去年秋天，一家做重载 AGV 的客户，把四个磨坏的脚轮寄了过来，问我们能不能做"更耐磨"的改性尼龙。
轮子边缘已经不是圆的，磨成一截一截的小平面，手一推就"咯噔、咯噔"。他电话里的原话我记到现在："同一个批次的料，一台车跑了三年没事，这台半年就磨成这样。"
他认定是买到了假料。我去了现场转了一圈，发现问题不在料上——在那台车的启停频次上。
这是我做移动机器人脚轮这些年最常见的误判：大家嘴上说要"耐磨"，但真正决定轮子寿命的，往往不是耐磨，是另外几件事。这篇就把 AGV 脚轮材料这件事，从头拆一遍。
一、AGV 的轮子，坏在"停"不在"跑"
AGV 和 AMR 的脚轮，和普通工业脚轮不是一回事。
普通脚轮的工况是"被动滚动"——被推着走，受力方向单一。AGV 是主动驱动 + 高频启停：起步要克服整车的惯性，刹车又要把它压回去。驱动轮每一趟都在"加速—减速"里循环。
这个差别带来一个结果：轮子的磨损，主要发生在启停瞬间，不是匀速行驶时。
启停瞬间，轮面与地面之间有一个很短的低速打滑过程。这时候接触点温度骤升、剪切应力集中，轮面被"搓"掉一层。频次越高，搓得越快。那台半年就磨平的车，就是停靠点位特别密——每一趟要走七八个站点，一天启停上千次。
所以看到"磨成平面"，不要先怀疑料。先算一个数：这台车一天启停多少次、单轮承载多少、地面是什么材质。
一句话：AGV 脚轮的选材，第一变量是启停频次，第二变量是地面，第三才轮到材料本身。
二、工况六维：AGV 脚轮被什么夹住
把约束摆开，才知道该盯哪几项。
载荷。 重载 AGV 的单轮承载可以到几百公斤量级，且是动载荷——刹车时单轮瞬时受力远高于静态。这一维必须取峰值，不能取均值。
速度与启停。 最高速度通常不高（1–2 m/s 量级），但加速度和启停频次差异极大。这决定轮面的热负荷与剪切。
地面。 环氧地坪、金刚砂、水泥、格栅、甚至车间油污地面。地面粗糙度直接决定磨耗倍率，同一款轮子换一种地面，寿命能差几倍。
环境。 电子厂、锂电车间、医药车间要求防静电/导电；冷库要低温冲击；食品区还要看冲洗与清洁剂。
寿命与精度。 寿命不只是"能不能转"，还包括轮径变化——磨掉一圈，车体高度就变了，对接精度跟着走。
合规。 防静电等级、阻燃要求、以及部分场景的洁净度要求，都要提前确认适用标准。
六维摆在一起，会看到一个结论：AGV 脚轮没有单项冠军。最耐磨的材料往往最硬，最硬的往往噪音最大、防静电最难做。
三、三条材料路线，并列看代价
| 路线 | 磨耗 | 抗压/承载 | 噪音 | 防静电可行性 | 适合场景 |
|---|
| 尼龙基（PA6 / PA66 + 增韧 + 耐磨体系） | 好 | 好 | 中 | 可做（加导电填料） | 重载、中高启停频次、需防静电 |
| 聚氨酯（PU）包胶轮 | 好 | 中 | 低 | 较难 | 洁净、静音、中等载荷 |
| 金属芯 + 包胶 / POM 轮 | 中 | 好 | 中高 | 可做 | 高承载、低速度、对噪音不敏感 |
看这张表，重点不在"哪个更好"，在代价在哪。
尼龙基的好处是可以做结构件本身——轮芯和轮面一体成型，不用包胶，不怕脱层，还能通过配方做防静电和阻燃。
代价是硬。硬意味着噪音比 PU 大，对地面平整度更敏感。
PU 包胶的好处是软、静音、不伤地面，在医药和电子洁净车间很受欢迎；代价是承载受限，且长期压缩永久变形后会出现"平点"。
还有一条容易被忽略的路：不同的轮子可以用不同的料。 驱动轮承担扭矩和启停剪切，用尼龙基更合适；从动轮和承重轮只承压，可以用更软的体系降噪。一辆车四个轮，本来就不必同一种材料。
AGV脚轮材料的账，最后要落到四个轮子上分头算。
驱动轮看剪切与启停，从动轮看压缩与噪音，承重轮看静载变形。
一辆车只用一种料，往往是把四笔账混成了一笔。
四、选型判据表（这是本篇最该收藏的一页）
门限值是方向性建议，不是验收标准——实际数值必须由你的载荷、启停频次和地面实测确定。
