变压器和电感骨架是最"传统"的一类塑料电气件，但选料从来没变简单过。
它要经受长期温升、要提供爬电距离、要固定端子、还要被漆包线反复勒紧。
四种失效模式，对应四条完全不同的解法，把它们混在一起谈，就容易只盯耐温而漏掉其他三条。
变压器骨架有一道独特的工序：浸漆。
浸漆烘烤的温度把骨架一起加热，普通牌号在这个环节变形。
一家音频变压器厂吃过亏，骨架烘烤后窗口位偏移，绕线全乱。
换耐温等级更高的料，烘烤过了，成本上去了，客户要压价。
两头夹击之下，浸漆工艺本身的优化反而成了出路。
骨架的选型故事，常常是这样的三方博弈。
一、先列四种失效
第一，软化变形。 长期温升超过材料承受能力，骨架蠕变、绕线柱歪斜，电感量漂掉。
第二，爬电起痕。 潮湿加污染加电压，表面形成碳化通道，最后击穿。
第三，端子开裂。 嵌件与塑料的热膨胀差加上绕线拉力，端子周围应力集中。
第四，绕线割伤。 骨架表面毛刺、圆角不足，把漆包线绝缘层勒破。
四条线的材料要求不一样：变形看耐温与抗蠕变，起痕看 CTI，开裂看嵌件结构，割伤看表面与结构设计。
二、耐温这一条，先对绝缘等级
骨架的耐温要求来自绕组绝缘等级，不是来自环境温度。
常见的对应关系是：B 级约 130℃、F 级约 155℃、H 级约 180℃。要按长期连续温度去对，而不是按短时峰值。
材料档位大致是：130℃ 用阻燃增强 PA66；155℃ 用高温体系 PA66 或 PA46；180℃ 以上通常要 PA6T、PA9T 或 PPS。
这里最容易犯的错是把 HDT 当长期耐温用。 HDT 是短时指标，长期表现要看热老化后的性能保留率，两者不是一回事。
三、薄壁长流程是骨架的注塑难点
骨架的壁厚常常只有零点几毫米到一毫米出头，流程却很长，中间还有端子要包住。
这决定了它对材料的流动性要求很高，而流动性好的料往往玻纤含量低、刚性又不够，形成一对矛盾。
处理办法通常是三条并行：提高料温与模温（在不降解的前提下）、调整浇口位置缩短流程、在关键位置加厚（用结构补材料）。
填充不足最典型的表现是端子附近缺料和末端发白，这类缺陷在整批件里往往只出现在个别模腔，很容易被抽检漏掉，必须靠首件与批量结合的检查方式。
还有一个结构性办法：把端子周围的壁厚适当加厚，用局部厚壁换取填充余量。薄壁件的设计原则不是"哪里都薄"，而是"该薄的地方薄、该厚的地方厚"。
四、端子嵌件：结构优先于材料
骨架端子上要焊线、要插接，还要被绕线张力持续拉着。
端子周围的应力来自三处：热膨胀差、绕线张力、装配应力。
处理顺序建议是：先改端子的咬合几何（滚花、环槽、倒钩）→ 再给塑料孔壁留足圆角与壁厚 → 再试嵌件预热 → 最后才考虑换料。
先换料往往解决不了问题，因为应力来自几何与温差，不是来自材料强度不够。
顺带说一个常见的返工场景：端子开裂后换了更韧的材料，结果变成端子松动。因为提高韧性往往要降低刚性，端子的保持力跟着下降。这是"按下葫芦浮起瓢"的典型，说明问题本不在材料。
正确做法是回到几何：咬合结构决定抗拉脱，孔壁厚度决定抗开裂，预热决定残余应力。三件事定完，再谈选材料。
还有一条容易被忽略的：绕线张力。骨架在绕线时承受持续的拉应力，张力越大，端子与绕线柱的负担越重。有些开裂案例把张力降一点就消失了——工艺参数的调整比换料便宜得多，也应该先试。
另外，骨架的表面质量直接影响绕线：毛刺、飞边、脱模拉伤都可能割破漆包线。这类问题在绝缘测试时不一定立刻暴露，但会在长期振动或温升后变成匝间短路。表面要求要写进图纸。
五、爬电距离与 CTI 要一起算
骨架的形状本身就是爬电距离的载体。
同样的表面距离，高 CTI 材料能承受更高的电压。所以"距离够不够"这个问题，答案取决于用什么材料。
这里有个实操顺序：先按目标电压与安规等级，确定需要的爬电距离；再根据可用的结构空间，倒推需要的 CTI 等级。倒推出来的 CTI 若超出常规材料，就要在结构与材料之间重新找平衡。
六、材料档次表
| 长期温升 | 材料方向 | 注意项 |
|---|
| 130℃ 以内 | 阻燃增强 PA66 | 关注 CTI 与熔接线 |
| 约 155℃ | 高温 PA66 / PA46 | 要长期老化数据 |
