上个月，一个做家具脚轮的客户寄来四个轮子和两个门窗滑轮。
轮子件是改性尼龙注塑的轮芯，米灰色，其中一个的轮面被压出一个明显的平台。门窗滑轮更小，壳体是同一个料，轮缘上磨出了一道浅槽。
他在电话里说得很直接："这个轮子不是磨坏的，是柜子放那儿没动，半年就自己压扁了。"
我问他三句：压扁的是承重轮还是转向轮？单轮承重多少、柜子多重？地面是什么、常年静置还是经常推？
前两句他答得清楚——压扁的是承重轮，单轮约三十五公斤，柜子常年不动。第三句他补了一句：地面是瓷砖。
这三句问完，"压"和"磨"就先分开了。
这篇把家用脚轮和门窗滑轮这两件分开讲，也把"压扁"和"磨掉一层"为什么是两本账讲清。
一、压扁和磨掉一层，是两条互不相干的线
用户描述脚轮坏掉，常用的词是"磨平了"。但从工程上看，这是两种完全不同的失效。
其一是磨耗。 轮子在地上滚，接触面被一次次剪切带走一点材料。它和行走距离、地面粗糙度、载荷正相关，和放了多久关系不大。
其二是压缩永久变形。 轮子长期被压在一个位置上，材料发生蠕变，卸力之后回不去。它和静置时长、载荷、温度正相关，和走了多少路关系不大。
那四个轮子属于后一种：柜子半年没动，轮子和地面的接触点始终是同一个点，接触应力一直压在那几平方毫米上。
算一笔账就清楚了。单轮三十五公斤，轮面接触宽度按两毫米估，接触面积大约几平方毫米到十几平方毫米。
换成压强，就是每平方毫米几公斤的量级——这个量放在钢上没事，放在塑料上，蠕变就是必然要面对的事。
所以那位客户把问题描述成"不耐磨"，方向就已经偏了：耐磨料解决不了放半年压扁的问题，抗蠕变才管。
一句话：脚轮的失效，先分"压"和"磨"。分不清这两条线，换几次料都还在原地。
二、工况六维：脚轮上的六个数字
载荷。 家用与办公家具的单轮承重常见 20 到 60 公斤；这是静载，动载系数还要往上乘。静置状态下的长期载荷，比行走时的峰值更值得盯。
移动与静置的比例。 这是脚轮和工业轮最不一样的一维。办公椅一天走几百米，储物柜可能一年移动两次。同一款轮子打这两类客户，参数取向完全相反。
地面。 瓷砖、木地板、地毯、水泥、门槛条。地面既决定磨耗，也决定噪音和地板保护要求。
温度与湿度。 阳台柜、卫浴柜、桑拿房附近的家具，长年处于高湿与温差里；尼龙类件的尺寸会跟着走。
外观与静音。 卧室和木地板场景对滚动噪声敏感，硬轮在空鼓地板上尤其明显。
气味与室内空气。 室内家具对气味有要求，落地按客户或平台指定的方法确认。
六维里，载荷、静置比例、地面、温度这四条能给出具体数，判断也就有依托。
顺带把一层的道理说明白。
塑料的蠕变，本质是分子链在恒定应力下缓慢地互相滑动、重新排列。温度越高，链段越容易动，蠕变越快。
这解释了两件事：为什么同一款轮子放在阳台比放在书房更容易压平；为什么提高结晶度、让晶区把非晶区锁得更紧，能明显改善压平。
换算成时间：常温室内的脚轮，压平往往在几个月到一年之间显形；到了四十度以上的环境，这个时间会缩短。
三、脚轮的三条路线，各自让掉什么
| 路线 | 抗蠕变（对压平） | 磨耗 | 静音与地板友好 | 承载上限 | 常见定位 |
|---|
| PA6 / PA66 + 耐磨体系 | 中到好（看结晶与刚性） | 好 | 中，偏硬 | 好 | 承重轮、办公椅轮 |
| POM 共聚 | 中 | 好，本征自润滑 | 中 | 中 | 小载荷、转向轮、滑块 |
| 轮芯 + TPE / TPU 包胶 | 看轮芯 | 好（胶层保护地面） | 好 | 看轮芯 | 木地板、静音场景 |
| 硬质 PVC / PP（对照） | 差，易压平 | 中 | 中 | 低 | 低价轻载件 |
关键不在哪条数值高，在每种方案把代价放在了哪一头。
PA6 体系的代价在吸水与蠕变。它综合性能均衡、成本可控，但长期恒定载荷下的压平要靠结晶度和刚性去补。
POM 的代价在承载与刚性。它摩擦低、尺寸稳，做小轮和转向件很合手；大载荷承重轮上，它的刚性不够用。
包胶轮的代价在工艺。胶层解决静音和地板保护，代价是多一道结合面，结合面的处理质量决定寿命——这一层零点几毫米的界面，往往就是整只轮子寿命的全部。
再补一句容易被忽略的：同一只柜子上的四个轮子，受力并不平均。重心偏移时，一只轮子可能吃掉六成的重量。
按平均值选型，装上去就有一只先坏。
四、门窗滑轮是另一件：小轮、高频、U 型槽
这位客户一起寄来的两个门窗滑轮，工况和脚轮完全不同，值得单独说。
载荷更小，频次更高。 推拉门滑轮单轮承重常见 20 到 40 公斤，但推拉频次高，一个家庭一天几十次，十年下来是十万次量级。
受力方向不一样。 脚轮是压在地面上滚，滑轮是在型材的 U 型槽里滚，轮缘和槽壁之间有侧向约束。槽的宽度和轮缘的厚度配不上，就会偏磨。
失效形态不一样。 脚轮先看压平，滑轮先看轮缘磨出凹槽。槽一深，门就往下垂，垂到一定程度就刮地。
