was made early, large-scale, and has many competitors.
What exactly does 8-10% water absorption mean ?
This is the most easily read as "one attribute." Let's convert it to that.
A 200-gram PA6 product, if you throw it in during the southern rainy season and reach equilibrium, it can absorb 16 to 20 grams of water—about one tablespoon.
With that spoonful of water, three things will happen:
First, the size will increase. After absorbing water, PA6 expands in volume, on the order of several thousandths. It doesn't sound big, but a 100-millimeter-long fitting hole increases from 100.00 to 100.30, and in terms of interference fit, it goes from "just right" to "jamming."
Second, the strength drops. Water molecules enter between the molecular chains, creating plasticizing effects—in simple terms, it loosens the tightly packed molecular chains. Pure dry PA6 at 80MPa can drop to 45-55MPa when saturated. This is not a theoretical value, but a measured sample soaked in water for several days.
Third, it will come back. When winter dries out, the water drains away and the part shrinks back. This rise and fall rub back and forth on the fit surface, with a limit on the number of times.
So customers in the south especially need to pay attention: What moisture content are the tolerances you mark on the plastic part drawings under what moisture content are you marking? Most people can't say.
The phrase "PA6 has the best toughness" needs to be discounted .
The textbook says that PA6 has better toughness than PA66 because its molecular chain has a lower amino group density (one every 6 carbons), fewer hydrogen bonds between molecules, and more room for chain segment movement.
This statement itself is correct, but it fails in three situations:
after adding glass fiber. The toughness gap between GF30's PA6 and GF30's PA66 is greatly smoothed out—the real toughness is determined by the glass fiber content and toughening system, not the substrate
at low temperatures. At room temperature, the "toughness" at -20°C may not exist at all
in places with notches. Sharp corners, sudden changes in wall thickness, weld lines—these places make PA6 just as brittle, and without warning .
In short: if the operating conditions allow, PA6 is the answer for cost performance. The problem is—many people can't tell if the operating conditions allow it.
2. PA6 or PA66? A single schedule
This is the real question to answer when choosing PA6. Look at this comparison:
| Dimension | PA6 | PA66 | Which is better |
|---|
| Melting point | 220℃ | 265℃ | PA66 |
| Long-term continuous temperature resistance | 100-120℃ | 120-140℃ | PA66 |
| Short-term peak tolerance | About 150℃ | About 180-200℃ | PA66 |
| Equilibrium water absorption | 8-10% | 8-9% | PA66 is slightly better |
| Toughness (notch impact) | Higher | Slightly lower | PA6 |
| Rigidity and strength | Slightly lower | Higher | PA66 |
| Oil and fuel resistance | Average | Better | PA66 |
| Surface quality (fiber float tendency) | Better | Worse | PA6 |
| Processing window | Wide, easy to process | Narrow, high melt temperature | PA6 |
| Relative price | ★ | ★★ | PA6 |
Where does the melting point difference at 45°C come from?
On the first row of the table, 220 versus 265, a 45°C difference. This difference is not a formulated value; it is determined by molecular structure and cannot be changed. On the molecular chain of
PA6, an amide group appears every 5 methylene groups; PA66 alternates every 4 (hexamethylenediamine group) and 4 (adipic acid group) groups. Amide groups form hydrogen bonds between molecules; the higher the hydrogen bond density, the tighter the chains are locked, requiring more energy to pull them apart—macroscopic manifestations indicate a higher melting point and better heat resistance.
PA66 has a higher amide density than PA6, so its hydrogen bond network is denser. This isn't something that "adding heat resistant can fix it"; the purpose of heat resistant is to slow aging, not to raise the ceiling.
That's why the customer's part at the beginning broke down. He chose PA6, located at 130°C. The upper limit that PA6 with a heat-resistant system can reach is about 140-150°C, but that's "can hold up without breaking immediately," not "stable long-term."
Long-term accounts are calculated annually.
Five situations where PA66 must be replaced
If any of the following occur, do not force it with PA6:
(1) Long-term continuous usage temperature >120°C
Engine compartment parts, lamp holders, and parts near heat sources should be directly used with PA66. This line is no exception, and it is not recommended to "try" heat-resistant masterbatch.
(2) Higher rigidity and strength are needed
Similarly, GF30 and PA66 have about 10-15% higher strength than PA6. If the structure has reached its limit and fiberglass can no longer be added, replacing with PA66 is a way to improve performance without changing the mold.
(3) Long-term oil and fuel
PA66 have significantly better oil resistance than PA6. For parts near the engine oil pan and fuel circuit, PA66 is preferred.