| 180℃ 以上 | PA6T / PA9T / PPS | 工艺窗口要重调 |
换档位的同时，干燥、模温、螺杆都要跟着换。 这一点在高温尼龙上尤其不能省。
七、验证清单
① 球压试验，确认长期温升下不软化。
② 耐压与绝缘电阻，调湿态测。
③ CTI，调湿态数据。
④ 端子拉拔力与扭转力，测嵌件结合。
⑤ 绕线试验，按实际张力绕满，检查割伤与变形。
⑥ 冷热循环后复测前四项。
六项里有三项是"调湿态"，这不是重复，而是因为骨架的真实工况就是吸湿环境。
最后提一句验证顺序：先做材料级（球压、CTI），再做零件级（拉拔、耐压），最后做整机级（绕线、老化）。顺序颠倒会浪费很多轮次，因为材料不过关时，零件级测试的意义有限。
浸漆工艺对骨架的影响，分三段看。
浸漆时漆液渗入塑料表面，某些溶剂体系会引起溶胀。
烘烤升温段是第一考验，热变形在这个阶段出现。
保温段是第二考验，长期温度考验耐热等级。
降温段看似安全，急冷也会留下内应力。
所以骨架料的验证要模拟完整浸漆曲线，不能只测一个烘烤温度。
有工厂和漆厂联合开发低温快固漆，把烘烤温度降了三十度。
材料与工艺互相让一步，成本和质量同时受益。
追问一：浸漆漆种对骨架影响多大？
差异很大。溶剂型漆的溶胀风险高，水性漆温和但烘烤要求不同。骨架料定点前，用实际漆种做浸泡和曲线验证，别用通用溶剂数据代替。漆和料是搭档，搭档要一起面试。
追问二：骨架的耐温等级怎么选牌号？
按烘烤温度加安全余量选，还要看 RTI 长期数据。短时扛得住烘烤，长期还要扛得住工作温度，两个温度两条线。只按烘烤温度选料，服役几年后的老化会补上那一课。
一单绕线乱槽的追查
烘烤后窗口位偏移，绕线全乱。追查发现那批漆的固化温度标高了十度，产线按标签执行，骨架超温变形。整改是漆厂修正标签加工艺端增加温度监控。供应链上游的标签错误，最后由下游的骨架变形买单。来料文件的核验，也是骨架质量管理的一部分。
骨架验证三步
实漆浸泡、完整曲线模拟、老化后尺寸与绝缘复测。三步走完，浸漆关就过了。
收一句：骨架是线圈的骨骼，浸漆是给骨骼洗澡，水温不对骨骼会变形。把漆当工艺参数来管，把料当结构伙伴来选，变压器的寿命就有了底座。
变压器骨架还有一个特殊场景：高频电感。开关电源里的电感骨架，频率上百千赫兹，材料在高频下的损耗成为新指标。普通牌号的高频损耗大，发热叠加绕组温升，骨架温度比预期高。高频场景要选低损耗配方，行业内成熟方案有专门的电感骨架牌号。频率每上一个台阶，材料清单就要重排一次。做电源的客户最认这条规律，因为他们烧过的骨架比谁都多。
骨架与包封工艺的配合也是一门课。包封料固化时放热，固化温度让骨架再受一次热考验。包封体的收缩应力压在骨架上，薄壁位置可能应力开裂。骨架料的验证要把包封曲线一起模拟，只按裸件验证的方案会漏风险。有工厂和包封料厂开联席会，把两个体系的验证合并成一套，效率翻倍。跨厂联席听起来麻烦，算总账它是最快的路。
清单收官
变压器与电感骨架定点资料包：实漆实包封曲线验证、烘烤与工作双温度数据、高频损耗数据（如适用）、应力开裂验证、老化后尺寸绝缘复测。骨架是与工艺绑定最深的一类塑料件，资料包里工艺数据占一半，这很正常。
变压器骨架的批次管理还有一个要点：介电批次抽测。骨架的介电性能有批间波动，批次抽测加留样对比，波动曲线就建立了。有工厂把三年的介电批次数据画成控制图，超限自动预警。介电数据的控制图，让骨架的批次管理从抽样放行升级为趋势管理。趋势管理的好处是能看见缓慢劣化，单批放行看不见的东西它看得见。
浸漆环节的清洁度也影响骨架。漆液里的杂质渗进骨架表面，介电薄弱点就在杂质处。漆液过滤和浸槽清洁要有制度，别让省下的过滤芯钱变成售后工单。有工厂的介电不良率季节性波动，追查是夏季漆液变质快，过滤周期没跟着调。工艺参数跟着季节走，这个原则在浸漆上同样灵验。
给变压器厂一个选料协同建议：把漆、包封料、骨架料三家供应商约到一起做联合验证。三个体系交界面上的问题，单独验证永远测不出来。联合验证一次的成本，低于任何一次批量返修。界面问题用联席会解决，这是电子制造业的成熟经验，值得每个骨架项目复制。
