前两天接了个电话，一家做背板连接器的客户，拿着一颗改性尼龙注塑的绝缘体来问。
是高速背板的胶芯，针脚密密麻麻排了两排。他说："背板跑着跑着误码率上来了，是不是胶芯变形了？"
我们没先答变形，先问了两件事：针脚共面度怎么验的，机房跑了多久。答案回来——共面度按干态测的，跑了一年。
这件事把信号连接器胶芯选料的核心说透了：它不扛大电压，却对"针脚齐不齐、表面脏不脏"极其敏感。 这一件要的从来不是 CTI 多高，是共面度和低析出两手都稳。
一、胶芯这一件，干的是"定位"不是"导电"
先说清胶芯在连接器里的角色。
导电靠针脚和端子，胶芯（绝缘体）干的是三件事：把针脚按精确位置定住、把相邻信号隔开、把外部应力和介质挡住。 它不导电，但信号的成败常常捏在它手里。
背板连接器把板卡插到背板上，针脚几十到几百根；高速连接器跑 25G、56G 甚至更高速率。这类件对胶芯的要求和高压连接器完全不同——
高压件看 CTI、看电弧；信号件看共面度、看析出、看介电。 把高压件的判据套到信号胶芯上，会漏掉真正要命的两项。
胶芯的精度不在"能不能绝缘"，在"针脚站得正不正"。
一颗密排胶芯上，针脚间距可能只有零点几毫米。
整排针脚里哪怕几根偏出公差，信号就断续。
所以胶芯的验收，先看共面度，再看别的。
一句话：信号连接器胶芯，CTI 不是主场，共面度和低析出才是。
二、工况六维：胶芯被什么约束
按六个维度摊开。
温度。 背板靠近芯片，长期 85–105℃ 常见，局部更高。比机房环境温度高，热老化和尺寸漂移要算。
信号与载荷。 不扛大电压，但针脚有插拔力和保持力，胶芯要抗蠕变不让针脚松。高速信号对介电常数和损耗敏感。
介质。 不直接泡液，但有机房空气、凝露、灰尘。低析出是重点——析出物沾到触点就麻烦。
寿命。 十年起步，插拔按千次。判据是共面度保持率和析出程度，不是初始强度。
外观与洁净。 低烟低毒、低黄变、低析出，机房环境要求。
合规。 阻燃 UL94 V0，多数无卤；信号件对 CTI 要求相对低，但靠近电源的位置仍要看。
六维里温度、信号速率、寿命都给了具体数字，件级精度就在这。
三、三条材料路线，共面度和析出要分开算
把胶芯候选路线并排放，看"共面度"和"析出"两列怎么走。
| 路线 | 组成 | 共面度 / 尺寸 | 析出与介电 | 代价 |
|---|
| PA66-GF | 脂肪族玻纤增强 | 吸湿翘曲、湿态漂 | 常规，介电一般 | 便宜、工艺熟 |
| PA6T/PA9T-GF | 半芳香族玻纤增强 | 低吸湿、尺寸最稳 | 低析出、低介电优 | 贵、加工窗口窄 |
| PA46-GF | 高温脂肪族 | 耐温好、吸湿仍高 | 常规 | 介于两者之间 |
三条路线没有"谁更好"，只有"哪条线更紧"。
PA66-GF 是通用主力，便宜好打，但吸湿 8%、各向异性翘曲，共面度湿态会漂。对针脚间距 0.4–2.0 mm 的密排胶芯，这点漂移就足以让部分针脚虚接。
PA6T/PA9T-GF 吸湿低、尺寸稳、介电常数和损耗因子更友好，是高速胶芯的常见正解。代价是单价和加工难度——熔点高、料温模温都要跟着调。
PA46-GF 耐温好，但吸湿仍偏高，共面度账和 PA66 类似。
一句话：信号胶芯的贵料，买的不是强度，是"针脚十年还齐、表面十年还干净"两样同时达标。
四、选型判据表：共面度和析出是主角
把约束落成可核对指标。下表门限是方向性建议，不是验收标准——实际由项目、工况和实测定。
| 指标 | 方向性门限 | 验证方法 / 标准 | 常见失效 | 通行解法 | 对应助剂体系 |
|---|
| 共面度（针脚平面） | 按 pitch 定，常 ≤ 0.05–0.1 mm | 三坐标 / 共面度仪 | 虚接、误码 | 低翘曲 + 浇口优化 | 成核剂（结晶控制） |
| 薄壁充填 | 满注无短射、无熔接线弱 | 短射 + X 光 | 局部缺料、强度弱 | 高流动体系 | 润滑剂（薄壁流动） |
| 吸湿尺寸稳定 | 湿态与干态差可控 | ISO 62 + 量测 | 共面度漂移 | 低吸湿基材 | 材料本征，不靠助剂 |
| 析出（触点污染） | 浸泡 / 老化后表面无迁移物 | 老化 + 表面分析 | 接触电阻升、腐蚀 | 低迁移配方 | 润滑剂（低析出选型） |
| 阻燃 | UL94 V0，优先无卤 | UL94 / GWIT | 安规返工 | 无卤阻燃体系 | — |