(4) High dimensional accuracy requirements
PA66 high crystallinity, slightly lower water absorption, and better dimensional stability than PA6. Parts with tolerances ± 0.05mm class must be carefully selected.
(5) If electrical safety regulations require
circuit breakers, contactors, and high-voltage components, PA66 has better heat resistance and electrical performance baselines, making it easier to achieve high GWIT and CTI.
Continue using four situations of PA6
(1) Long-term ≤ of 120°C general structural parts—this is PA6's main domain, replacing it with PA66 is wasteful.
(2) Requires high-toughness, impact-resistant parts—PA6 naturally has better toughness and is more suitable for covers, shells, and clips.
(3) Parts with high surface quality requirements—PA6 has less tendency to float fibers than PA66, making it more user-friendly for appearance
(4) Thin-walled, complex runners, and difficult parts to process—PA6 has good flowability, wide windows, and higher yield .
That "saved 0.8 yuan " account
A factory making power tool casings replaced PA66-GF30 with PA6-GF30. The reason is very practical: with the same fiberglass content, PA6 is cheaper, saving 70 to 0.8 yuan per kilogram.
One shell uses about 180 grams of material. Calculated annually, it saved over 60,000 yuan annually. The finance side was very satisfied.
The problem was in the seventeenth month.
That batch of machines was sold to the northwest. During outdoor work at noon in summer, the inside temperature of the casing can reach over 90°C after exposure to the sun. PA6 running at this temperature for a long time shows visible aging speed—first the surface loses its shine and turns white, then fine silver lines start to appear at the buckle area, and finally, three pieces shattered in one drop test.
Their reported repair rate is 3.2%. It doesn't sound like much, but that's a scale of several thousand units, plus logistics, labor, replacement costs, and dissatisfaction from the channels—the total is over twenty times 60,000 .
A shortcut to judgment: temperature is the dividing line, everything else is fine-tuning. First, check if the long-term temperature exceeds 120°C; this line can filter out 80% hesitation.
And the reverse holds: if the temperature doesn't exceed 120°C, you have no reason to spend extra money. What you save is real money.
3. Four PA6 modification routes
Route 1: Reinforcement (maximum usage)
| Glass fiber content | Tensile strength | Bending modulus | Notch impact | Thermal distortion temperature | Shrinkage rate |
|---|
| Pure PA6 | 70-80MPa | Baseline | Maximum | Benchmark | 1.5-2.0%. |
| GF15 | About 100-120 MPa | About 2 times | Decrease by about 40% | 40℃ | 0.8-1.2% |
| GF30 | About 160-180 MPa | About 3 times | Decrease by about 50% | 70℃ | 0.3-0.7% |
| GF50 | About 190-210 MPa | About 4 times | Decrease by about 60% | 90℃ | 0.2-0.4% |
(Typical trend values, refer to the specific grade TDS for accuracy)
How to choose the content:
GF15: Light reinforcement, mainly improves rigidity and dimensional stability while maintaining good toughness. Suitable for shells and covers
GF30: Main range. The answer for most structural parts, rigidity is sufficient and toughness is acceptable
GF50: High rigidity and high creep resistance, but notch impact drops to about half of GF30, noticeable floating fibers and warpage
Before leaving GF50, let me ask: Is rigidity really insufficient, or is the structural design insufficient? Often, reinforcing ribs are cheaper than fiberglass.
Fiberglass, this small matter: length is easier to overlook than content .
For products labeled as GF30, the performance of the product may be 20% worse. The reason is the length retention of fiberglass.
Fiberglass will be sheared inside the screw. The stronger the cut, the higher the back pressure, and the narrower the gate, the more severe the breakage. If the fiberglass inside the particles is 3mm, the product may only have 0.2 to 0.4mm left in the product—and between 0.2 and 0.4, the impact strength difference is quite significant.
This isn't something the material factory can fully control; it's half about the formula and half on your machine.
So there's a very rustic but effective way to judge: break the punched part and check the cross-section. If the extracted parts are long and clearly drawn, it means the length is still preserved; If the broken parts are like powder and the fibers are very short, then you need to check back pressure and gate.
I've seen experienced craftsmen in injection molding workshops do it this way—pick a gate from the scrap bin, drop it on the ground, and you can roughly tell the gist just by listening. This judgment isn't written in any standard, but it can be accurate seven or eight times out of ten.
Route 2: Toughening
PA6 has good toughness, but can still break brittle in low-temperature and notch-sensitive scenarios.