变压器骨架的批量管理再补一点：留样与追溯。骨架的介电数据要能追溯到材料批次和成型参数，留样是追溯的实物锚点。有工厂的留样室按批次排列，客户审核时这张王牌最能加分。客户对供应商的信任，一半来自走进留样室的那一刻。留样成本每件几毛，信任价值无法计价。
骨架与安规认证的关系要说清。变压器类产品有过安规认证，骨架作为绝缘系统的一部分在认证范围内。换料要重新评估安规，这项成本要提前算进项目。有工厂换料时漏了安规重评，整批货卡在海关。认证与材料的关系，是骨架类项目排期表上的第一行。排产之前先把认证路径画完，项目才不会中途断粮。
变压器厂的选料收官建议：把漆、包封、骨架三个体系的联合验证当成标配动作，把安规路径当成排期表第一行。两件事做到位，骨架类项目的失败率断崖式下降。行业的成熟经验就摆在那里，抄作业不丢人，返工才丢人。
骨架项目最后补一个跨文化经验。日本客户对骨架的介电批次数据要求按批附报告，欧美客户要长期数据曲线，国内客户最看重响应速度。三种客户三种期望，服务方案要跟着分档。有工厂做了三套资料包模板，接不同客户的单子都不慌。理解客户的检查逻辑，比多跑几次现场都管用。
骨架的失效还有一类冷门场景要提：鼠咬与虫害。户外和仓储环境的骨架被鼠咬破损，介电失效来得毫无征兆。防鼠设计在户外变压器上不是笑话，是实打实的失效项。外壳的孔洞设计和材料硬度都影响鼠咬风险。行业经验库里的冷门条目，往往就是某次奇怪售后的唯一答案，值得慢慢攒。
结语
变压器骨架选料，判断链是：
先对绝缘等级定耐温档 → 再按电压与空间定 CTI → 再按端子与绕线定结构 → 最后才是选具体牌号。
四种失效各管一条线，缺哪条都会在验证阶段暴露出来。
Transformer and inductor cores are the most 'traditional' type of plastic electrical components, but selecting materials has never been any easier.
It has to withstand long-term temperature rises, provide creepage distance, secure terminals, and be repeatedly tightened by enameled wire.
Four failure modes correspond to four completely different solutions. If you mix them together in discussion, it's easy to focus only on temperature resistance and overlook the other three.
The transformer frame has a unique process: varnishing.
The temperature of baking after dipping in paint heats the frame together, and ordinary grades deform at this stage.
An audio transformer factory suffered a loss; after the core frame was baked, the window position shifted, and all the winding was messed up.
Switch to materials with a higher temperature rating; after baking, the costs went up, and the customer wants a price reduction.
Under a pincer attack from both sides, the optimization of the varnishing process itself has instead become the way out.
The story of choosing a framework is often a three-way game.
