"这批件装不上去。"
听到这句话，多数人的第一反应是模具不准、公差没控好。
但尼龙件的尺寸问题，模具往往不是主因。
更常见的原因是材料吸湿、玻纤取向、以及成型条件漂移。而这些问题的排查方式，和改模完全不同——先改模，等于把钱花在了错误的地方。
装配尺寸超差的排查，通常从一条产线停线消息开始。
车门模块装配工位报干涉，件和图纸都对，装不进去。
质量部门把供应商、模具、来料检验挨个问了一遍，人人都有数据，人人都说自己没问题。
最后把三个批次的件放到恒温间里泡了两天，干涉消失了。
尺寸跟着湿度走，这件事写进了那家厂的进料检验规程。
一、三类原因，先分开
尼龙件的尺寸超差，可以归到三个大类：
| 类别 | 典型原因 | 特征 |
|---|
| 材料类 | 吸湿膨胀、玻纤取向收缩 | 尺寸随时间变化；不同方向差异大 |
| 模具类 | 型腔尺寸、收缩率补偿不当 | 尺寸稳定，但整体偏移 |
| 工艺类 | 保压、模温、冷却不均 | 批次之间波动；穴间差异 |
分清的标志是"尺寸是稳定的偏，还是变化的偏"：
稳定地偏一个固定值 → 看模具与收缩率补偿。
随着时间变化 → 看吸湿。
批次之间忽大忽小 → 看工艺。
方向性差异明显（流动方向 vs 垂直方向） → 看玻纤取向。
二、吸湿：最被忽略、也最常见
尼龙是吸湿性材料。吸水之后会膨胀，尺寸跟着变大。
这个规律可以粗略估：吸水率每增加 1%，尺寸大约变化 0.2-0.3%。 这个量级已经超过很多装配公差。
关键在于：吸湿不是瞬间完成的，是几周甚至几个月的过程。
所以会出现一个很典型的现象：出厂检验合格，装车一段时间后装配间隙变了。 这不是磨损，是吸湿。
还有一个常见误解：以为只有 PA6、PA66 吸湿严重，长碳链就没事。长碳链（PA11、PA12）吸水率确实低得多，但不是零，精密件上同样要算。
处理办法是调湿：在装配前让件先吸到接近使用环境的平衡含水率。 这样尺寸"提前变完"，装机后就不再飘。
调湿时间要按最厚截面算——水分的扩散时间大致与壁厚的平方相关，壁厚翻倍，吸透时间明显拉长。
三、玻纤取向：方向性差异的根源
玻纤增强件的收缩率是各向异性的：
沿流动方向：玻纤取向，收缩率小
垂直流动方向：收缩率大
结果是同一个件的不同方向，尺寸变化不一样。表现出来就是翘曲或者某个方向的尺寸超差。
这里有一条很实用的判据：
翘曲的方向，通常和熔体流动方向一致。 看到翘曲，先查浇口位置和流道布局，不要急着换料。这多半是工艺与模具问题，换料治不好。
四、模具与工艺：什么时候才该怀疑它们
该怀疑模具的情况：
尺寸稳定地偏一个固定量
所有方向都偏，没有明显规律
试模时的尺寸和量产一致，都在同一个偏差上
该怀疑工艺的情况：
同一模具、同一材料，批次之间波动
一模多穴之间尺寸不一致
换了机台或干燥条件后尺寸变了
其中一模多穴的穴间差异在小型件上尤其常见：流道长度不同、冷却条件不同、保压传递不均，都会让不同穴的收缩率不一样。
五、排查顺序：从便宜到贵
顺序很重要，先做不用改模的那些动作：
第一步：复测尺寸，区分"稳定偏"还是"变化偏"。 这一步零成本，但能直接分流出方向。
第二步：查干燥与含水率。 干燥不到位不仅影响力学性能，也影响尺寸稳定性。
第三步：做吸湿平衡试验。 把件放在恒温恒湿环境里放到平衡，再复测尺寸。这一步能确认是不是吸湿导致。
第四步：查模温与保压的一致性。 包括多穴之间、不同批次之间。
第五步：才轮到核模具型腔尺寸与收缩率补偿。
顺序颠倒，最常见的后果是：改了模，问题还在，而且多了一个变量。
六、预防比排查更划算
几件在设计阶段就能定的事：
1. 明确使用环境的湿度，按平衡吸湿量预留尺寸余量
2. 把调湿写进工艺文件，不要靠自然放置
3. 浇口位置与流道平衡在开模前定好，尽量让流动对称
4. 公差设定时区分方向，流动方向与垂直方向的公差不能一刀切
再加一条设计端建议：图纸上标注尺寸时，把"测量条件"一并写清楚——是成型后立即测，还是吸湿平衡后测，还是装配状态下测。
同一个件，这几个状态下能差出好几个丝。标注不清，供需双方就会各测各的，各说各的合格。 很多所谓"供应商尺寸不合格"的争议，其实出在测量条件没对齐。
尺寸超差的排查，先把尺寸的来源分层。
注塑出模的尺寸是第一层，冷却收缩定型决定。