| 指标 | 方向性门限 | 验证方法 / 标准 | 常见失效 | 通行解法 | 对应助剂体系 |
|---|
| 磨耗量（对地面） | 参考阿克隆磨耗或自建台架对比 | GB/T 1689 / 现场复测 | 轮面磨平、轮径变小 | 调整耐磨体系 + 降低启停冲击 | 耐磨填料（二硫化钼 / 硅油类） |
| 压缩永久变形 | 长期承载后回弹在合理区间 | GB/T 7759 | 轮面出现平点、车体下沉 | 提高结晶度与刚性 | 成核剂（结晶均匀性） |
| 缺口冲击（含低温） | 覆盖使用环境最低温 | GB/T 1043 | 轮缘崩块、轮芯开裂 | 增韧体系 | — |
| 表面电阻（防静电） | 按场景定表面电阻区间 | GB/T 1410 / 现场测 | 静电积累、放电 | 导电填料体系 | 导电填料 |
| 轮缘圆度 / 同轴度 | 按件精度定，通常 0.05 mm 量级起 | 三坐标 / 圆度仪 | 行驶跳动、噪音 | 浇口与取向设计 | — |
| 长期热氧保留率 | 按车间最高温考核 | ISO 527 | 轮面发白、脆化 | 稳定化体系 | 抗氧剂 |
| 耐油污与清洁剂 | 浸泡后尺寸外观无异常 | 介质浸泡实测 | 表面溶胀、变软 | 选耐介质基材 | — |
怎么用这张表：不要逐行打分。先看第二行"压缩永久变形"——重载 AGV 的轮子，先垮在"压下去回不来"，再垮在"磨掉一层"。很多项目只盯磨耗，结果装车两个月车体高度就变了。
五、五种常见失效，和它们真正的根因
失效一：半年磨成平面，同期另一台车三年没事。
根因通常不是料，是启停频次与地面差异。停靠点位密的、地面是金刚砂或带油污的，磨耗会成倍上升。先对齐两台车的工况，再谈换料。
失效二：轮面出现"平点"，车体下沉。
根因是压缩永久变形，不是磨耗。长期静载（尤其在高温车间）下，轮子被压成一个平台。解法是提高刚性、加结晶助剂，或者把长时停靠的轮改成分担结构。
失效三：轮芯开裂或嵌件松动。
根因在结构：轮芯壁厚不均、嵌件周围熔接线是弱区，刹车扭矩一冲就裂。这一类换料解决不了，要动模具与嵌件设计。
失效四：同一批轮子颜色深浅不一，性能也散。
这多半是分散问题——导电填料或耐磨填料在混料阶段没混匀。看到这个现象，先查混料工艺和母粒化，不要急着换料。导电填料的分散，比色差更值得盯：分散不均会让表面电阻在轮子上各处不一样。
这里有一条要直说的：AGV 脚轮的失效排查，先算工况，再看结构，最后才怀疑材料。 把顺序倒过来，会一直换料、一直坏。
六、加工与验证：几件必须提前定的事
干燥。 尼龙必烘。含水率超标会在熔融时水解降解，件表面看着没事，内部强度已经掉了。干燥窗口按实际含水率定，不照抄推荐值。
熔接线。 轮子是厚壁回转体，浇口位置决定熔接线落在哪。熔接线如果正好落在受力最大的轮缘，冲击时会先崩在这里。 浇口位置要和受力方向一起定。
圆度与收缩。 厚壁件的收缩不好算，玻纤或填料还会带来方向性差异。轮芯与轮缘如果一体成型，收缩差会在冷却时形成内应力，后期表现为"用着用着开裂"。
厚壁冷却。 轮子是厚壁回转体，内外冷却速度差得大。
模温太高容易缩孔，太低又留下内应力，窗口比薄壁件窄得多。
试模时要把模温拉成几条曲线各试一轮，别只试一个点。
验证顺序。 建议这样排，顺序不要换：
1. 材料级：磨耗、压缩永久变形、冲击、表面电阻
2. 件级：圆度、嵌件拉脱、静载压平测试
3. 台架：模拟实际启停频次的滚轮试验
4. 整车：在实际地面跑完整工况，复测轮径变化
5. 环境叠加：温度 + 油污 + 长时静载，最后一项最容易被跳过
七、边界：什么时候这事不该谈
以下四种情况，这个件走改性尼龙不建议推进：
其一，长期在强溶剂或高浓度油污中浸泡的件。 尼龙的耐介质有边界，超了会溶胀变形，这类该回到金属或特种弹性体。
其二，要求极低噪音的洁净场景。 尼龙基偏硬，噪音天生高于 PU 包胶。要求静音的场合，硬做尼龙是把成本花在了错的方向上。