1. First, list the four types of failures
First, softening deformation. When the long-term temperature rise exceeds the material's tolerance, the framework creeps, the winding posts tilt, and the inductance drifts.
Second, tracking. Moisture combined with pollution and voltage forms a carbonized path on the surface, eventually leading to breakdown.
Third, terminal cracking. The stress concentration around the terminal is caused by the difference in thermal expansion between the insert and the plastic, combined with the winding tension.
Fourth, winding cuts. Burrs on the frame surface and insufficient rounding can damage the insulation layer of the enameled wire.
The material requirements for the four lines are different: for deformation, consider temperature resistance and creep resistance; for marks, consider CTI; for cracking, consider the insert structure; for cuts, consider surface and structural design.
2. Regarding temperature resistance, first consider the insulation class
The temperature resistance requirement of the skeleton comes from the insulation grade of the winding, not from the ambient temperature.
The common correspondence is: Class B about 130°C, Class F about 155°C, Class H about 180°C. It should be matched according to long-term continuous temperature, not according to short-term peak values.
The material grades are roughly: 130℃ uses flame-retardant reinforced PA66; 155℃ uses high-temperature system PA66 or PA46; above 180℃ usually requires PA6T, PA9T, or PPS.
The most common mistake here is treating HDT as suitable for long-term heat resistance. HDT is a short-term indicator; long-term performance depends on the retention rate of properties after thermal aging, and the two are not the same.
3. Thin-walled long processes are the injection molding challenges of the framework
The wall thickness of the skeleton is often only a few tenths of a millimeter to just over one millimeter, but the process is very long, and in the middle, the terminals also need to be covered.
This determines that it has very high requirements for the fluidity of the material, but materials with good fluidity often have low glass fiber content and insufficient rigidity, forming a contradiction.
The usual handling methods are three parallel approaches: increase the material temperature and mold temperature (without causing degradation), adjust the gate position to shorten the flow path, and thicken key areas (using structure to add material).
The most typical表现 of insufficient filling is material shortage near the terminals and whitening at the ends. Such defects often appear only in individual cavities within a whole batch, making them easy to miss in spot checks, and inspection combining first article and batch checks is required.
There is another structural approach: appropriately thicken the wall around the terminals, using locally thick walls to compensate for the filling allowance. The design principle for thin-walled parts is not "thin everywhere," but "thin where it should be thin, and thick where it should be thick."
4. Terminal Inserts: Structure Over Material
Wires need to be soldered and connected on the skeleton terminals, and they also need to be continuously pulled by the winding tension.
The stress around the terminal comes from three sources: thermal expansion differences, winding tension, and assembly stress.
The recommended processing order is: first modify the terminal's clamping geometry (knurled, ring groove, barb) → then leave enough fillet and wall thickness for the plastic hole → then try preheating the insert → finally consider changing the material.
Changing the material first often does not solve the problem, because the stress comes from geometry and temperature difference, not from insufficient material strength.
By the way, here is a common rework scenario: after the terminal cracked, it was replaced with a tougher material, but it resulted in the terminal becoming loose. Increasing toughness often requires reducing rigidity, causing the terminal's retention force to decrease. This is a typical case of 'pressing the gourd and the ladle rises,' indicating that the problem was not originally with the material.
The correct approach is to return to geometry: the occlusion structure determines tensile resistance, the thickness of the hole wall determines cracking resistance, and preheating determines residual stress. Once these three things are decided, then talk about material selection.
There is one more thing that is easily overlooked: winding tension. The core experiences continuous tensile stress during winding; the greater the tension, the heavier the burden on the terminals and the winding post. In some cracking cases, reducing the tension slightly made it disappear—adjusting process parameters is much cheaper than changing materials, and should be tried first.
In addition, the surface quality of the skeleton directly affects winding: burrs, flash, and demolding scratches can all cut the enameled wire. These issues may not be immediately revealed during insulation testing, but can lead to inter-turn short circuits after long-term vibration or temperature rise. Surface requirements should be specified in the drawings.
5. Creepage distance should be calculated together with CTI
The shape of the skeleton itself is the carrier of the creepage distance.
For the same surface distance, materials with a higher CTI can withstand higher voltages. So the question of whether the 'distance is enough' depends on what material is used.