Super toughness PA6: can withstand 50-70 kJ/m² of notch impact at room temperature, and still does not crack at -40° C
Distinguish between room temperature and low temperature: ordinary elastomer toughening fails at low temperatures, and below -30°C, use core-shell structures for toughening
Cost: rigidity, heat resistance, and weather resistance decrease simultaneously. If toughening agent is added above 15%, the parts will become so soft that they affect dimensional stability
To judge low-temperature toughness, look at a number: -40°C notch impact strength (ISO 179). At room temperature 60 kJ/m², but material dropping to -5°C at -40°C is fragile and broken in cold regions.
There is a real geographic dividing line here.
The same piece was sold in Zhanjiang, Guangdong for three years without a single complaint, but in Jiamusi, Heilongjiang, the first batch was broken. The manufacturer's first reaction was "There is a problem with this batch of material." After checking around, the batch was very stable—the temperature had changed, and the original selection was made at room temperature.
Toughening agents themselves are also materials and have their own vitrification temperature. Ordinary elastomer toughening agents harden and become brittle at low temperatures, so they lose their "buffering" effect in the matrix, which is basically wasted.
So for cold region parts, refrigerated logistics parts, and outdoor northern equipment, don't just ask "Is toughening done?" but "What type of toughening system is used?" and then directly ask for the -40°C data line.
Route 3: Flame Retardant
Logic for choosing between two systems:
| With halogen (bromine, antimony) | No halogen (phosphorus, hypophosphite, MCA) |
|---|
| Efficiency | High, small addition | Low, large addition |
| Impact on mechanical properties | Small | Toughness and flow decrease significantly |
| Cost | Low | High 20-40% |
| Compliance | Partially restricted by regulations | Environmentally compliant |
| Application | General industrial parts | Home appliances, exports, automotive interiors |
An extra cost must be noted for PA6 flame retardancy: PA6 itself has good toughness, but after adding flame retardants, its toughness decreases more obviously than PA66. When making flame-retardant PA6, it is often necessary to add some toughening system at the same time, which makes the PA6 flame-retardant solution more complex than PA66.
When selecting flame-retardant PA6, three aspects must be considered, not just one:
UL94: tests self-extinguishing property (entry-level)
GWIT: tests glow wire ignition (really crucial for home appliances and breaker safety standards)
CTI: Tests leakage marks (the dividing line between high-voltage and photovoltaic components)
These three are correct because they are not the same thing.
UL94 tests lighting the sample, removing the ignition source, and seeing how long it takes to extinguish itself. It tests "what happens if it catches fire."
GWIT It tests pressing a hot wire onto the sample to see if it catches fire. It simulates "whether the component heating will ignite the casing"—this is the real safety requirement for home appliances and low-voltage electrical appliances.
CTI It tests whether a conductive path is formed by dripping electrolyte and voltage on the sample surface. It tests "whether dirty and damp surfaces will cause electricity creepage"—the dividing line between high-voltage and photovoltaic components.
One material can make UL94 V-0 look beautiful, GWIT is only 700°C, so even when installed in circuit breakers, it still won't pass certification. This isn't a cut by manufacturers; the three tests are not the same problem at all.
Route 4: Filling and Function
Mineral Filling: Low warpage, dimensional stability, good surface, but the cost is weight gain and limited strength gain .
Glass Microbeads: Improve anisotropy, suitable for precision parts
Wear-resistant modification (PTFE / MoS₂): Used for gears, sliders, and bushings. Pay attention to matching pairing components
Thermal conductivity / electrical conductivity: functional requirements, small usage but high unit price
4. Typical Applications and Material Correspondence
| Application | Common Grade Direction | Key Requirements |
|---|
| Power Tool Housing | PA6-GF30 (toughened) | Impact resistance, appearance, rigidity |
| Home appliance structural parts | PA6-GF15/GF30 | Cost, rigidity, heat resistance 80-100° C |
| Automotive interior parts | PA6 toughened / PA/ABS | Toughness, low odor, paintable |
| Gears, sliders | PA6 GF MoS₂ or PA6 toughened | Wear resistance, fatigue strength |
| Electronic housings | PA6-GF15 flame retardant | UL94 V0, GWIT |
| Home hardware | PA6-GF30 / pure PA6 | Cost, dimensional stability |
| Plumbing fittings | PA6-GF30 | Hydrolysis resistance (requires specific system) |
| Sports equipment | PA6 toughened, PA6-GF | Impact resistance, toughness |
The genuinely useful way to use this table is to read it in reverse: first determine the aspect you care about most, then look up the corresponding material, rather than starting from the application.
Because for the same 'power tool housing,' handheld and benchtop, with and without drop requirements, cost-sensitive and high-end, the direction of material selection is completely different.