吸湿之后的尺寸是第二层，PA 系材料吸湿会膨胀，百分之一含水量能带来可观的尺寸变化。
后收缩是第三层，出模后几天的时效收缩，退火与否差别明显。
排查时要按层取证：出模尺寸、调湿后尺寸、装机时尺寸，三个点各留记录。
三层数据一摆，超差发生在哪一层，一目了然。
大多数扯皮的根源，是三层数据从来没分开记过。
追问一：吸湿带来的尺寸变化有多大？
PA66 在标准湿态下体积变化可到百分之零点几，对精密件是显著的。关键配合尺寸要给出干态、湿态两个名义值，公差带覆盖两者之间。设计图纸上只标一个值的，等于把问题留给装配线。
追问二：退火对后收缩有用吗？
有用，但要选对时机和温度。退火能加速时效收缩、稳定内应力，温度过高会诱发新的变形。批量件建议先做退火工艺验证，把退火参数和尺寸稳定曲线一起固化进工艺文件。
一单装配干涉的追查
车门模块干涉，三方数据都合格。最后按温度湿度分层对比：出模尺寸全部合格，库房存放两周后的尺寸漂了。仓库靠窗，夏季高温高湿，件在库房里就已经调湿了。整改是库房分区控湿，配合尺寸按调湿态复核。存放条件进了选型资料，这是那单最大的收获。
尺寸超差排查流程图
出模尺寸→存放条件→调湿状态→装配工艺→量具一致性，五层按序排查，每层留数据。五层走完还没答案的，回头查模具磨损。顺序别乱，乱了就是三方扯皮。
尺寸问题的管理，最后要落在制度上。
来料检验的尺寸项目，按干态还是湿态，要有明文。
仓储的温湿度范围，要有边界和记录。
装配线的环境，超范围时的处理流程要预先写好。
三件事写在文件里，尺寸问题的处理就从扯皮变成执行。
一家做座椅的工厂设了调湿态检验室，来料按调湿态复测关键尺寸。
第一年觉得多此一举，第二年梅雨季就成了全厂唯一不停线的装配线。
制度的价值不在平时，在极端季节里。
三个延伸问题
量具本身要不要控温？要，量具和被测件的温差会引入误差，标准做法是同温放置后测量。
塑料件的测量基准怎么定？按图纸基准，但夹持力要统一，夹太紧薄壁件会变形出假数据。
超差件能不能让步接收？按功能影响评估，配合尺寸让步前先做装配试装验证，不能凭感觉放行。
制度四件套
检验状态明文、仓储环境边界、装配环境流程、量具同温规范。
四件套齐了，尺寸扯皮概率断崖式下降。
尺寸管理的进阶版是数字化。
把关键尺寸做成测量数据流，每批自动入库。
数据流跑起来之后，尺寸漂移在图表上一眼可见。
趋势比单点值有信息量：缓慢漂移指向模具磨损，阶跃漂移指向换料或工艺变更。
一家内饰件厂靠数据流把尺寸投诉提前两周预警。
预警的价值不是修件，是留出换工艺的时间窗。
最后一组追问
数据流要投入多少？测量设备已有，加一套数据采集和看板，投入不大。
小厂做不了数字化怎么办？先做手抄趋势表，每周画折线，效果就有七八成。
吸湿件的测量要在什么状态下？约定一个状态点：调湿后测，或者出模几小时内测，前后必须一致。
尺寸管理的终点不是测得准，是看得早。
尺寸这一篇的最后，讲讲和供应商的相处方式。
尺寸争议最忌讳各说各话，数据口径要先统一。
建议和供应商约定统一的测量状态、量具和方法。
口径统一之后，九成的争议会在第一次对数时消解。
剩下的一成，再按流程逐层排查。
收官三点
尺寸问题多数不是谁错，是谁都没说清状态。
三层数据分开记，是尺寸管理的第一课。
把口径写进协议，是让专业代替情绪的最短路径。
尺寸排查这一篇，最后补充一个沟通模板。
三方对数会之前，各自把数据按统一模板填好。
模板字段：测量状态、环境温湿度、量具编号、测点位置、数值。
一次对数会把原来三天的争议压缩到一个下午。
模板本身不值钱，值钱的是所有人都按它来。
管理尺寸争议，管理的是人的口径。
这一篇的完整知识地图
出模、存放、调湿三层定数据来源，五层排查定责任路径，口径统一定沟通效率，制度四件套定长期秩序，趋势预警定响应速度。
尺寸管理的全部内容，就是让每个尺寸都有身份、有历史、有出处。
超差不可怕，说不清的超差才可怕。
把这一篇交给新来的质量工程师，能省他两年摸索。
尺寸排查还有一类高频场景要单独讲：季节性批量超差。夏季高温高湿来临时，某几批件的配合尺寸集体漂移，检验记录却都合格。原因是检验环境的温湿度没控住，量具和件的膨胀方向不一致。对策是检验室加装控温控湿，超范围时数据标注环境条件。一家厂在检验记录里加了环境栏之后，季节性争议直接消失，因为数据自己把原因说清楚了。环境信息进记录，是尺寸管理里成本最低、见效最快的一个动作。