| 介电（高速件） | Dk / Df 按速率定 | IPC / 谐振腔法 | 信号衰减、串扰 | 低介电基材 | 材料本征，不靠助剂 |
| 长期耐温 | 105℃×长期 保持 | ISO 527 | 发白、脆化 | 热稳定体系 | 抗氧剂（热氧） |
| 熔接线强度 | 针脚位置不落在熔接线 | 短射 + 力学测试 | 局部开裂 | 浇口与流道优化 | 偶联剂（界面） |
| 灼热丝 / 安规 | 按场景定 GWIT | GWIT / GWFI | 安规返工 | 阻燃协效体系 | 阻燃剂（协效） |
怎么用这张表：共面度一行和析出一行分开打。只盯阻燃不盯共面度，胶芯 V0 过了，背板照样误码。 高速件还要把介电那行补上。
一个提醒：共面度这项，必须用调湿态数据定设计。PA66 吸湿 8%，胶芯吸湿后针脚整体偏移，共面度掉档——用干态测的共面度定装配，机房跑一年湿度上来，针脚就虚了。
五、五条常见误判，和真实根因
判反一：胶芯只看阻燃 V0。
信号连接器询盘里最常见的开场是"要 V0 的料"。问得对，但只问了一半。胶芯 V0 过了，共面度漂了、析出脏了，背板照样误码或接触不良。 我们接胶芯单，一定把共面度和析出和阻燃并列提，少一项都不定型。
判反二：把析出当成基材的锅。
客户发现触点接触电阻升了，第一反应是"料不好"。很多时候根因在配方里的低分子助剂——润滑剂加过量、或选了耐温不够的品种，长期热下往表面迁移，沾到金触点形成绝缘或腐蚀层。 换基材不换配方，照样析。看到析出，先查助剂清单和老化后的表面成分，别急着换料。
判反三：用干态共面度定设计。
尼龙吸湿会涨会翘。PA66 吸湿 8%，密排胶芯吸湿后整体偏移，共面度掉档。用干态测的共面度定装配，湿态就虚接。 共面度和尺寸都要按调湿态留余量。
判反四：拿 PA66 的工艺打半芳香族。
PA6T/PA9T 熔点 300℃ 以上，料温模温干燥和 PA66 不是一个窗口。工艺不换，薄壁填不满、内应力大，共面度和析出都受影响。同牌号不同工艺，实测表现能差出一档。
判反五：把薄壁短射当成材料流动性不够。
短射先要查三件事：料温够不够、模温够不够、浇口位置对不对。
这三件没查就换高流动料，常常换了还是短射，反而牺牲了耐温和析出。
薄壁件的账，一半在料上，一半在模具上。
一条时间线（行业常见的共面度劣化路径）：胶芯注塑、干态共面度合格 → 装机房运行热循环吸湿 → 共面度缓慢漂移 → 部分针脚虚接、误码率升 → 一年某槽位频繁掉线 → 追溯是吸湿尺寸没按湿态留余量。问题在选材阶段就埋了，只是慢。
六、加工与验证：薄壁和析出要分开盯
信号胶芯是精密注塑件，两个坑分开控。
干燥。 尼龙必烘，半芳香族吸湿低但仍要烘，含水超标薄壁填不满、降解。干燥按实测含水率定窗口。
浇口与流道。 密排针脚胶芯浇口位置决定流动平衡和翘曲。先定浇口，再谈换料。
模温。 半芳香族模温要到 130–150℃ 量级，表面致密、尺寸稳。
低析出工艺。 助剂混匀不团聚，加工温度不超耐温，料筒滞留时间控——停留久析出风险升。
模具与针脚。 密排胶芯的针脚位置靠模具定位件来定，定位件磨损后会整体偏移。
定位件要按模次定期检查，到点了就该复测一次共面度。
模具磨损是慢的，等误码率上来才查，往往已经跑了几十万模。
胶芯的薄壁区还有一个隐藏约束：针脚的插拔力。
保持力不够，针脚会松；保持力够了，孔又容易被撑大。
这一对力要靠材料和孔的公差一起配。
上机前先测一次保持力，再定孔径，比反复换料省事。
插拔力的账，一半在料上，一半在模具公差上。
验证顺序。 建议这样排：
1. 材料级：弯曲模量、热氧保持率、介电（高速件）
2. 尺寸级：干湿态共面度、薄壁充填
3. 析出级：老化 + 表面分析
4. 阻燃级：UL94 / GWIT
5. 系统级：上背板前最后验
顺序不能换。 前一项没过就往下走，后面数据没有解释意义。
七、边界：什么时候胶芯不该用改性尼龙
这一段可能比前面更值钱。
其一，超高速、超低损耗场景。 这类对介电和翘曲的要求，往往指向 LCP 或特种材料，不是尼龙常态区间。
其二，长期温度超过 150℃ 的靠近热源位置。 普通体系不够，要半芳香族高耐温档或 PPS 类。
其三，强电弧或高压位置。 那是高压连接器的账，CTI 和电弧要重新算，别套信号件结论。
其四，年用量小到摊不平精密注塑模和验证。 密排胶芯要开模、做浇口优化、跑共面度和析出，年用量几百个不成立。