还有一条是脚轮上没有的：滑轮常和金属轴、金属型材配对。
塑料轮对钢轴的摩擦副，和塑料对瓷砖的摩擦副不是一回事。选材逻辑要按对偶件重新过一遍。
把这两件放在一起看，会得到一个很实际的结论：它们可以共用同一条材料路线，但不能共用同一套验收指标。
用脚轮的报告去覆盖滑轮，是模板化交付最常见的一种。
五、选型判据表（脚轮与滑轮建议收藏这一页）
门限值是方向性建议，不是验收标准。实际数值必须由载荷、静置比例、地面与型材配合确定。
| 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 |
|---|
| 压缩永久变形 | 长期承载后压下量在设计余量内 | GB/T 7759，或按实际温度加严 | 轮面压出平台、柜体下沉 | 提高结晶度与刚性 | 成核剂（结晶均匀性） |
| 磨耗量（对地面或对型材） | 按寿命折算，磨痕不超设计余量 | GB/T 1689（常见参照）或自建往复台架 | 轮缘磨出凹槽、轮径变小 | 耐磨体系 + 对偶面搭配 | 耐磨填料 |
| 轮缘圆度与径向跳动 | 按件精度定，常见 0.05 mm 量级起 | 三坐标 / 圆度仪 | 滚动跳动、异响、偏磨 | 浇口与取向设计 | — |
| 轴孔与轴承座变形 | 长期承载后配合不松 | 持久载荷后复测孔径 | 轮子发旷、运转偏摆 | 提高刚性 + 嵌件设计 | 偶联剂（界面） |
| 缺口冲击（含低温） | 覆盖仓储与使用最低温 | GB/T 1043.1 简支梁 | 运输跌落后轮缘崩块 | 增韧体系 | 增韧剂 |
| 湿态尺寸与配合 | 吸湿后仍能装入轴与型材槽 | 调湿前后实测 | 装不进、运转发紧 | 低吸水基体 + 调湿交付 | 成核剂 |
| 滚动噪音 | 按卧室与办公两类场景分别定 | 自建噪音台架 | 低频异响、地板共鸣 | 包胶或降低硬度 | — |
| 耐地面清洁剂与油污 | 擦拭或浸泡后尺寸外观无异常 | 介质擦拭 + 浸泡实测 | 表面溶胀、粘灰 | 选耐介质基体 | 抗氧剂（抑制老化） |
怎么用这张表：先看第一行，再看第二行。
第一行管"放着不动"，第二行管"动着走"。两条线的先后，取决于这个家具是常静置还是常移动。
表里有一处要提醒：门窗滑轮的轮缘厚度与型材槽宽的配合，很多项目没有现成标准可依。没有标准可依时，把配合尺寸和磨耗判据写进技术协议，不要省掉这一项。
六、五种失效，和它们真正的根因
失效一：放着不动，半年自己压扁。
根因是压缩永久变形。这一条上，客户最常走错的一步就是换成"更耐磨"的料。
耐磨和抗蠕变在配方上是两个方向：前者靠填料和表面硬度，后者靠结晶度和刚性。拿耐磨料去治压平，等于拿错药。
失效二：四个轮子里，总是某一个先坏。
根因多半是载荷分配。家具重心偏、地面不平、某一个轮子被门框挡住使不上力，都会让个别轮子长期超载。
先量载荷分布，再谈换料。
失效三：包胶轮用一段时间，胶层和轮芯之间分层。
根因在结合面的工艺。轮芯表面处理不到位、脱模剂残留没清干净，胶粘不牢，剥落就从这零点几毫米的界面起头。
这一类是工艺问题，不是材料配方能兜住的。
失效四：轮子用一段时间，表面粘灰、发黏。
（助剂侧归因）这一条多数出在助剂迁移上。润滑剂用量偏高或选型偏向外润滑，会在件表面慢慢形成一层析出物，落上灰尘就变成黏腻的一层。
排查时先看表面成分，再回看润滑体系，不要先换基材。
同一批件摆在一起比，深浅不一的情形也不罕见，耐磨填料的分散状态就写在颜色上，先把混料那道工序看一遍，再谈换基体。
失效五：滑轮轮缘磨出凹槽，推拉门下沉刮地。
根因通常是两条叠加：轮缘与型材槽的配合偏松，加上轮缘材料的抗磨不足。
解法是先修配合，再考虑换料。 配合不改，换料只能把凹槽磨得慢一点，不能让它不出现。
一条时间线，这类件最常见的走法：
柜子出厂时推起来很顺 → 客户装在阳台上，一放就是两个季节 → 第一个梅雨季过后，柜体开始往一侧斜 → 拆件发现承重轮压出平台 → 复测发现那只轮子长期吃了六成载荷 → 调整轮子布置与载荷分配，同时把轮芯换成抗蠕变更强的体系 → 同型号在室内场景继续用原方案，两档并行。
从头到尾，用户说的"磨平了"都不是磨耗。
七、加工与验证：轮子件上有几件事必须提前定
干燥。 尼龙类必烘，含水率超标会在熔融时水解降解，轮子表面看着没事，内部强度已经掉了。
模温与结晶。 抗压平的能力很大一部分来自结晶状态，模温偏低时表层结晶不足，压平更容易从表层先起。
厚壁与冷却。 轮子是厚壁回转体，内外冷却差大，容易留下内应力。内应力不直接导致压平，但会让轮子在长期载荷下更早出问题。
嵌件与轴孔。 轴孔周围是最容易变形的区域。设计上要给足肉厚和圆角，嵌件要预热，避免冷嵌件周围形成应力集中。
验证顺序。 建议这样排，顺序不要换：
1. 材料级：压缩永久变形、磨耗量、缺口冲击