5. Four key points in PA6 processing
Key Point 1: Drying is more important than anything else
PA6 High water absorption rate, insufficient drying is the number one cause of problems with PA6 components.
Condition: 80-100°C × 4 hours (adjust depending on equipment and ambient humidity)
Target moisture content: < 0.1%-0.2%
Key reminder: The return material must be dried separately, and the return material absorbs water more quickly .
Excessive moisture content results in melt hydrolysis and molecular weight chain breakage—impact strength drops sharply, and it's not visible visually.
The last half of the sentence is the most critical. It's completely invisible from the outside.
once saw a factory that worked night shifts to save trouble, crushing the gates and scrap parts that day and pouring them into the hopper, mixing them with new material, and starting the machine to process the next day. The color was normal, the dimensions were qualified, and the finished products looked neat when piled in the finished product area.
The problem came three months later at the client: a batch of parts broke when snapped on the assembly line. Tracing the rebound batch, it turned out to be the goods from that night.
The reason was during those few minutes inside the screw. When water meets 250°C melt, it cuts the PA6 molecular chains—this is hydrolysis. When molecular weight drops, toughness collapses. This kind of damage is chemically related; you can't see it from the parts you drive, and you can't measure the size either. Only impact testing can reveal it.
So there's an old saying in the industry: half the quality of PA6 depends on the formula, half in the oven.
Key Point 2: Material temperature and mold temperature
| Material | Material temperature | Mold temperature |
|---|
| PA6 / PA6-GF | 240-260°C | 80-100°C |
| toughening PA6 | 230-250°C | 60-80°C |
| Flame retardant PA6 | 230-250°C | 70-90° C |
Mold temperature is the switch for PA6 surface quality. If the mold temperature is below 60°C, floating fibers, poor gloss, and insufficient weld line strength will all appear. When making exterior parts, the mold temperature must be raised.
What does floating fiber look like: a dense, white, misty texture appears on the surface of dark-colored parts, as if covered with frost. It doesn't feel uneven to the touch, but when lighted on, it disperses. It's not just a simple defect—it means the resin layer on the surface is too thin, and the fiberglass directly hits the subcutaneous area.
The most effective solution is often one thing: raise the mold temperature. When the mold temperature is high, the melt flowing in the cavity has a higher front temperature, making the fiberglass less likely to be "frozen" on the surface and pushed back by subsequent melt.
The trade-off is a longer cycle. This is a straightforward economic calculation: the extra few seconds multiplied by daily mold cycles, multiplied by electricity cost and production capacity. Some factories decide to endure the calculation and accept slight floating fibers; while others produce exterior parts and have to pay for the extra few seconds spent.
Key Point 3: The dimensions of fiberglass parts should be measured by
PA6-GF parts; the data difference between 24 and 48 hours after injection molding can be 0.1-0.2%. Sending samples for inspection right after production is basically not tested.
Moreover, fiberglass parts have different longitudinal and transverse shrinkage rates (GF30 may reach 1:2 or even 1:3), so under long conditions, warping is inevitable—this is a material characteristic, not a poor process.
This "1:2" ratio is worth explaining.
Glass fibers in the melt align along the flow direction; once arranged in this direction, it suppresses shrinkage in that direction (the fibers do not shrink), while perpendicular to the fibers are not constrained and shrinkage remains normal.
The result is a long PA6-GF30 piece, shrinking along the flow direction may only be 0.3%, but perpendicular to the flow direction is 0.7%. If the two sides shrink by a different amount, the piece naturally bends to the side.
This is not a problem that can be fixed by adjusting the function—it is anisotropy, written in the physical essence of glass fiber reinforcement. There are only three things you can do: change the gate position and quantity to make the fiber arrangement more chaotic, add mineral filler (glass microbeads) to reduce orientation, or factor in warpage during the design phase.
Key Point 4: Mold steel and gate
PA6 do not cause much mold wear by themselves, but wear is significantly worsened after adding glass fiber. For materials above GF30, it is recommended to use hardened steel for molds, and gate design should avoid high shear (to prevent glass fiber breakage).
Six, the five most common pitfalls
Pit 1: Using PA6 to match PA66's heat resistance
long-term > 120°C conditions, even PA6 with heat-resistant systems can't hold up. You won't notice problems in the short term, but after a year it starts to become brittle. Temperature is a hard boundary and can't be broken through by formulas. (The gearbox cover mentioned at the beginning is exactly this)
Pit 2: Only look at dry state strength when selecting materials
PA6 After absorbing water, strength may drop by more than 30%. The data sheet looks good when dry, but in actual southern rainy seasons, it operates in wet mode. When selecting load-bearing components, wet data must be asked.
has a very simple on-site verification: the same piece fits smoothly during the summer rainy season, but starts loosening after drying in winter—this means the size is off; Conversely, if the assembly works normally in winter but gets stuck in summer, it's mostly due to suction and swelling. Both directions point to the same thing.