其五，要求零析出可见、触点极敏感。 这类要把低析出预期提前对齐，必要时换更洁净基材或特殊牌号。
其六，针脚间距极小且要求长期零漂移。 这类件对共面度的余量要求，已经贴近尼龙湿态漂移的量级。
要继续做，先做湿态共面度可行性，不能只比样条数据。
把这五条写在前头，不是劝退，是省时间。样品阶段共面度全过、量产因析出或湿态漂移回退的项目，见过不止一个——回退代价比当初不做高得多。
还有一条常被漏掉的：胶芯的批次一致性。
同一牌号不同批次，吸湿和流动窗口会有小差，密排件对这种差很敏感。
建议把批次留样写进采购要求，出问题时才有得比。
八、换料风险清单（从原方案换到改性尼龙胶芯，要动什么）
| 环节 | 要动什么 | 容易漏的点 |
|---|
| 模具 | 浇口按流动平衡定，不套旧位 | 玻纤取向导致翘曲 |
| 干燥 | 半芳香族按高要求烘 | 回用料带入水分 |
| 料温 / 模温 | 按 PA6T/PA9T 窗口调 | 只按 PA66 参数给 |
| 保压与脱模 | 薄壁针脚位置重点控 | 短射、熔接线弱 |
| 调湿 | 共面度按湿态留余量 | 干态齐、湿态偏 |
| 色差 | 外观件色板提前确认 | 老化黄变预期 |
| 验证顺序 | 材料 → 尺寸 → 析出 → 阻燃 → 系统 | 前一项未过就往下走 |
| 模具定位件 | 按模次定期复测共面度 | 磨损带来的针脚整体偏移 |
九、一页纸汇报表（给要向上汇报的人）
`
项目：背板 / 高速连接器胶芯 · 材料路线评估
结论方向：低吸湿玻纤尼龙可作为候选，能否落地取决于两条线
一、必须守住的三条
1. 共面度按湿态定，不按干态
2. 析出单列验证，老化后做表面分析
3. 高速件补介电指标，不漏 Dk / Df
二、前置条件（任一不满足则建议暂缓）
· 长期工作温度 ≤ 105℃ 量级（局部另核）
· 与针脚有共面度配合方案
· 年用量足以摊薄精密注塑与验证投入
三、下一步动作
1. 做干湿态共面度与薄壁充填
2. 做老化 + 析出表面分析
3. 高速件测 Dk / Df
风险提示：本路线主要不确定性在共面度与析出，不在初始强度。
`
十、读者常问的三句
问：和进口胶芯尼龙差在哪？
只讲两件能对照的事：同一指标，看它标没标测试状态（干态还是湿态、老化前还是老化后）；同一件上，看它给没给共面度和析出数据。胶芯对状态极敏感，状态不明的数字不宜直接比。有些通用胶芯走国产 PA66-GF 路线已经成熟，有些高速密排件仍建议半芳香族——具体到你的 pitch 和速率，要看温度和析出两样。
问：PA66-GF 够不够，要不要上半芳香族？
看密度和速率。针脚疏、速率低、温度不高的胶芯，PA66-GF 够用也划算；针脚密排、跑高速、长期贴近芯片的，PA6T/PA9T-GF 的低吸湿和低析出更值当。 升级前先确认湿态共面度真超差、介电真到门槛，再换不迟。多数通用胶芯，把钱花在浇口和共面度上比花在基材上升级更值当。
前两天接了个电话，第一句是"你们的尼龙耐多少度"。
这句话没法直接答。耐温要看长期连续使用温度，不是短期峰值；还要看负载、介质、有没有增强。同一句话，答案能从 100℃ 讲到 200℃ 以上。
宁波市科隆新材料有限公司，做改性尼龙（PA6 / PA66 / PA46 / PA11 / PA12 / PA6T / PA9T 及尼龙合金）、改性 PPO / PPS / 热塑性弹性体，以及各大化工巨头尼龙树脂、副牌料、大包料现货。另：长期收尼龙原料、水口回料与各类尼龙废料，有正规处置渠道。
配方里的助剂体系按件的工况配——常规助剂常备现货，特殊型号按需配套；你报工况和牌号，料和助剂一次配齐。
A couple of days ago, I received a phone call from a customer who makes backplane connectors, holding an insulator made of modified nylon injection molding to ask about it.
It is the glue core of the high-speed backplane, with pins densely arranged in two rows. He said: 'The backplane's error rate increased while running; could it be that the glue core has deformed?'