2. 件级：轮缘圆度、轴孔配合、静载压平测试
3. 台架级：按实际载荷与频次做滚动或往复试验
4. 整机级：装到家具上，在真实地面跑完整工况
5. 环境叠加：温度 + 湿度 + 长时静载，最后一项最常被跳过
这里有个内行细节：静载压平测试要按实际使用温度做，不能只在常温做。
常温下压不出明显平台的料，在阳台温度下可能一个月就变形。温度和时长是一起起作用的两个变量，缺哪一个，测出来的都是装饰数据。
八、边界：什么情况下这两件不该走塑料路线
其一，单轮长期承重超过百公斤量级的位置。 这个量级下塑料的抗蠕变很难兜住，改走金属轮或加大轮径分担是更省事的做法。
其二，长期处于高温区域的家具。 靠近暖气、桑拿、户外暴晒的位置，蠕变会明显加速，要考虑换体系或改结构。
其三，对地板保护要求极高的场景。 抛光实木地板、软木地板这类地面，硬质塑料轮的划伤风险高，走软质包胶更稳妥。
其四，年用量小到摊不平模具与验证。 轮子和滑轮通常要开专用模、做静载与环境验证，量太小，从成本上不成立。
把这四条先摆出来，是为了让项目少走一轮。 试错的钱不在样品上，在整批换模和客户信任上。
换料风险清单（从原方案换到改性尼龙轮件，要动的东西）
| 环节 | 要动什么 | 容易漏的点 |
|---|
| 模具 | 轮缘厚度、轴孔公差与浇口位置按载荷重定 | 沿用原 PU 轮的壁厚 |
| 干燥 | 按实测含水率定窗口 | 回用料掺入带入的水分 |
| 调湿 | 轴孔与型材配合尺寸按调湿态给 | 干态合格，装配后发紧 |
| 料温 / 模温 | 厚壁件要慢冷，模温比常规件高 | 只按牌号推荐值给 |
| 保压与脱模 | 厚壁收缩大，保压曲线与脱模方式重定 | 脱模太早留下内应力 |
| 包胶 / 嵌件 | 结合面处理与嵌件预热要单独定 | 表面处理与脱模剂残留 |
| 验证顺序 | 材料 → 件级 → 台架 → 整机 → 环境叠加 | 只做常温静载，不测使用温度 |
一页纸汇报表（给要向上汇报的人）
| 场景 | 推荐路线 | 关键指标 | 验证标准 | 需先确认的条件 |
|---|
| 常静置的储物柜脚轮 | 高结晶度改性尼龙 | 压缩永久变形、抗蠕变 | GB/T 7759 按使用温度 | 单轮长期载荷 |
| 办公椅与常移动件 | 耐磨体系 + 合理轮径 | 磨耗量、圆度 | GB/T 1689 或往复台架 | 地面材质 |
| 木地板静音场景 | 轮芯 + 软质包胶 | 噪音、地板保护 | 自建噪音台架 + 现场试用 | 地面类型 |
| 推拉门窗滑轮 | 低摩擦耐磨体系 | 轮缘磨耗、配合间隙 | 往复台架 + 配合实测 | 型材槽宽 |
| 百公斤级重载 | 金属轮或加大轮径 | 载荷、蠕变 | 对应产品标准 | 是否属重载路径 |
风险提示：本路线的主要不确定性在长时静载下的压缩永久变形，不在初始磨耗。
读者常问的三句
问：脚轮到底选尼龙还是 POM？
看载荷和静置情况。小载荷、常移动、要低摩擦，POM 合手；中大载荷、长期承重，改性尼龙的刚性和抗蠕变更能兜。同一只柜子上两种料混着用也常有。
问：包胶轮一定比全塑料轮好吗？
看场景。木地板、卧室、要求静音，包胶的好处很实在；但它多了一道结合面，工艺不到位反而先从这里坏。加分项能不能拿到，取决于结合面的处理质量。
问：滑轮能不能直接套用脚轮的料？
料可以共用，指标不能共用。滑轮看轮缘磨耗和与型材槽的配合，脚轮看压平与地面磨耗。报告要用各自的口径出，不能一张表打通两个件。
结语
回到那批压出平台的轮子。
后来我们先做的事是量载荷：那只压扁的轮子确实长期吃着六成的重量。轮子布置调了，轮芯也换了一版结晶度更高的体系。
同一型号在室内书房那批没动，两档并行。
开头那三句问话之所以先问出来，是因为它们分别落在三个不同的面上：问位置、问载荷、问是"放着"还是"动着"。 分清了静与动，压和磨就不会再混成一件事。
脚轮与滑轮的判断链，说到底只有三条：
使用方式（静置还是移动）定路线 → 单轮实际载荷定等级 → 湿态尺寸与调湿定配合。
三条定完，"用哪个料"自然就有答案了。
如果你手上正有一只脚轮或滑轮要定料，把三样东西发过来就能给方向：单轮长期承重、常静置还是常移动、地面或型材的材质。
三年五年之后还那样 —— 这四个字听起来像承诺，其实是把静置时长、载荷分配和温度一起算进工况表之后的结果。
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。
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。
这类件的选料与试模，可以一起聊。
Last month, a customer who makes furniture casters sent four wheels and two window and door pulleys.