其三，地面平整度差、又有高速运行要求的场合。 硬轮对地面起伏敏感，跳动大、噪音大、寿命短，这类更适合软轮或充气结构。
其四，年用量小到摊不平模具。 这个件通常要开专用模具，用量太小，从经济上就不成立。
其五，长期在高温车间且长时间静载停靠。 高温加静载，压缩永久变形会明显加速。
这类件要么把刚性提上去，要么在停靠位加支撑分组分担，单靠换料解决不了。
把这五条写在前面，不是劝退，是省时间。 我见过不少项目在样品阶段很顺，装车才发现方向错了——回退的成本，比一开始就不做高得多。
换料风险清单（从原方案换到改性尼龙脚轮，要动的东西）
| 环节 | 要动什么 | 容易漏的点 |
|---|
| 模具 | 轮缘厚度与浇口位置按受力重定 | 沿用原 PU 轮的结构厚度 |
| 干燥 | 按实测含水率定窗口 | 回用料掺入带入的水分 |
| 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，保压曲线要重定 | 脱模太早导致内应力 |
| 嵌件 | 嵌件周围熔接线要避让受力区 | 嵌件预热与包胶温度 |
| 色差 | 导电填料件颜色偏深且批次有差 | 外观要求提前放宽或另定 |
| 验证顺序 | 材料 → 件级 → 台架 → 整车 → 环境 | 跳过台架直接装车 |
一页纸汇报表（给要向上汇报的人）
`
项目：AGV / AMR 脚轮 · 材料路线评估
结论方向：改性尼龙基可作为候选路线，能否落地取决于四项前置条件
一、必须守住的三条
1. 按峰值动载荷选型，不用静态均值
2. 表面电阻按场景先定区间，再选配方
3. 台架要模拟实际启停频次，不是匀速跑
二、前置条件（任一不满足则建议暂缓）
· 地面以环氧地坪或平整水泥为主
· 对噪音的容忍度可接受（非静音优先场景）
· 年用量足以摊薄专用模具
· 有实测启停频次与单轮载荷数据
三、下一步动作
1. 取现役轮，测轮径磨耗速率与压平量
2. 做压缩永久变形对比（含车间最高温）
3. 同地面同频次做一次对照台架
风险提示：本路线的主要不确定性在长时静载下的压缩变形，不在初始磨耗。
`
读者常问的三句
问：AGV 脚轮到底选尼龙还是聚氨酯？
不看材料本身，看三件事：载荷、噪音要求、要不要防静电。重载 + 要防静电，尼龙基的理由更充分；中等载荷 + 静音优先，PU 更合适。很多车可以混着用——驱动轮尼龙、从动轮 PU。
问：加了耐磨填料是不是就一定耐磨？
方向对，但不是加的越多越好。填料加多了，韧性和加工性会掉，轮缘反而更容易崩。磨耗是个体系问题，和结晶、润滑、对偶地面都有关，单靠一种填料顶不下来。
问：同一个牌号，为什么我这边磨得比别人快？
先对齐四个数：启停频次、单轮峰值载荷、地面材质、车间温度。
这四个数只要有一项差一倍，轮子寿命就可能差出几倍。
比磨耗之前先对齐工况，不然比的是两台车，不是两款料。
结语
AGV 脚轮的选材，说到底是一道工况题，不是材料题。
判断链只有三条：
启停频次定磨耗 → 载荷与温度定变形 → 场景定防静电与噪音。
三条都定完，"用什么料"这个问题自然就有答案了。
如果你手上正有一个移动机器人脚轮或驱动轮要定料，把三样东西发过来就能给方向：单轮峰值载荷、一天的启停次数、地面材质。
补一句：脚轮的账，最后是四项一起算的。
磨耗、变形、防静电、噪音，只能各让一点，换一个整体能用的组合。
没有哪一款料能四项全占。
「脚轮用什么料」这句话我们每周都听到 —— 但真答对的前提取决于启停频次和地面，不是牌号。
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。
这类件的选料与试模，可以一起聊。
Last fall, a customer who makes heavy-duty AGVs sent over four worn-out caster wheels and asked us if we could make 'more wear-resistant' modified nylon.