再补一个退火件的处理细节。退火后的件要在干燥环境里冷却到室温再包装，带着余热封装会在袋子里凝露，吸湿从此开始。有工厂的调湿态数据异常，追查发现是包装环节的凝露问题。工艺链上每一环都可能给数据埋雷，排查时要有全链视角。包装这个最不起眼的环节，恰恰是很多尺寸谜案的终点站。
尺寸排查再讲一个换季场景的应对。春秋换季时仓库和产线的温湿度都在变，同一批件早上和下午的测量值都可能不同。检验规程里因此要写测量时间窗：出模后几小时内测，或者恒温放置后测。时间窗一固定，数据就稳定了。有工厂的质检员换班时间不同，测出的尺寸带着班次特征，追查半个月的谜题，答案就藏在排班表里。数据背后的变量清单越全，谜题越少。
补一个和研发协同的做法。设计变更时同步更新关键尺寸清单，新尺寸新状态新测法三件套跟着变更走。有工厂吃过亏：设计改了配合尺寸，检验规程没跟上，旧测点测了个寂寞，超差件一路放行到装配线。变更管理的最后一米，永远落在检验文件上。这一米走通了，尺寸体系才算闭环。走不通，前面所有的精度努力都会在某个变更节点上归零。
结语
尼龙件尺寸超差的排查链：
先分清稳定偏还是变化偏 → 再排吸湿与取向 → 最后才查模具。
如果你手上有一批件正在尺寸问题上纠缠，把三样东西发过来：超差的方向与量值、是出厂就不合格还是用一段时间后变化、使用环境湿度。
This batch of pieces can't be loaded on.
Upon hearing this sentence, most people's first reaction is that the mold is inaccurate or the tolerances are not well controlled.
But for the size issues of nylon parts, the mold is often not the main cause.
More common reasons are material moisture absorption, fiberglass orientation, and shifts in molding conditions. The way to troubleshoot these issues is completely different from modifying the mold — changing the mold first is like spending money in the wrong place.
The inspection of assembly dimension deviations usually starts with a line stoppage message from a production line.
The car door module assembly station reports interference; both the part and the drawing are correct, but it can't be installed.
The quality department questioned the suppliers, molds, and incoming material inspections one by one; everyone had data, and everyone said they had no problems.
Finally, I put the items from the three batches into the constant temperature room to soak for two days, and the interference disappeared.
Size changes with humidity; this matter was written into the incoming material inspection procedures of that factory.