The wheel component is a wheel core made of modified nylon injection molding, light gray in color, with one of the wheel surfaces pressed to form a noticeable flat area. The door and window pulleys are smaller, the housing is made of the same material, and a shallow groove has been worn on the wheel rim.
He spoke very directly on the phone: 'This wheel didn't wear out; the cabinet just sat there without moving, and in half a year it got flattened by itself.'
I asked him three questions: Is it the load-bearing wheel or the steering wheel that gets crushed? How much weight does a single wheel bear, and how heavy is the cabinet? What is the ground like, and is it stationary most of the time or frequently pushed?
He answered the first two sentences clearly—the ones being compressed are the load-bearing wheels, each wheel weighs about thirty-five kilograms, and the cabinet doesn’t move throughout the year. For the third sentence, he added a remark: the floor is tiled.
After asking these three questions, 'press' and 'grind' were separated first.
This article discusses household casters and door and window pulleys separately, and also explains why 'flattening' and 'wearing off a layer' are two different matters.
1. Flattening and grinding off a layer are two unrelated lines.
The user described the caster as broken, and the commonly used term is 'worn flat.' But from an engineering perspective, these are two completely different failures.
The first is wear. When the wheels roll on the ground, the contact surfaces are repeatedly sheared and carry away some material. This is directly related to travel distance, ground roughness, and load, but not much to how long it has been left standing.
The second is compressed permanent deformation. When a wheel is pressed in one position for a long time, the material undergoes creep and cannot return after the force is removed. It is positively correlated with the duration of being stationary, load, and temperature, and has little to do with how far it has traveled.