The edge of the wheel is no longer round, worn into small flat sections, and when pushed by hand it goes 'clunk, clunk.' I still remember his exact words on the phone: 'The same batch of material, one car ran for three years without any issue, but this one wore like this in just six months.'
He was convinced that he had bought fake material. I went to the site and took a look around and found that the problem was not with the material—it was with the start-stop frequency of that car.
This is the most common misunderstanding I have encountered over the years in making mobile robot casters: everyone verbally says they want 'wear-resistant' wheels, but what truly determines the lifespan of a wheel often isn't wear resistance, it's a few other factors. This article will break down the topic of AGV caster materials from the beginning.
1. The wheels of the AGV break down when 'stopping', not when 'running'
The casters of AGVs and AMRs are not the same as ordinary industrial casters.
The typical working condition of ordinary caster wheels is 'passive rolling'—they are pushed to move, with a single direction of force. AGVs are actively driven and experience high-frequency start-stop cycles: starting requires overcoming the inertia of the entire vehicle, and braking has to push it back. The drive wheels undergo a 'acceleration-deceleration' cycle on every trip.
This difference leads to a result: wheel wear occurs mainly at the moments of starting and stopping, not during steady-speed driving.
At the moment of starting and stopping, there is a very brief low-speed slipping process between the wheel surface and the ground. At this time, the contact point temperature rises sharply, shear stress concentrates, and a layer of the wheel surface is 'rubbed' off. The higher the frequency, the faster it gets rubbed off. The car that wears out in just half a year is one that stops at particularly dense points—each trip passes through seven or eight stations, with thousands of start-stop cycles in a single day.
So when you see 'worn flat,' don't immediately doubt the material. First, calculate one number: how many times this vehicle starts and stops in a day, how much load each wheel carries, and what the ground material is.
In one sentence: The selection of AGV caster wheels, the first variable is start-stop frequency, the second variable is the floor, and only then does the material itself come into play.
2. Six-dimensional working condition: What is clamping the AGV caster wheel
Only by setting the constraints aside can you know which items to focus on.
Load. The single-wheel load capacity of a heavy-duty AGV can reach several hundred kilograms, and it is a dynamic load—the instantaneous force on a single wheel during braking is much higher than the static load. This dimension must take the peak value, not the average.
Speed and start-stop. The maximum speed is usually not high (on the order of 1–2 m/s), but the acceleration and frequency of start-stop vary greatly. This determines the thermal load and shear on the wheel surface.
Ground. Epoxy floors, terrazzo, cement, gratings, and even workshop oil-stained floors. The roughness of the ground directly determines the wear rate; changing the floor type with the same wheel can cause the lifespan to differ by several times.
Environment. Electronics factories, lithium battery workshops, and pharmaceutical workshops require anti-static/conductive measures; cold storage needs low-temperature shock resistance; food areas also need to consider washing and cleaning agents.
Lifespan and accuracy. Lifespan is not just about 'whether it can rotate,' it also includes changes in wheel diameter—once a layer is worn down, the vehicle height changes, and the docking accuracy changes accordingly.
Compliance. The anti-static rating, flame retardant requirements, and cleanliness requirements for certain scenarios all need to have applicable standards confirmed in advance.