Here is a practical procedure: First, according to the target voltage and safety standard level, determine the required creepage distance; then, based on the available structural space, work backward to determine the required CTI level. If the back-calculated CTI exceeds conventional materials, it is necessary to find a new balance between structure and materials.
6. Material Grade Table
| Long-term temperature rise | Material direction | Notes |
|---|
| Within 130℃ | Flame-retardant reinforced PA66 | Focus on CTI and fusion lines |
| About 155℃ | High-temperature PA66 / PA46 | Need to age data long-term |
| Above 180℃ | PA6T / PA9T / PPS | The process window needs to be readjusted |
At the same time as changing gears, drying, mold temperature, and the screw must also be changed. This is especially important for high-temperature nylon.
7. Verification Checklist
① Ball pressure test to confirm that it does not soften under long-term temperature rise.
② Withstand voltage and insulation resistance, measured in a humidity-conditioned state.
③ CTI, data under humidified conditions.
④ Terminal pull-out force and torque, test the assembly of the insert.
⑤ Winding test: wind fully according to the actual tension, and check for cuts and deformation.
⑥ Retest the first four items after the hot and cold cycles.
Three of the six items are 'humidity control state'. This is not a repetition, but because the actual working condition of the framework is a moisture-absorbing environment.
One last point about the verification sequence: first conduct the material-level tests (ball pressure, CTI), then the component-level tests (pull, pressure resistance), and finally the whole machine-level tests (winding, aging). Reversing the order will waste many rounds, because if the material does not pass, the component-level tests are of limited significance.
The impact of the varnishing process on the framework can be considered in three parts.
When dipping in paint, the paint penetrates the surface of the plastic, and certain solvent systems can cause swelling.
The heating stage of baking is the first test, and thermal deformation occurs at this stage.
The insulation section is the second test, a long-term temperature test to assess the heat resistance level.
The cooling phase seems safe, but rapid cooling can leave internal stress.
Therefore, verification of frame materials must simulate a complete dipping curve; it is not possible to measure only one baking temperature.
Some factories and paint manufacturers jointly developed low-temperature fast-curing paint, lowering the baking temperature by thirty degrees.
Materials and processes yield one step to each other, benefiting both cost and quality.
Follow-up question one: How much does the type of impregnation paint affect the skeleton?
There is a big difference. Solvent-based paint has a high risk of swelling; water-based paint is milder but requires different baking requirements. Before determining the skeleton material, use the actual paint type for soaking and curve verification, rather than replacing it with generic solvent data. Paint and material are partners, and partners must interview together.
Follow-up question two: How do you choose the temperature resistance grade for the frame?
Choose based on baking temperature plus safety margin, and also check long-term RTI data. It can withstand baking for a short time and can withstand operating temperature in the long term—two temperatures, two lines. If you select materials only based on baking temperature, aging after a few years of service will catch up on that lesson.
Tracking the tangled groove winding in a single order
After baking, the window position shifted and the winding was completely messy. Investigation found that the curing temperature of that batch of paint was ten degrees higher, the production line followed the label, and the frame was overheated and deformed. Rectification involved the paint factory correcting labels and adding temperature monitoring at the process end. If the upstream label was incorrect, the downstream frame deformation would pay the price. Verifying incoming material documents is also part of skeleton quality management.
Three steps for skeleton verification
solid paint soaking, complete curve simulation, and re-testing dimensions and insulation after aging. Once these three steps are completed, the soaking test is over .
to sum up: the skeleton is the coil's skeleton; soaking with varnish is to bathe the bones. If the water temperature is incorrect, the bones will deform. Treat the paint as a process parameter and the material as a structural partner when chosen, thus giving the transformer a base for lifespan.
Transformer skeleton has another special scenario: high-frequency inductors. The inductor skeleton in switching power supplies has frequencies in the hundreds of kilohertz, and material loss at high frequencies becomes a new indicator. Ordinary grades have high high-frequency losses, and with heat generation and winding temperature rise, the skeleton temperature is higher than expected. For high-frequency scenarios, low-loss formulas should be chosen; mature industry solutions have specialized inductor skeleton grades. Every time frequency rises, the material list must be rearranged. Power supply customers recognize this rule the most because they have burned more skeletons than anyone else. The coordination between
skeleton and encapsulation process is also a skill. When encapsulation material cures, heat is released, and curing temperature allows the skeleton to undergo another thermal test. The shrinkage stress of the encapsulation body presses on the skeleton, and the thin wall area may crack due to stress. Verification of skeleton materials requires simulating the encapsulation curve together; using only bare parts for verification can cause risk leakage. Some factories and encapsulation material factories hold joint meetings to combine verification of the two systems into one set, doubling efficiency. Cross-factory joint meetings sound troublesome, but calculating the total is the fastest way.