Those four wheels belong to the latter type: the cabinet hasn't been moved for half a year, and the points where the wheels contact the ground have always been the same, with the contact stress constantly pressing on those few square millimeters.
It becomes clear after doing the math. A single tire weighs thirty-five kilograms, and if the tread contact width is estimated at two millimeters, the contact area is roughly a few square millimeters to a dozen square millimeters.
Converting to pressure, it is on the order of several kilograms per square millimeter—this amount is fine for steel, but for plastic, creep is an inevitable issue.
So that client described the problem as 'not wear-resistant,' and the direction was already off: wear-resistant materials can't solve the issue of being flattened after six months; only creep resistance matters.
In one sentence: The failure of casters can be first divided into 'pressure' and 'abrasion'. If you can't distinguish between these two lines, no matter how many times you change the material, it will remain in the same place.
2. Six Dimensions of Operating Conditions: The Six Numbers on the Caster
Load. The single-wheel load capacity of household and office furniture is commonly 20 to 60 kilograms; this is the static load, and a dynamic load factor should also be multiplied on top. The long-term load in a stationary state is more worth monitoring than the peak caused during movement.
The ratio of movement to stationary. This is the most distinct dimension between casters and industrial wheels. An office chair may travel hundreds of meters a day, while a storage cabinet might only be moved twice a year. Using the same type of wheel for these two kinds of customers, the parameter orientation would be completely opposite.
Floor. Tiles, wooden flooring, carpet, cement, threshold strips. The floor determines both wear and tear, as well as noise and floor protection requirements.
Temperature and humidity. Furniture near balcony cabinets, bathroom cabinets, and saunas is exposed to high humidity and temperature fluctuations year-round; the size of nylon items will change accordingly.
Appearance and quietness. Bedrooms and wooden floor settings are sensitive to rolling noise, and hard wheels are particularly noticeable on hollow floors.
Odor and indoor air. Indoor furniture has requirements for odor, and on-site confirmation is done according to the method specified by the customer or platform.
In six dimensions, load, idle ratio, ground, and temperature can provide specific numbers, giving a basis for judgment.
By the way, explain the reasoning of one layer clearly.
The creep of plastics is essentially the slow sliding and rearrangement of molecular chains under constant stress. The higher the temperature, the easier the chain segments move, and the faster the creep occurs.
This explains two things: why the same wheel is more easily flattened on the balcony than in the study; and why increasing crystallinity, allowing the crystalline regions to lock the amorphous regions more tightly, can significantly improve flattening.
Converted to time: For caster wheels at normal room temperature, flattening usually becomes apparent within a few months to a year; in environments above forty degrees, this time will be shortened.
3. The three paths of the caster, what each yields
| Route | Creep resistance (to flattening) | Wear | Silent and floor-friendly | Load capacity | Common positioning |
|---|
| PA6 / PA66 Wear-Resistant System | Medium to good (depending on crystallinity and rigidity) | Good | Medium hard | Good | Load-bearing wheels, office chair wheels |
| POM Copolymer | middle | Okay, intrinsically self-lubricating | middle | middle | Small load, caster wheel, slider |
| Wheel core TPE/TPU coating | Check the wheel core | Good (glue layer protects the floor) | Good | Check the wheel core | Wood flooring, silent scene |
| Rigid PVC / PP (comparison) | Poor, easily flattened | middle | middle | Low | Low-cost light-duty parts |
The key is not which number is higher, but where the cost is placed in each option.
The drawback of the PA6 system lies in water absorption and creep. Its overall performance is balanced and costs are controllable, but under long-term constant load, flatness must be compensated by crystallinity and rigidity.
The cost of POM lies in its load-bearing capacity and rigidity. It has low friction and stable dimensions, making it suitable for small wheels and steering components; for large load-bearing wheels, its rigidity is insufficient.
The cost of a coated wheel lies in its craftsmanship. The rubber layer addresses noise reduction and floor protection, but the trade-off is an additional bonding surface. The quality of the treatment of this bonding surface determines its lifespan—this interface, just a few tenths of a millimeter thick, often accounts for the entire lifespan of the wheel.
One more easily overlooked point: the four wheels on the same cabinet do not bear the load equally. When the center of gravity shifts, one wheel may take on 60% of the weight.
If you choose based on the average, one will fail as soon as it's installed.
4. The pulleys of doors and windows are another item: small wheel, high frequency, U-shaped groove
The two door and window pulleys sent by this customer together have completely different working conditions from the casters, and are worth mentioning separately.
Smaller load, higher frequency. The load-bearing capacity of a single sliding door pulley is commonly 20 to 40 kilograms, but the sliding frequency is high. A household may operate it dozens of times a day, which adds up to the magnitude of 100,000 times over ten years.