We didn't answer about the deformation first; we asked two things first: how the pin coplanarity was tested, and how long the data center had been running. The answers came back — coplanarity was measured in a dry state, and it had been running for a year.
This matter clearly explains the core of choosing the material for the signal connector's insulation core: it can't withstand high voltage, but it is extremely sensitive to whether the pins are aligned and whether the surface is clean. What is needed here has never been a high CTI; it requires both coplanarity and low exudation to be reliable.
1. The gel core part is about 'positioning', not 'conducting electricity'.
First, clarify the role of the ferrule in the connector.
Conductivity relies on the pins and terminals, while the plastic core (insulator) does three things: it holds the pins in precise positions, separates adjacent signals, and blocks external stress and media. It does not conduct electricity, but the success of the signal is often in its hands.
The backplane connector inserts the board card into the backplane, with tens to hundreds of pins; high-speed connectors run at 25G, 56G, or even higher speeds. The requirements for the adhesive core of this type of component are completely different from those of high-voltage connectors—
For high-voltage components, check CTI and arcs; for signal components, check coplanarity, precipitation, and dielectric properties. Applying the criteria for high-voltage components to signal gel cores will miss the two truly critical items.
The precision of the glue core is not about 'whether it can insulate', but about 'whether the pins are aligned properly'.
On a densely packed rubber core, the pin spacing may be only a few tenths of a millimeter.