1. Three types of causes, separate them first
The dimensional deviation of nylon parts can be classified into three major categories:
| Category | Typical cause | Feature |
|---|
| Materials | Moisture absorption expansion and fiberglass orientation shrinkage | Size changes over time; large differences in different directions |
| Mold category | Improper cavity size and shrinkage compensation | Dimensionally stable, but overall shifted |
| Crafts | Pressure holding, mold temperature, uneven cooling | Fluctuations between batches; differences between wells |
The distinguishing sign is whether the 'size' tends to be stable or tends to change:
Stably偏向 a fixed value → look at the mold and shrinkage compensation.
As time changes → watch for moisture absorption.
The batch sizes fluctuate → Check the process.
Directional differences are obvious (flow direction vs vertical direction) → look at the glass fiber orientation.
2. Moisture absorption: Most overlooked, yet most common
Nylon is a moisture-absorbing material. It swells after absorbing water, and its size increases accordingly.
This rule of thumb can be roughly estimated: for every 1% increase in water absorption, the dimensions change by about 0.2-0.3%. This magnitude already exceeds many assembly tolerances.
The key point is: moisture absorption does not happen instantly; it is a process that takes weeks or even months.
So a very typical phenomenon appears: the product passes factory inspection, but after being loaded onto a vehicle for a period of time, the assembly gaps change. This is not wear; it is moisture absorption.
There is another common misconception: people think that only PA6 and PA66 are highly hygroscopic, and long-chain nylons are fine. Long-chain nylons (PA11, PA12) indeed have much lower water absorption, but it is not zero, and it must also be considered for precision parts.
The method of handling is moisture adjustment: before assembly, let the parts absorb moisture until they reach the equilibrium moisture content close to the intended usage environment. This way, the dimensions will "change in advance," and after installation, they will no longer fluctuate.
The humidification time should be calculated based on the thickest section — the diffusion time of moisture is roughly proportional to the square of the wall thickness. If the wall thickness doubles, the saturation time increases significantly.
3. Glass Fiber Orientation: The Root of Directional Differences
The shrinkage of glass fiber reinforced components is anisotropic:
In the flow direction: fiberglass orientation, small shrinkage
Vertical flow direction: high shrinkage
The result is that different directions of the same part have different dimensional changes. This manifests as warping or dimensional deviations in a certain direction.
Here is a very practical criterion:
The direction of warpage usually aligns with the flow direction of the melt. When you notice warpage, first check the gate position and runner layout, and don't rush to change the material. This is mostly an issue with the process and mold, and changing the material won't solve it.
4. Molds and Processes: When Should You Start Suspecting Them
Situations in which the mold should be suspected:
Dimensionally stablely offset by a fixed amount
All directions are off, with no obvious pattern
The dimensions during mold testing are consistent with mass production, all within the same deviation.
Situations where the process should be suspected:
Same mold, same material, fluctuations between batches
The dimensions between multiple cavities in the same mold are inconsistent
The dimensions changed after switching the machine or drying conditions.
The differences between cavities in multi-cavity molds are particularly common in small parts: differences in runner length, cooling conditions, and uneven pressure transfer can all cause different cavities to have different shrinkage rates.
5. Inspection order: from cheap to expensive
The order is important; start with the actions that do not require changing the mold:
Step 1: Re-measure the dimensions to differentiate between 'stable deviation' and 'variable deviation.' This step has zero cost, but it can directly guide the direction.
Step 2: Check dryness and moisture content. Inadequate drying not only affects mechanical properties but also affects dimensional stability.
Step 3: Conduct a moisture absorption equilibrium test. Place the parts in a constant temperature and humidity environment until equilibrium is reached, then re-measure the dimensions. This step can confirm whether moisture absorption is the cause.
Step 4: Check the consistency of mold temperature and holding pressure, including between multiple cavities and between different batches.
Step 5: It's time to compensate for the cavity dimensions of the core mold and the shrinkage rate.
The order is reversed, and the most common consequence is: the mold is changed, the problem still exists, and there is an additional variable.
6. Prevention is more cost-effective than screening
A few things that can be decided during the design phase:
1. Clearly determine the humidity of the usage environment and reserve dimensional allowances according to the equilibrium moisture content.
2. Include humidity adjustment in the process documentation, do not rely on natural placement
3. Gate location and runner balance should be set before mold opening, trying to make the flow symmetrical.
4. When setting tolerances, distinguish the directions; the tolerances for the flow direction and the perpendicular direction cannot be uniformly applied.