The directions of the forces are different. Casters roll on the ground under pressure, while pulleys roll in the U-shaped groove of the profile, with lateral constraints between the flange and the groove wall. If the width of the groove does not match the thickness of the flange, it will cause uneven wear.
The failure modes are different. For caster wheels, first check for flattening; for pulleys, first check if the wheel rim has worn grooves. Once the groove is deep, the door will sag, and when it sags to a certain extent, it will scrape the ground.
There is another one that casters do not have: pulleys are often paired with metal shafts and metal profiles.
The friction pair between a plastic wheel and a steel axle is not the same as the friction pair between plastic and tiles. The selection logic needs to be reviewed according to the mating parts.
Putting these two together will lead to a very practical conclusion: they can share the same material route, but cannot share the same set of acceptance criteria.
Using a caster report to cover a pulley is one of the most common forms of templated delivery.
5. Selection Criteria Table (It is recommended to bookmark this page for casters and pulleys)
Threshold values are directional recommendations, not acceptance criteria. The actual values must be determined by the load, the resting ratio, and the coordination between the ground and the profiles.
| Indicator | Directional Threshold | Verification Method / Standard | Common Failures | Common solution | Corresponding auxiliary agent system |
|---|
| Compression set | The settlement after long-term loading is within the design tolerance. | GB/T 7759, or tighten according to the actual temperature | Wheel surface extrusion platform, cabinet sinking | Increase crystallinity and rigidity | Nucleating agent (crystallization uniformity) |
| Wear (on the ground or on the profile) | When converted based on lifespan, wear marks do not exceed the design margin | GB/T 1689 (commonly referenced) or self-built reciprocating test bench | Rim worn grooves, wheel diameter reduced | Wear-resistant system Paired surfaces matching | Wear-resistant packing |
| Rim roundness and radial runout | Determined by piece accuracy, commonly starting at the 0.05 mm level | CMM / Roundness Tester | Rolling vibration, abnormal noise, uneven wear | Gate and Orientation Design | — |
| Shaft hole and bearing seat deformation | No looseness after long-term load | Re-measure the aperture after sustained load | Wheel is loose and running with wobble | Improve rigidity Insert design | Coupling agent (interface) |
| Gap impact (including low temperature) | Coverage of storage and minimum usage temperature | GB/T 1043.1 Simply Supported Beam | Flange chipping after transport drop | Toughening system | Toughening agent |
| Wet dimensions and tolerances | Can still be installed into shafts and profile grooves after absorbing moisture | Measured before and after humidity adjustment | Cannot fit in, operation is tight | Low water-absorption substrate Moisture-conditioned delivery | Nucleating agent |
| Rolling noise | Set separately according to the two types of scenarios: bedroom and office | Self-built noise test bench | Low-frequency abnormal noise, floor resonance | Coating or reducing hardness | — |
| Floor cleaner and grease | No abnormal changes in size or appearance after wiping or soaking | Media wiping Soaking actual measurement | Surface swelling, ash adhesion | Select a resistant medium substrate | Antioxidant (inhibits aging) |
How to use this table: first look at the first row, then look at the second row.
The first line says 'tube stays still', the second line says 'tube moves along'. The order of the two lines depends on whether this piece of furniture is usually stationary or frequently moved.
There is one point to remind you: the fit between the thickness of the pulley's rim for doors and windows and the width of the profile groove — many projects do not have an existing standard to follow. When there is no standard to follow, include the fit dimensions and wear criteria in the technical agreement; do not omit this item.
Six, Five Types of Failures, and Their True Root Causes
Failure 1: Leaving it untouched causes it to get flattened by itself in half a year.
The root cause is permanent compression deformation. In this regard, the step that customers most often get wrong is switching to 'more wear-resistant' material.
Wear resistance and creep resistance are two different directions in formulation: the former relies on fillers and surface hardness, while the latter relies on crystallinity and rigidity. Using wear-resistant materials to treat flattening is like using the wrong medicine.
Failure Two: Among the four wheels, there is always one that breaks down first.
The root cause is mostly load distribution. If the furniture's center of gravity is off, the floor is uneven, or a certain wheel is blocked by a door frame and can't bear weight, it will cause individual wheels to be overloaded for a long time.
First measure the load distribution, then talk about changing the material.
Failure three: After using the rubber-coated wheel for a period of time, delamination occurs between the rubber layer and the wheel core.
The root cause lies in the processing of the bonding surface. If the surface treatment of the wheel core is inadequate or the release agent residue is not thoroughly cleaned, the adhesive will not bond firmly, and peeling will start from this interface of just a few tenths of a millimeter.