Even if a few pins in the entire row are out of tolerance, the signal will be intermittent.
So for the acceptance of the adhesive core, first look at the coplanarity, then look at the others.
In a nutshell: For signal connector gel cores, CTI is not the main focus; coplanarity and low outgassing are.
2. Six-dimensional working condition: What constrains the rubber core
Spread out across six dimensions.
Temperature. The backplane is close to the chip, commonly 85–105°C for long periods, and locally even higher. It is higher than the data center ambient temperature, so thermal aging and dimensional drift must be considered.
Signals and payload. It cannot withstand high voltage, but the pins must have insertion and retention force, and the rubber core must resist creep to prevent the pins from loosening. High-speed signals are sensitive to dielectric constant and loss.
Medium. It does not directly soak in liquid, but is exposed to data center air, condensation, and dust. Low deposition is key—deposits on contacts cause trouble.
Service life. Starting from ten years, insertion and removal counted in thousands of times. The criteria are the retention rate of coplanarity and the degree of precipitation, not the initial strength.
Appearance and cleanliness. Low smoke and low toxicity, low yellowing, low precipitation, suitable for computer room environment requirements.
Compliant. Flame retardant UL94 V0, mostly halogen-free; the CTI requirements for signal components are relatively low, but it still needs to be considered when near the power supply.
In the six dimensions, temperature, signal rate, and lifespan all have specific numbers provided; the accuracy at the component level is right here.
3. Three material paths, coplanarity and precipitation should be calculated separately
Place the candidate glue core routes side by side and look at how the columns 'coplanarity' and 'precipitation' go.
| Route | compose | Coplanarity / Dimension | Precipitation and Dielectric | Cost |
|---|
| PA66-GF | Aliphatic glass fiber reinforced | Moisture absorption warping, wet state drifting | Conventional, dielectric is average | Cheap and well-crafted |
| PA6T/PA9T-GF | Semi-aromatic glass fiber reinforced | Low moisture absorption, most dimensionally stable | Low precipitation, low dielectric excellence | Expensive, narrow processing window |
| PA46-GF | High-temperature aliphatic | Good temperature resistance, still high moisture absorption | Regular | Somewhere between the two |
There is no 'which is better' among the three routes, only 'which route is tighter'.
PA66-GF is a common main material, inexpensive and easy to process, but it absorbs 8% moisture and exhibits anisotropic warping, causing flatness to drift when wet. For densely packed rubber cores with pin spacing of 0.4–2.0 mm, this amount of drift is enough to cause partial pin contact failure.
PA6T/PA9T-GF has low moisture absorption, stable dimensions, and more favorable dielectric constant and loss factor, making it a common choice for high-speed adhesive cores. The cost is higher unit price and processing difficulty — high melting point, requiring adjustments to both material and mold temperatures.
PA46-GF has good temperature resistance, but its moisture absorption is still relatively high, and its coplanarity accounting is similar to PA66.
In one sentence: The expensive material of the signal adhesive core is not bought for strength, but for meeting both 'pins remain intact for ten years' and 'surface stays clean for ten years' at the same time.
4. Selection Criteria Table: Coplanarity and Precipitation Are the Main Focus
Turn constraints into verifiable indicators. The thresholds in the table below are directional suggestions, not acceptance standards—they are actually determined by the project, working conditions, and actual measurements.