When the six dimensions are put together, one sees a conclusion: AGV caster wheels do not have a single champion. The most wear-resistant materials are often the hardest, and the hardest ones often have the greatest noise and are the most difficult to make anti-static.
Three, three material routes, comparing the costs side by side
| Route | Wear | Pressure Resistance / Load-bearing | Noise | Anti-static feasibility | Suitable scenarios |
|---|
| Nylon-based (PA6 / PA66 toughened wear-resistant system) | Good | Good | middle | Can be made (with conductive filler added) | Heavy load, medium-high start-stop frequency, needs anti-static |
| Polyurethane (PU) Coated Wheel | Good | middle | Low | Relatively difficult | Clean, quiet, medium load |
| Metal core with rubber coating / POM wheel | middle | Good | Medium-high | Can do | High load, low speed, not sensitive to noise |
Look at this table; the focus is not on 'which is better,' but on where the cost lies.
The advantage of nylon-based materials is that structural components themselves can be made——the wheel core and wheel surface are integrally molded, no need for rubber coating, not afraid of delamination, and can also be made anti-static and flame-retardant through formulations.
The cost is hardness. Hard means that the noise is greater than PU and it is more sensitive to ground flatness.
The advantages of PU-coated wheels are that they are soft, silent, and do not damage the floor, making them very popular in pharmaceutical and electronic cleanrooms; the drawback is that their load capacity is limited, and after long-term compression, permanent deformation can occur, resulting in 'flat spots'.
There is also a commonly overlooked approach: different wheels can use different materials. The driving wheels bear torque and the shear stress of starting and stopping, so a nylon-based material is more suitable; the driven wheels and load-bearing wheels only bear pressure, so a softer system can be used to reduce noise. A car has four wheels, and they don't necessarily need to be made of the same material.
The account of the AGV caster material should ultimately be allocated individually to the four wheels.
Check the driving wheel for shearing and start-stop, the driven wheel for compression and noise, and the load-bearing wheel for static load deformation.
A car uses only one type of material, which often means mixing four separate accounts into one.
4. Selection Criteria Table (This is the page you should most definitely save in this article)
The threshold value is a directional recommendation, not an acceptance standard—the actual values must be determined by your load, start-stop frequency, and on-site measurements.
| Indicator | Directional Threshold | Verification Method / Standard | Common Failures | Common solution | Corresponding auxiliary agent system |
|---|
| Wear Amount (on the Ground) | Refer to Akron wear test or self-built test bench comparison | GB/T 1689 / On-site Retesting | Wheel surface is ground flat, wheel diameter becomes smaller | Adjust the wear-resistant system to reduce start-stop impact | Wear-resistant packing (molybdenum disulfide / silicone oil type) |
| Compression set | Rebound after long-term load is within a reasonable range | GB/T 7759 | Flat spots on the wheel surface, vehicle body sinking | Increase crystallinity and rigidity | Nucleating agent (crystallization uniformity) |
| Gap impact (including low temperature) | Cover the minimum temperature of the usage environment | GB/T 1043 | Rim chipping, wheel core cracking | Toughening system | — |
| Surface Resistance (Anti-Static) | Set surface resistance range according to the scenario | GB/T 1410 / On-site Measurement | Static accumulation and discharge | Conductive filler system | Conductive filler |
| Flange roundness / coaxiality | Determined by piece accuracy, usually starting at the 0.05 mm level | Coordinate Measuring Machine / Roundness Tester | Driving vibrations and noise | Gate and Orientation Design | — |
| Long-term thermal-oxygen retention rate | Assessed according to the highest temperature in the workshop | ISO 527 | The surface of the wheel is pale and brittle | Stabilization system | Antioxidant |
| Resistant to oil and detergents | No abnormalities in size or appearance after soaking | Medium Soaking Test | Surface swelling and softening | Select a resistant substrate | — |
How to use this table: Do not score line by line. First, look at the second row 'Compression Permanent Deformation' — overload the AGV's wheels, first failing at 'pressed down and won't come back,' then failing at 'worn down one layer.' Many items only focus on wear, and as a result, the vehicle's body height changes after just two months of installation.
Five, five common failures and their true root causes
Failure 1: Worn flat in half a year, while another car from the same period had no issues for three years.