This kind of issue is a process problem, not something that can be covered by the material formula.
Failure four: After using the wheels for a period of time, the surface accumulates dust and becomes sticky.
(Additive-side attribution) This issue mostly arises from additive migration. When the lubricant dosage is too high or the selection leans towards external lubrication, a layer of exudate gradually forms on the surface of the parts, which becomes a sticky layer when dust settles on it.
When troubleshooting, first look at the surface components, then review the lubrication system, and do not replace the base material first.
When pieces from the same batch are placed together for comparison, it is not uncommon to see variations in shade. The dispersion of the wear-resistant filler is reflected in the color. First, review the mixing process, then discuss changing the matrix.
Failure 5: The pulley rim is worn into a groove, causing the sliding door to sink and scrape the floor.
The root cause is usually a combination of two factors: the fit between the rim and the profile groove is relatively loose, coupled with insufficient wear resistance of the rim material.
The solution is to first adjust the fit, and then consider changing the material. If the fit is not changed, changing the material can only slow down the abrasion of the groove, but it cannot prevent it from appearing.
A timeline, this is the most common approach for this type of item:
The cabinet was easy to push when it left the factory → The customer installed it on the balcony, and it stayed there for two seasons → After the first rainy season, the cabinet started to lean to one side → When disassembled, it was found that the load-bearing wheel had pressed out the platform → Retesting revealed that the wheel had been bearing 60% of the load for a long time → Adjusted the wheel arrangement and load distribution, and replaced the wheel core with a stronger creep-resistant system → The same model continues to use the original scheme in indoor settings, running both options in parallel.
From beginning to end, the 'smoothed out' that the user mentioned is not wear.
7. Processing and Verification: There are a few things regarding wheel parts that must be decided in advance
Dry. Nylon materials must be dried; if the moisture content exceeds the standard, it will hydrolyze and degrade when melted. The surface of the wheel may look fine, but the internal strength has already decreased.
Mold temperature and crystallization. A large part of the ability to resist flattening comes from the crystallization state. When the mold temperature is relatively low, the surface crystallization is insufficient, making the flattening more likely to start from the surface.
Thick walls and cooling. The wheel is a thick-walled rotating body, with a large difference in cooling between the inside and outside, making it easy to retain internal stress. Internal stress does not directly cause flattening, but it can cause the wheel to fail earlier under long-term load.
Inserts and shaft holes. The area around the shaft holes is the most prone to deformation. In design, sufficient wall thickness and fillets should be provided, and inserts should be preheated to avoid stress concentration around the cold inserts.
Verification order. It is recommended to arrange it like this, do not change the order:
1. Material level: compressive permanent deformation, wear amount, notch impact
2. Part level: rim roundness, shaft hole fit, static load flatness test
3. Test Bench Level: Perform rolling or reciprocating tests according to actual load and frequency
4. Complete machine level: Install on the furniture and run the full conditions on a real floor
5. Environmental superposition: temperature humidity long-term static load, the last item is most often skipped
Here's an insider detail: the static crush flatness test should be conducted at the actual operating temperature, not just at room temperature.
Materials that do not show a noticeable plateau when pressed at room temperature may deform in a month at balcony temperature. Temperature and duration are two variables that work together; without either, the measurements obtained are just decorative data.
8. Boundaries: In what situations these two items should not go the plastic route
Firstly, positions where a single wheel bears a long-term load exceeding the hundred-kilogram level. At this magnitude, it is difficult for plastic to resist creep, so switching to a metal wheel or increasing the wheel diameter to share the load is a more convenient approach.
Secondly, furniture that is in high-temperature areas for a long time. Positions close to heating, saunas, or outdoor exposure to the sun will significantly accelerate creep, so one should consider changing the system or altering the structure.
Thirdly, scenarios that require extremely high floor protection. For floors such as polished solid wood and cork floors, there is a high risk of scratches from hard plastic wheels, so using soft rubber-coated wheels is safer.
Fourth, the annual usage is too small to justify spreading the cost of molds and validation. Wheels and pulleys usually require dedicated molds and static load and environmental validation, but if the quantity is too small, it is not feasible from a cost perspective.
Putting these four points out first is to save the project from going through an extra round. The money for trial and error isn’t in the samples, but in changing the mold for the entire batch and in the client’s trust.