One more design suggestion: when marking dimensions on the drawings, clearly specify the 'measurement conditions'—whether it is measured immediately after molding, after moisture equilibrium, or in the assembled state.
For the same part, there can be several microns of difference under these different conditions. If the markings are unclear, both the supplier and the buyer will measure it their own way and each will claim their results are qualified. Many so-called 'supplier dimension nonconformity' disputes actually arise from misaligned measurement conditions.
For troubleshooting oversized dimensions, first categorize the sources of the dimensions.
The dimensions of the injection-molded part are determined in the first layer, which is defined by cooling shrinkage and setting.
The size after moisture absorption is the second layer. PA series materials will expand when absorbing moisture, and a water content of one percent can bring about considerable dimensional changes.
Post-shrinkage is the third stage, the aging shrinkage that occurs a few days after demolding, with a significant difference whether annealed or not.
During the inspection, evidence should be collected layer by layer: record the dimensions after molding, the dimensions after humidity adjustment, and the dimensions at the time of installation, keeping a record for each of the three points.
With the three layers of data laid out, it’s immediately clear at which layer the deviation occurred.
The root of most bickering is that the three layers of data have never been recorded separately.
Follow-up Question 1: How significant are the dimensional changes caused by moisture absorption?
PA66 Under standard wet conditions, volume change can reach several tenths of a percent, which is significant for precision parts. The key fit dimensions must specify two nominal values: dry and wet, with tolerance zones covering between them. If only one value is marked on the design drawing, it means leaving the problem to the assembly line.
Follow-up Question 2: Is annealing useful for post-shrinkage?
Yes, but the timing and temperature must be chosen correctly. Annealing accelerates aging shrinkage and stabilizes internal stress; excessively high temperatures can trigger new deformations. For batch parts, it is recommended to verify the annealing process first, solidifying both annealing parameters and dimensional stability curves into the process documents.
Tracking Interference in One Assembly Order
Door Module Interference, all three parties are qualified. Finally, layered comparison by temperature and humidity: all outputting dimensions were qualified, but after two weeks of storage, the dimensions had drifted. The warehouse is near the window, with high temperature and humidity in summer, so parts were already humidified in the warehouse. Rectification involved controlling humidity in the warehouse zones, and checking the dimensions according to humidity conditioning. Storage conditions were included in the selection materials, which was the biggest takeaway from that order.
Dimensional Over-Tolerance Investigation Flowchart
Mold Size → Storage Conditions→ Humidity Control Status→ Assembly Process → Measuring Tool Consistency, Five Layers Checked in Sequence, Data Kept on Each Layer. If you still have no answer after five layers, go back to check mold wear. Don't mess up the order; if it gets messed up, it's just a three-party dispute.
Managing Size Issues Ultimately, Rules Down to Regulations.
For incoming material inspection dimensions, whether dry or wet must be clearly documented.
The temperature and humidity range in warehousing must have boundaries and records.
The environment of the assembly line should be written in advance for handling cases that exceed the range.
Three things are written in the file, so handling size issues shifts from arguing to execution.
A factory that makes seats set up a humidity control testing lab, re-testing key dimensions according to the humidity conditioning state of incoming materials.
The first year felt unnecessary, but the second year, during the rainy season, it became the only assembly line in the entire factory that ran non-stop.
The value of the system lies not in normal times, but in extreme seasons.
Three extended questions
Should the measuring tool itself be temperature-controlled? Yes, the temperature difference between the measuring instrument and the part being measured will introduce errors. The standard practice is to measure at the same temperature.
How to determine the measurement reference for plastic parts? According to the reference drawn in the drawings, but the clamping force must be consistent; if the clamping is too tight, thin-walled parts will deform and produce false data.
Can over-tolerance parts be accepted? Evaluate by function and impact, and before conceding dimensions, conduct assembly trial installation verification; do not let it pass based on intuition.
Four-piece system set
Inspection status explicit, warehouse environment boundaries, assembly environment flow, measuring tool temperature standards.
With all four pieces set, the probability of size disputes drops sharply.
The advanced form of size management is digitalization.
Turn key dimensions into measurement data streams, with each batch automatically stored in storage.
Once the data stream runs, dimensional drift becomes obvious on the chart.
Trends are more informative than single point values: slow drift points to mold wear, step drift points to material changes or process changes.
An interior parts factory relies on data streams to issue two weeks of advance warning about size complaints.