The root cause is usually not the material, but the frequency of start and stop and the differences in the floor. If the docking points are dense and the floor is made of diamond grit or is oily, wear will increase exponentially. Align the working conditions of the two machines first, then talk about changing the material.
Failure 2: Flat spots appear on the wheel surface, and the vehicle body sinks.
The root cause is permanent deformation due to compression, not wear. Under long-term static load (especially in high-temperature workshops), the wheel is pressed into a flat surface. The solution is to increase rigidity, add crystallization additives, or change wheels that are parked for long periods to a load-sharing structure.
Failure three: core cracking or loose insert.
The root cause lies in the structure: the wheel hub wall thickness is uneven, and the weld lines around the insert are weak areas, so the brake torque causes immediate cracking. This type of issue cannot be solved by changing materials; it requires modifying the mold and insert design.
Failure 4: The wheels from the same batch vary in color and their performance is inconsistent.
This is mostly a dispersion problem—the conductive filler or wear-resistant filler was not evenly mixed during the mixing stage. When seeing this phenomenon, first check the mixing process and masterbatching, and don't rush to change the material. The dispersion of conductive fillers is more important to monitor than color difference: uneven dispersion will cause the surface resistance to vary across different parts of the wheel.
Here's something that needs to be said directly: For troubleshooting AGV caster failures, first calculate the working conditions, then look at the structure, and only finally suspect the material. If you reverse the order, you will keep changing materials and it will keep breaking.
6. Processing and Verification: Several Things That Must Be Decided in Advance
Dry. Nylon must be baked. If the moisture content exceeds the standard, it will hydrolyze and degrade during melting. The surface may look fine, but the internal strength has already dropped. The drying window should be determined based on the actual moisture content, not by blindly following the recommended values.
Weld lines. The wheel is a thick-walled rotating body, and the location of the gate determines where the weld lines will appear. If the weld line happens to fall on the rim that experiences the greatest force, it will break there first upon impact. The gate location should be determined together with the direction of the force.
Roundness and shrinkage. The shrinkage of thick-walled parts is difficult to calculate, and fiberglass or fillers can also cause directional differences. If the wheel hub and wheel rim are molded as a single piece, the shrinkage difference will create internal stress during cooling, which later manifests as 'cracking with use'.
Thick-wall cooling. The wheel is a thick-walled rotating body, and the cooling rate difference between the inside and outside is large.
If the mold temperature is too high, it is prone to shrinkage holes; if it is too low, internal stress remains. The window is much narrower than the thin-walled part.
During mold testing, the mold temperature should be adjusted into several curves and tested one round each; don't just test a single point.
Verification order. It is recommended to arrange it like this, do not change the order:
1. Material grade: wear, permanent compression deformation, impact, surface resistance
2. Part level: roundness, insert pull-out, static load flattening test
3. Test Bench: Roller test simulating actual start-stop frequency
4. Complete vehicle: Run the full working condition on actual ground and remeasure wheel diameter changes
5. Environmental overlay: temperature oil contamination long-term static load, the last item is the easiest to be overlooked
7. Boundaries: When This Matter Should Not Be Discussed
In the following four situations, it is not recommended to proceed with this part using modified nylon:
First, parts that are soaked for a long time in strong solvents or high-concentration oil stains. Nylon has limits in chemical resistance; exceeding them will cause swelling and deformation. Such parts should be replaced with metal or special elastomers.
Secondly, it requires extremely low-noise clean environments. Nylon is relatively hard, and its noise level is naturally higher than PU coating. For situations that require quietness, using hard nylon is spending the cost in the wrong direction.
Thirdly, in situations where the ground is uneven and there are high-speed operation requirements. Hard wheels are sensitive to ground undulations, have large vibrations, loud noise, and short lifespan, making this type more suitable for soft wheels or pneumatic structures.
Fourth, the annual usage is too small to justify the tooling costs. This part usually requires a dedicated mold, and if the usage is too low, it is not economically feasible.
Fifth, long-term work in high-temperature workshops and prolonged static docking. High temperature combined with static load will significantly accelerate permanent compression deformation.
For this type of part, you either increase its rigidity or add support groupings at the docking position to share the load; simply changing the material won't solve the problem.