Pitfall 3: If you think PA6 has good toughness, you don't need toughness
PA6 Room temperature toughness is indeed good, but it can break brittle in low temperatures and sensitive notch-sensitive scenarios. Moreover, adding fiberglass greatly weakens PA6's toughness advantage—GF30's PA6 is not much tougher than GF30's PA66.
Pitfall 4: For flame-retardant PA6, only look at UL94
Home appliances. The real problem is GWIT, while high-voltage parts are stuck with CTI. UL94 V0 passes but GWIT doesn't meet standards, so certification still won't pass.
Pitfall 5: Put it on the machine without drying
PA6 High water absorption, this is more serious than any other material. Feed right at startup, don't cover the hopper, don't dry the material back—these three actions are enough to ruin an entire batch.
7. Boundary Statement: These tasks PA6 cannot do
| Operating conditions | Conclusion | Alternative direction |
|---|
| Long-term continuous >140 °C | PA6 systems have ceiling | PA66 + thermal stability, or PA46/PA6T |
| Long-term contact with high-temperature water (>80°C) | Hydrolysis degradation | PA612 / PA1010, or specialized hydrolysis-resistant systems |
| Long-term contact with fuel engine oil | Insufficient oil resistance | PA66 or long carbon chains |
| tolerance ±0.05mm Long-term stability | water absorption leads to uncontrollable size | PA66, PA9T |
| high CTI 600V + thin-walled | PA6 systems that are difficult to balance | PPA or PA9T |
| 20 years long-term outdoor + bearing | aging risk | complete weather-resistant system or material replacement |
A real feeling in the industry: when customers say "good toughness," nine out of ten times they actually mean "don't break again"—but the word "toughness" itself can't be directly selected. The most common misuse in our PA6 inquiries is using PA6 to replace PA66 heat-resistant parts. The reasons are usually very practical: they have PA6 in stock, PA6 is cheap, "adding heat-resistant modification should work." In the short term, it can be effective, but under long-term high-temperature conditions, it will compensate for it through deformation and aging—PA6's heat resistance ceiling isn't in additives, but in molecular structure, it can't be changed.
Back to those three sentences
Do you still remember the three questions I asked the client at the beginning? Now it can be explained.
First sentence: Is the fracture at room temperature or at low temperature?
Brittle fractures at room temperature are mostly due to insufficient toughness or aging and degradation; Low-temperature brittle fractures are often due to choosing the wrong type of toughening system—ordinary elastomers become brittle at low temperatures and need to replace the core-shell structure to work.
The same word "brittle" means two different medicines. If you ask one wrong question, you won't get it right even after half a year.
Second sentence: Is the fracture surface uniform or uneven?
Uniform edge, whitening, and no strings are typical brittle fractures; If there are brushed wires, tough pits, and the whitened areas of the cross-section appearing fibrous, it means the material underwent obvious plastic deformation before breaking—usually overload, not a material issue.
Checking the fracture is faster than reading any report. Five seconds gives you a direction, and then you verify it with data.
Third sentence: Does every die break, or does it break occasionally?
The customer couldn't answer this question at the time, but later he checked the records and found it was concentrated in a few batches.
This narrowed the problem from "the material as a whole is poor" to "variables in a certain batch during production." Later, it was found that those batches used recycled material with a higher recycling ratio—then circled back to the drying section.
See, three sentences on one phone call narrowed the inspection scope from "material replacement" to "batch inspection."
The real answer is simple: that location is long-term 130°C, switch to PA66-GF30, add a heat-resistant stabilization system. It's been over a year since the modification and hasn't stopped again.
Conclusion
PA6 It's the most worthwhile material to use—it's cheap, easy to process, and tough, and when used correctly, it's the king of cost-effectiveness.
But it has two hard boundaries:
One is temperature (long-term 120°C), the other is water absorption (8-10%).
Cross the temperature boundary, switch to PA66; Cross the water absorption boundary, switch to a longer carbon chain.
If you stick to these two points, PA6 can solve 80% of your general structural component needs; If you can't hold them, no matter how good the formula is, it won't save you.
And the remaining 20% is often not the formula's issue—it's about whether you convert the words "quite hot" into a specific number during the ten minutes of material selection