The value of early warnings is not part repairs, but leaving a window for process changes.
Last group follow-up question
How much investment is needed for data streaming? Measuring equipment is already available, plus a set of data collection and dashboards, so the investment is not large.
What if small factories can't do digitalization? First, make a handwritten trend chart and draw line lines weekly; the effect is 70-80%.
Under what conditions should moisture-absorbing parts be measured? Set a state point: measure after humidity adjustment, or test within a few hours after molding; the front and back must be consistent.
The end point of dimension management is not accurate measurement, but early observation.
At the end of this dimension article, talk about how to interact with suppliers.
The biggest taboo in size disputes is having your own arguments; data standards must be unified first.
It is recommended to agree with suppliers on unified measurement status, measuring instruments, and methods.
After standardizing the measurements, ninety percent of disputes will be resolved by the first logarithm.
The remaining 10% should be checked layer by layer according to the process.
Final Three Points
Most size issues are not about who is at fault; it's that no one clearly explained the status.
Recording the three layers of data separately is the first lesson in size management.
Writing the calibration into the agreement is the shortest way for professionals to replace emotions.
For this article on size inspection, I will add a communication template at the end.
Before the three-party logarithm meeting, each should fill in the data according to a unified template.
Template fields: measurement status, ambient temperature and humidity, measuring instrument number, measurement point location, value.
A single logarithm will compress the disputes that had been over three days into one afternoon.
The template itself is not valuable; what matters is that everyone follows it.
Managing size disputes is managing the caliber of people.
The complete knowledge map of this article
Mold extraction, storage, and humidity regulation define data sources at three layers, responsibility paths at five levels, communication efficiency on unified standards, four-piece system sets long-term order, trend warnings determine response speed.
The entire content of dimensional management is to ensure every size has an identity, history, and origin.
Over-tolerance is not scary; what is inexplicable is what is.
Give this article to a new quality engineer and save him two years of trial and error.
There is another high-frequency scenario to discuss separately: seasonal batch over-tolerance. When summer heat and humidity arrive, the matching dimensions of certain batches of parts drift collectively, but the inspection records all pass. The reason is that the temperature and humidity of the inspection environment are not controlled, and the expansion direction of the measuring tool and the part is not aligned. The countermeasure is to install temperature and humidity control in the laboratory, and when the scope is exceeded, the data is labeled as environmental conditions. After a factory added an environment column to its inspection records, seasonal disputes disappeared because the data itself clearly explained the reasons. Recording environmental information is the lowest-cost and fastest action in size management.
Add one more detail on handling annealed parts. Annealed parts must be cooled to room temperature in a dry environment before packaging, and the residual heat will condense inside the bag, starting moisture absorption. Some factories have abnormal humidity control data, and investigation reveals condensation issues in the packaging stage. Every link in the process chain can plant data pitfalls, so investigations require a full-chain perspective. Packaging, the most inconspicuous step, is precisely the endpoint of many dimensional mysteries.
Size inspection also talks about handling seasonal change scenarios. During spring and autumn transitions, warehouse and production line temperature and humidity change, and measurements of the same batch in the morning and afternoon may differ. Therefore, inspection procedures specify measurement time windows: test within a few hours after mold release, or test after constant temperature storage. Once the time window is fixed, the data stabilizes. Some factories have quality inspectors with different shift changes, measuring dimensions with shift characteristics, and after half a month of puzzles, the answers are hidden in the schedule. The more complete the variable list behind the data, the fewer puzzles there are.
adds a method of collaborating with R&D. When design changes, the key dimension list is updated simultaneously, and the new dimensions, new status, new measurement method, three-piece set follows the change. Some factories have suffered losses: design changes to matching dimensions, inspection procedures don't keep up, old measurement points are measured, and over-tolerance parts are released all the way to the assembly line. The last meter of change management always falls on the inspection documents. Once this meter is fully implemented, the dimensional system is considered closed. If it doesn't work, all previous precision efforts will reset to zero at some change point.
Conclusion
Nylon part dimension over-tolerance troubleshooting chain:
First distinguish between stable and variable bias → then eliminate moisture absorption and orientation → Finally, check the mold
If you have a batch of parts that are giving you trouble with size issues, send over three things: the direction and magnitude of the deviation, whether they were nonconforming from the factory or changed after some use, and the humidity of the usage environment.