Writing these five points at the beginning is not to discourage, but to save time. I have seen quite a few projects go smoothly at the sample stage, only to find out the direction is wrong during vehicle installation—the cost of going back is much higher than not doing it in the first place.
Material Change Risk List (From the original plan to modified nylon casters, things that need to be changed)
| link; segment; part | What do you want to move? | Points that are easy to overlook |
|---|
| Mold | The flange thickness and gate position are redefined according to the force applied | Maintain the original PU wheel's structural thickness |
| Dry | Determine the window based on the measured moisture content | Recycled materials mixed with the water content brought in |
| 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 walls shrink heavily, and the holding pressure curve needs to be readjusted. | Demolding too early causes internal stress |
| Insert | The weld lines around the insert should avoid the stress area | Insert Preheating and Encapsulation Temperature |
| Color difference | The color of the conductive filler parts is relatively dark and there are variations between batches | Appearance requirements to be relaxed in advance or set separately |
| Verification order | Material → Component → Test Bench → Complete Vehicle → Environment | Skip the test bench and install directly on the vehicle |
One-page report sheet (for people who need to report upwards)
`
Project: AGV / AMR Casters · Material Route Assessment
Conclusion direction: Modified nylon-based can be considered as a candidate route, and whether it can be implemented depends on four prerequisite conditions
1. Three Rules That Must Be Followed
1. Select based on peak dynamic load, not static average.
2. Determine the range of surface resistance according to the scenario, then select the formula.
3. The test bench should simulate the actual start-stop frequency, not run at a constant speed.
2. Prerequisites (if none are met, postponement is recommended)
· Floor mainly made of epoxy flooring or leveled cement
· Noise tolerance acceptable (non-silent priority scenarios)
· Annual usage sufficient to spread specialized molds
· Actual measured start/stop frequency and single-wheel load data
3. Next steps
1. Take an active wheel, measure wheel diameter wear rate and flattening amount
2. Perform compression permanent deformation comparison (including workshop maximum temperature)
3. Perform a comparison test at the same floor frequency with the same test frame
Risk warning: The main uncertainty of this route lies in compression deformation under long-term static load, not initial wear.
'
Three Questions Readers Often Ask
Question: Should AGV casters be made of nylon or polyurethane?
Don't look at the material itself, but look at three things: load, noise requirements, and whether anti-static is needed. Heavy load + anti-static—nylon-based is a more reasonable reason; Medium load + prioritize quietness, PU is more suitable. Many cars can mix and match—drive wheel nylon, driven pulley PU.
Question: Does adding wear-resistant filler mean it's definitely wear-resistant?
The direction is correct, but more isn't always better. If too much filler is added, toughness and workability decrease, and the rim is more likely to crack. Wear is a systemic issue, related to crystallization, lubrication, and paired flooring. One filler alone can't handle it.
Question: Why does my side wear faster than others for the same grade?
First, align four numbers: start-stop frequency, peak load on a single wheel, ground material, and workshop temperature.
If any one of these four numbers differs by a factor of two, the wheel's lifespan can differ by several times.
Before comparing wear, first align the working conditions; otherwise, you're comparing two cars, not two materials.
Conclusion
AGV Ultimately, caster material selection is a working condition problem, not a material problem.
There are only three judgment chains:
Start and stop frequency determines wear, → load and temperature determine deformation, → Scenario determines static electricity and noise.
Once all three are set, the question of "what materials to use" naturally comes to an answer.
If you have a mobile robot caster or drive wheel to specify, send over three items and you can give direction: single-wheel peak load, daily start/stop count, and floor material.
Just to add: the caster count is calculated all four at once.
Wear, deformation, anti-static, noise—only allow a little of each, and switch to a combination that works as a whole.
No material can cover all four aspects.
We hear the question "What materials are used for casters?" — but the real answer depends on the frequency of start/stop and the floor, not the grade.
What we do is very specific: converting resins like PA6, PA66, PA46, PA11, PA12, PA6T, PA9T, and nylon alloys into a truly usable product; We also do modified PPO, PPS, and thermoplastic elastomers.
Also deals in nylon resin, sub-brand materials, and bulk materials from major chemical giants, and long-term collection of nylon raw materials, sprue return materials, and various nylon scraps, with official disposal channels.
For material selection and mold trials for these types of parts, we can chat together