| Indicator | Directional threshold | Verification Method / Standard | Common Failures | Common solution | Corresponding auxiliary agent system |
|---|
| Coplanarity (pin plane) | Determined by pitch, usually ≤ 0.05–0.1 mm | Coordinate Measuring Machine / Coplanarity Tester | False connection, error code | Low warpage Gate optimization | Nucleating agent (crystallization control) |
| Thin-wall filling | Fully filled with no short shots and no weld line weakness | Short-range X-ray | Localized material deficiency, weak strength | High mobility system | Lubricant (thin-wall flow) |
| Moisture absorption dimensional stability | The difference between wet and dry states is controllable | ISO 62 Measurement | Coplanarity Drift | Low moisture-absorbing substrate | Intrinsic properties of the material, without relying on additives |
| Precipitation (contact pollution) | No migratory substances on the surface after soaking/aging | Aging Surface Analysis | Increased contact resistance and corrosion | Low migration formula | Lubricant (Low Precipitation Selection) |
| Flame retardant | UL94 V0, halogen-free preferred | UL94 / GWIT | Safety standard rework | Halogen-free flame retardant system | — |
| Dielectric (high-speed component) | Dk / Df determined by rate | IPC / Resonant Cavity Method | Signal attenuation, crosstalk | Low dielectric substrate | The material is intrinsic and does not rely on additives |
| Long-term heat resistance | 105℃ × long-term storage | ISO 527 | Pale and brittle | Thermally stable system | Antioxidant (thermal-oxidative) |
| Weld line strength | The pin position does not fall on the weld line | Short shot Mechanical test | Localized cracking | Gate and runner optimization | Coupling agent (interface) |
| Hot wire / Safety regulations | Set GWIT by scenario | GWIT / GWFI | Safety standard rework | Flame retardant synergistic system | Flame Retardant (Synergist) |
How to use this table: Separate the row for coplanarity and the row for precipitation. Focus only on flame retardancy, not coplanarity. If the glue core passes V0, the backplane can still have errors. For high-speed components, you also need to fill in the dielectric row.
A reminder: For coplanarity, the design must be based on data in a humidified state. PA66 absorbs 8% moisture, and after the core absorbs moisture, the pins shift overall, causing a drop in coplanarity — if you use dry-state measurements to determine coplanarity for assembly, after a year in the data center with rising humidity, the pins will become loose.
5. Five Common Misjudgments and the Real Causes
Judgment 1: Only the flame-retardant V0 of the glue core is considered.
The most common opening in inquiries about signal connectors is 'We need V0 material.' That's a correct question, but it's only half of it. Even if the plastic core is V0 compliant, if the coplanarity is off or there are contaminations, the backplane will still have errors or poor contact. When we accept orders for plastic cores, we always make sure to mention coplanarity, contamination, and flame retardancy together; missing any one of them can prevent the product from being finalized.
Counter Judgment Two: Treating the precipitate as the base material pot.
When a customer finds that the contact resistance has increased, their first reaction is that the "material is bad." Often, the root cause is the low-molecular additives in the formulation—the lubricant is added in excess or a type with insufficient temperature resistance is chosen. Under prolonged heat, it migrates to the surface, contaminating the metal contacts and forming an insulating or corrosive layer. Changing the substrate without changing the formulation will still result in deposits. When deposits are observed, first check the list of additives and the surface composition after aging, and don't rush to change the material.
Reverse Judgment Three: Design is determined using the dry-state coplanarity measurement.
Nylon absorbs moisture and can swell and warp. PA66 absorbs 8% moisture, and the dense-molded rubber core shifts as a whole after absorbing moisture, causing the coplanarity to drop. If coplanarity is measured in the dry state for assembly, it will be loosely connected in the wet state. Both coplanarity and dimensions need to allow for tolerances according to the conditioned moisture state.
Judgment four: Use the PA66 process to make semi-aromatic.
PA6T/PA9T has a melting point above 300℃. The material temperature, mold temperature, and drying window are not the same as PA66. If the process is not changed, thin walls cannot be filled completely, internal stress is high, and both coplanarity and precipitation are affected. Even for the same grade with different processes, the actual performance can differ by one level.
Contradiction 5: Mistaking thin-walled short shots for poor material flow.
Before short-shotting, you first need to check three things: whether the material temperature is sufficient, whether the mold temperature is sufficient, and whether the gate position is correct.
These three items were replaced with high-flow materials without inspection, and often, even after the replacement, short shots still occurred, while temperature resistance and precipitation were sacrificed.
For thin-walled parts, half is on the material and half on the mold.
A timeline (common coplanarity degradation path in the industry): molded core injection, dry-state coplanarity qualified → installed in the server room, undergoes thermal cycling and moisture absorption → coplanarity slowly drifts → some pins have poor contact, bit error rate rises → after a year, a certain slot frequently disconnects → trace back, the issue is that moisture absorption dimensions did not leave a margin for the wet state. The problem was buried at the material selection stage; it’s just slow.