Checklist Summary
Transformer and Inductor Framework Fixed-Point Data Package: Solid paint and encapsulation curve verification, dual temperature data for baking and operation, high-frequency loss data (if applicable), stress cracking verification, post-aging dimensional insulation retesting. The skeleton is the type of plastic component most closely tied to the process, and process data accounts for half of the package, which is normal.
Another key point in batch management of transformer skeletons: dielectric batch sampling. The dielectric performance of the skeleton fluctuates between batches; by comparing batch sampling and retained samples, the fluctuation curve is established. Some factories have drawn three years of dielectric batch data into control charts, with automatic warning over limits. Dielectric data control charts upgrade skeleton batch management from sampling and release to trend management. The benefit of trend management is that you can see slow degradation; when a batch is released, you can see what can't be seen.
The cleanliness of the dipping stage also affects the skeleton. Impurities in the paint penetrate the skeleton surface, and the weak point of dielectric is at the impurities. Filtration and tank cleaning must have systems; don't let the saved filter cartridge money turn into after-sales work orders. Some factories have dielectric defect rates that fluctuate seasonally; tracing the fact that the paint deteriorates quickly in summer and the filtration cycle is not adjusted accordingly. Process parameters follow the seasons; this principle works equally for impregnating paint.
A material selection and coordination suggestion for transformer factories: arrange a joint verification of the paint, encapsulation materials, and framework suppliers. Problems at the interface between the three systems can never be detected by individual verification. The cost of joint verification is lower than any batch rework. Interface issues are solved by joint meetings, which is a mature experience in electronics manufacturing and worth replicating for every skeleton project.
One more point on batch management of transformer skeleton: sample retention and traceability. The dielectric data of the skeleton must be traceable down to material batches and molding parameters; retained samples are the physical anchor point for traceability. Some factories arrange sample rooms by batch; this trump card is the biggest bonus during customer audits. Half of the customer's trust in suppliers comes from the moment they enter the sample room. The cost of retaining samples is just a few cents per item, and trust value is invaluable.
The relationship between skeleton and safety certification must be clearly explained. Transformer products have had safety certifications, and the frame as part of the insulation system is within the scope of certification. Material replacements require re-evaluation of safety regulations, and this cost should be included in the project in advance. Some factories missed safety re-evaluation during material replacement, causing the entire batch to be stuck at customs. The relationship between certification and materials is the first line on the framework project schedule. Before scheduling, the certification path must be drawn out first to avoid supply shortages midway.
Manufacturer's final material selection advice: treat joint verification of paint, encapsulation, and framework systems as standard measures, and treat safety pathways as the first line of the schedule. If these two things are done properly, the failure rate of framework projects will plummet. The industry's mature experience is right there; copying operations is not shameful, rework is.
Frame project final cross-cultural experience. Japanese clients require batch reports for dielectric batch data for frameworks; European and American clients require long-term data curves; domestic customers value response speed most. Three types of customers have three expectations, and service plans should be tiered accordingly. One factory has made three sets of document package templates, so taking orders from different clients is easy. Understanding the customer's inspection logic is more effective than going to multiple sites.
Frame failure Another niche scenario to mention: rodent bites and insect damage. Outdoor and warehouse environments where frames are damaged by rodent bites can cause dielectric failure without warning. Rodent-proof designs in outdoor transformers are not a joke; they are real failures. The design of housing holes and material hardness both affect the risk of mouse bites. Niche entries in the industry experience database are often the only answer to a strange after-sales incident worth accumulating slowly.
Conclusion
Transformer Framework Material Selection: The judgment chain is:
first set the temperature resistance level for insulation grade →, then set CTI → by voltage and space, then determine structure by terminals and winding→ and finally select the specific grade.
Each of the four failures has its own pipeline; any missing one will be exposed during the verification phase