Material Change Risk List (From the original plan to modified nylon wheel components, things that need to be changed)
| link; segment; part | What needs to be moved? | Points that are easy to overlook |
|---|
| Mold | Rim thickness, shaft hole tolerance, and gate location are redefined according to the load. | Maintain the wall thickness of the original PU wheel |
| Dry | Set the window according to the measured moisture content | Recycled materials mixed with the water content brought in |
| Humidity control | The bore and profile fitting dimensions are given according to the conditioned state | Dry state qualified, tight after assembly |
| Material Temperature / Mold Temperature | Thick-walled parts should be cooled slowly, and the mold temperature is higher than that of conventional parts. | Give only according to the recommended value by grade |
| Pressure Holding and Demolding | Thick-wall shrinkage is large, and the holding pressure curve and demolding method are redefined | Demolding too early leaves internal stress |
| Overmolding / Insert | The combination of surface treatment and insert preheating needs to be determined separately | Surface Treatment and Release Agent Residue |
| Verification order | Material → Component Level → Test Bench → Complete Machine → Environmental Superposition | Only perform static load at room temperature, do not test operating temperature |
One-page report form (for people who need to report upwards)
| Scene | Recommended Route | Key indicators | Verification Standard | Conditions that need to be confirmed first |
|---|
| Casters for frequently stationary storage cabinets | High-crystallinity modified nylon | Compression set and creep resistance | GB/T 7759 According to Operating Temperature | Single-cycle long-term load |
| Office chairs and frequently moved parts | Wear-resistant system Reasonable wheel diameter | Wear amount, roundness | GB/T 1689 or reciprocating test bench | Ground material |
| Silent wood flooring scenario | Core Soft coating | Noise, floor protection | Self-built noise test bench on-site trial | Ground type |
| Sliding door and window pulley | low-friction and wear-resistant system | Flange wear, fitting clearance | Reciprocating test bench with actual measurements | Profile slot width |
| Hundred-kilogram class heavy load | Metal wheels or enlarged wheel diameter | Load, creep | Corresponding Product Standards | Whether it is an overloaded path |
Risk warning: The main uncertainty of this route lies in the permanent compressive deformation under long-term static load, not in the initial wear.
Three questions readers often ask
Question: For casters, should I choose nylon or POM?
Look at the load and idle conditions. For small loads, frequent movement, and low friction, POM fits well; for medium to large loads and long-term bearing, modified nylon's rigidity and creep resistance are more reliable. It's also common to mix the two materials on the same cabinet.
Question: Are rubber-coated wheels necessarily better than all-plastic wheels?
It depends on the situation. For wooden floors and bedrooms, with a requirement for quietness, the advantages of rubber coating are very practical; but it adds an extra joint, which, if the craftsmanship is not up to standard, might fail from there first. Whether you get extra points depends on the quality of the treatment of the joint.
Question: Can pulleys directly use the materials of casters?
Materials can be shared, but indicators cannot be shared. For pulleys, check the rim wear and the fit with the profile groove; for caster wheels, check the crushing and wear on the ground. Reports should be issued according to their respective standards, not combined into a single table for both parts.
Conclusion
Back to those wheels pressed out by the platform.
Later, the first thing we did was measure the load: that crushed wheel indeed bore 60% of the weight for a long time. The wheel arrangement was adjusted, and the wheel core was also replaced with a version of a system with higher crystallinity.
The same model in the indoor study hasn't been touched; the two settings are running in parallel.
The reason the first three questions are asked at the beginning is that they each pertain to a different aspect: asking about position, asking about load, and asking whether it is 'stationary' or 'moving'. Once the distinction between still and moving is clear, pressure and friction will no longer be confused as the same thing.
When it comes to judging casters and pulleys, there are ultimately only three criteria:
Usage method (stationary or mobile) determines the route → Single-wheel actual load determines the grade → Wet dimensions and moisture adjustment determine the fit.
Once the three rules are set, the answer to 'which material to use' naturally emerges.
If you currently have a caster or pulley to determine the material for, sending over three things can provide guidance: whether it is a single wheel for long-term load, usually stationary or frequently moving, and the material of the floor or profile.
Three or five years later still the same — these four words sound like a promise, but in fact, they are the result of taking into account the duration of idle time, load distribution, and temperature together in the operating conditions table.
What we do is very specific: we take resins like PA6, PA66, PA46, PA11, PA12, PA6T, PA9T, and nylon alloys, and turn them into a form that can actually be used for a certain part; we also do modified PPO, PPS, and thermoplastic elastomers along the way.
Also operates the nylon resin, secondary brand materials, and bulk materials of major chemical giants.
The additive system in the formula is matched according to the working conditions of the parts—conventional additives are kept in stock, special models are matched as needed; you report the working conditions and grade, and the material and additives are prepared together at once.
The material selection and mold testing for these types of parts can be discussed together.