6. Processing and Verification: Keep thin walls and precipitation under separate monitoring
The signal gel core is a precision injection-molded part, with the two cavities controlled separately.
Dry. Nylon must be baked; semi-aromatic types have low moisture absorption but still need to be baked. If the water content exceeds the standard, thin walls cannot be filled completely and degradation occurs. Drying should be set according to the measured moisture content.
Gate and runner. The gate position of the densely arranged pin-core determines flow balance and warpage. Determine the gate first, then discuss material change.
Mold temperature. Semi-aromatic mold temperature needs to reach the 130–150°C range, with a dense surface and stable dimensions.
Low precipitation process. Additives are mixed evenly without agglomeration, processing temperature does not exceed heat resistance, and the residence time in the barrel is controlled—the longer the residence, the higher the risk of precipitation.
Molds and pins. The pin positions of closely packed rubber cores are determined by the mold's positioning parts, and after the positioning parts wear out, the overall position will shift.
The locating parts should be checked regularly according to the mold cycle, and when it's time, the coplanarity should be re-measured once.
Mold wear is slow, and it is only checked when the error rate rises, by which time it has often been used for hundreds of thousands of cycles.
There is another hidden constraint in the thin-walled area of the gel core: the insertion and removal force of the pins.
If the holding force is insufficient, the stitches will loosen; if the holding force is sufficient, the holes are easily stretched larger.
This pair of forces needs to be matched together with the material and hole tolerances.
Test the holding force once before going on the machine, then determine the aperture size; this is more convenient than repeatedly changing the material.
The insertion and extraction force depends half on the material and half on the mold tolerance.
Verification order. It is recommended to arrange it like this:
1. Material grade: bending modulus, thermo-oxidative retention, dielectric (for high-speed parts)
2. Dimensional level: coplanarity in dry and wet states, thin-wall filling
3. Precipitation Stage: Aging Surface Analysis
4. Flame Retardant Grade: UL94 / GWIT
5. System level: Final inspection in front of the upper backplane
The order cannot be changed. If the previous item hasn't passed, just move on, the later data has no explanatory significance.
7. Boundaries: When the adhesive core should not use modified nylon
This section might be more valuable than the previous one.
First, ultra-high-speed, ultra-low-loss scenarios. The requirements for dielectric properties and warpage in such cases often point to LCP or special materials, not the normal range for nylon.
Second, locations near the heat source where the long-term temperature exceeds 150℃. Ordinary systems are not sufficient; semi-aromatic high-temperature resistant grades or PPS types are needed.
Third, strong electric arcs or high-voltage locations. That concerns high-voltage connectors; CTI and arcs need to be recalculated, and do not apply the conclusions for signal components.
Fourth, the annual usage is too small to justify the cost of precision injection molds and validation. Dense-arranged rubber cores require mold opening, gate optimization, running flatness checks, and extraction, which is not feasible with an annual usage of only a few hundred units.
Fifth, it requires zero visible precipitation and extremely sensitive contacts. In this case, the expected low precipitation should be aligned in advance, and if necessary, a cleaner substrate or a special grade should be used.
Sixth, the pin spacing is extremely small and long-term zero drift is required. The tolerance for coplanarity for such components is already approaching the magnitude of nylon wet-state drift.
To continue, first assess the feasibility of wet state coplanarity; you can't just compare the spline data.
Writing these five points at the beginning is not to discourage, but to save time. I have seen more than one project in the sample stage pass all coplanarity checks, only to be reverted during mass production due to precipitation or wet-state drift — the cost of rollback is much higher than not doing it well in the first place.
There is also one that is often overlooked: the batch consistency of the gel core.
For the same grade but different batches, there can be slight differences in moisture absorption and flow window, and densely packed parts are very sensitive to such differences.
It is recommended to include batch samples in the purchase requirements, so there is something to compare when problems arise.
8. Material Change Risk List (From the original plan to modified nylon core, what needs to be changed)
| link; segment; part | What needs to be moved? | Points that are easy to overlook |
|---|
| Mold | The gate is determined according to flow balance and does not follow the old position. | Fiber orientation causes warping |
| Dry | Semi-aromatic baked according to high standards | Recycled material introduces moisture |
| Material Temperature / Mold Temperature | Adjust according to PA6T/PA9T window | Only provide according to PA66 parameters |
| Pressure Holding and Demolding | Key control of thin-wall pin positions | Short shot, weak weld line |
| Humidity control | Coplanarity according to wet-state allowance | Dry state is even, wet state is uneven |
| Color difference | Advance confirmation of exterior color samples | Expected aging and yellowing |
| Verification order | Material → Size → Precipitation → Flame Retardant → System | If the previous item fails, just move on. |
| Mold positioning component | Periodic re-measurement of coplanarity according to the module number | Overall stitch displacement caused by wear |
9. One-page report form (for those who need to report upward)
`
Project: Backplane / High-Speed Connector Adhesive Core · Material Route Evaluation
Conclusion direction: Low-moisture-absorption glass fiber nylon can be considered as a candidate, but whether it can be implemented depends on two lines.
1. Three Rules That Must Be Followed
1. Coplanarity is determined in the wet state, not in the dry state
2. Perform single-column verification, and carry out surface analysis after aging
3. Supplement dielectric specifications for high-speed components, without omitting Dk / Df
2. Precondition (It is recommended to postpone if any are not met)
· Long-term operating temperature ≤ 105℃ range (locally verified separately)
· Plan for coplanarity fitting with pins
· The annual consumption is sufficient to dilute the investment in precision injection molding and validation
3. Next Steps
1. Perform co-planarity test for dry and wet states and thin-wall filling
2. Perform aging Precipitation surface analysis
3. Measure Dk / Df of high-speed components
Risk reminder: The main uncertainties of this route lie in coplanarity and precipitation, not in initial strength.
`
10. Three Questions Frequently Asked by Readers
Question: How does it differ from imported rubber-core nylon?
Let's only talk about two comparable things: for the same indicator, see whether it indicates the test condition (dry state or wet state, before or after aging); for the same part, see whether it provides coplanarity and deposition data. The core material is extremely sensitive to the condition, so numbers with unknown conditions should not be directly compared. Some general-purpose cores using the domestic PA66-GF route are already mature, while some high-speed dense parts still recommend semi-aromatic—specifically for your pitch and speed, you need to consider both temperature and deposition.
Question: Is PA66-GF sufficient, or should we go for semi-aromatic?
Look at density and speed. For glue cores with sparse pins, low speed, and not high temperature, PA66-GF is sufficient and cost-effective; for densely packed pins, high-speed operation, and long-term proximity to chips, the low moisture absorption and low exudation of PA6T/PA9T-GF are more worthwhile. Before upgrading, first confirm that the wet-state coplanarity is truly out of spec and the dielectric actually reaches the threshold; only then is it not too late to switch. For most general-purpose glue cores, spending money on the gate and coplanarity is more worthwhile than upgrading the base material.
A couple of days ago, I received a phone call, and the first sentence was, 'How heat-resistant is your nylon?'
This sentence cannot be answered directly. The temperature resistance depends on the long-term continuous usage temperature, not the short-term peak; it also depends on the load, the medium, and whether there is reinforcement. For the same sentence, the answer could range from 100℃ to over 200℃.
Ningbo Kelon New Materials Co., Ltd. specializes in modified nylon (PA6 / PA66 / PA46 / PA11 / PA12 / PA6T / PA9T and nylon alloys), modified PPO / PPS / thermoplastic elastomers, as well as nylon resins from major chemical companies, secondary materials, and bulk material in stock. Additionally, we have long-term procurement of nylon raw materials, sprue material, and various nylon waste, with formal disposal channels.
The auxiliary system in the formula is matched according to the working conditions per item — conventional auxiliaries are kept in stock, and special models are matched as needed; you report the working conditions and grade, and the materials and auxiliaries are prepared together at once.
