结论：限制说明不是套话，它界定了报告能证明什么
检测报告上通常有一些限制性的说明——「本报告仅对来样负责」「未经允许不得部分复制」「报告无检测专用章无效」。
很多人把这些当成套话跳过，但它们实际上界定了报告的适用范围 ——报告能证明什么、不能证明什么、在什么条件下有效。
读懂这些说明，才能正确使用报告，也才能判断别人提供的报告能支撑什么。
常见的限制条款
| 条款 | 含义 |
|---|---|
| 仅对来样负责 | 结论只针对送检的那些样品 |
| 送样检验 | 样品由委托方提供，非抽样 |
| 不得部分复制 | 摘取部分内容可能造成误导 |
| 需盖章有效 | 无章的复印件不能作为依据 |
| 仅限委托项目 | 只测了委托的项目 |
| 依据的标准与版本 | 按某个特定版本判定 |
| 有效期或时效 | 部分报告有时效要求 |
「仅对来样负责」的含义
这是最重要也最常被忽略的一条。 它的意思是：
结论只针对送检的那几台样品。 不是对整批产品、也不是对该型号的所有产品。
不能推广到量产品。 除非能论证送检样品代表量产状态。
不能推广到其他批次。 后续批次的一致性需要其他证据。
不能推广到其他型号。 除非有覆盖论证。
「样品代表性」是这条限制的核心。 送检样品如果是挑选出来的、或者是试制样品、或者工艺与量产不同，报告的结论与量产品的实际情况可能有差距。
所以采购时看到报告，还应当核对交付产品与报告中样品的一致性。
「送样检验」与「抽样检验」
送样检验。 样品由委托方送来，实验室不参与挑选。样品的代表性由委托方保证。
抽样检验。 实验室或第三方按规则抽取样品。代表性更好。
报告中会注明是哪一种，这个信息对判断数据的代表性有意义。
送样检验不等于数据不可信，但它意味着样品的选择不受实验室控制。
「仅限委托项目」的含义
报告只覆盖实际测了的项目。 没测的项目报告不涉及。
不能引申为全面符合。 报告中测了六项合格，不能说明产品符合该标准的全部要求。
核对项目清单。 使用报告时应当对照标准的项目清单，看是否有遗漏。
这是超范围引用最常发生的地方 ——报告覆盖部分项目，却被用来证明全面符合。
标准与版本的限制
按特定版本判定。 报告依据某个版本的标准，标准更新后结论可能变化。
版本差异的影响。 新版本可能有新增项目或更严的要求。
时效的考虑。 标准更新后，旧版本的报告是否仍被接受，要看具体要求。
建议关注报告依据的标准版本，特别是在标准刚更新的时期。
超范围使用的风险
用于其他型号。 缺少覆盖论证。
声称未测项目符合。 超出报告范围。
引用时省略条件。 数据脱离测试条件。
部分摘取造成误导。 只引用有利的部分。
用过期的报告。 超出时效要求。
这几种使用都可能在审查、招标、或者纠纷中带来问题，而它们在使用报告时是可以避免的。
正确使用报告的建议
完整阅读。 包括限制说明。
核对适用范围。 型号、规格、配置。
核对项目覆盖。 是否覆盖需要证明的内容。
核对条件。 测试条件与实际使用或要求的对应。
核对时效。 是否在要求的时效内。
引用时带条件。 引用数据时一并说明测试条件。
不超范围引申。 报告没说的不要声称。
「引用时带条件」是简单但有效的做法。 说「续航多少公里」不如说「在某载重、某路面、某温度下续航多少公里」，后者准确且不易被质疑。
作为报告使用方的核对清单
拿到一份报告，建议核对：报告编号与出具机构；样品型号与规格是否与实际产品一致；样品来源（送样还是抽样）；测试项目是否覆盖需要的范围；依据标准及版本是否适用；实测数据及条件；报告日期与时效；以及限制说明的内容。
这份清单几分钟能过完，但能避免使用报告时的大部分问题。
不同用途对报告的要求
| 用途 | 对报告的要求 |
|---|---|
| 注册申报 | 覆盖全部适用要求、机构资质符合 |
| 招标投标 | 覆盖招标参数、时效符合 |
| 采购验收 | 与交付产品对应 |
| 内部研发 | 数据详细、条件清楚 |
| 质量追溯 | 与批次关联 |
| 宣传使用 | 表述严格对应数据 |
「宣传使用」的要求最严，因为超范围引用在宣传中是明确的风险。
报告的保存与管理
保存期限。 与产品生命周期和法规要求对应。
电子与纸质。 两种形式的管理。
版本管理。 如果报告有修订版本。
关联管理。 报告与产品型号、批次、变更的关联。
检索。 需要时能快速找到。
建议建立报告台账，记录每份报告的编号、日期、覆盖型号、覆盖项目、依据标准、以及到期时间。
常见的使用错误
跳过限制说明。 不知道报告的边界。
只看结论页。 忽略样品信息和条件。
摘取部分引用。 造成误导。
用于不同型号。 缺少覆盖论证。
忽略报告时效。 使用过期报告。
这五项是使用检测报告时最常见的问题，而读一遍限制说明就能避免大部分。
与供应商沟通的要点
拿到报告有疑问时，值得问清楚的：未测项目是什么情况；样品与交付产品的差异；宣传参数的报告依据；以及是否有更完整的报告。
这几个问题的答案能看出供应商的准备程度，也能避免后续的争议。
我们的做法
我们的报告会明确说明样品的来源、测试的项目范围、依据的标准版本、以及适用的限制。这些说明不是免责，而是界定报告能证明什么。
对于不在认可范围内的项目，我们会明确标注，避免报告被超范围引用。
如果委托方对报告的使用范围有疑问，或者需要针对特定用途出具说明，可以与我们沟通——明确的界定对双方都有好处。
如果你在使用或核对检测报告时有疑问，可以直接联系：132 4819 8029。检测能力见服务介绍，产品分类见手动轮椅检测与电动轮椅检测，案例见案例。
English version
Conclusion. Test reports normally carry limiting statements: the report relates only to the samples submitted, partial reproduction is not permitted, the report is invalid without the testing seal. Many readers skip these as boilerplate, when in fact they define the report's scope: what it can demonstrate, what it cannot, and under what conditions it holds. Understanding them is what allows a report to be used correctly and what allows someone else's report to be judged for what it supports.
Common limiting clauses. Relates only to the samples submitted means the conclusion applies to those samples. Submitted samples means the client supplied them rather than the laboratory drawing them. Partial reproduction not permitted means extracting parts may mislead. Valid only with the seal means unsealed copies are not evidence. Limited to the commissioned items means only those items were tested. The standard and version relied upon means judgement followed a particular edition. And validity or currency means some reports carry time requirements.
What relating only to the samples means. This is the most important clause and the most overlooked. It means the conclusion applies to those particular samples, not to a batch or to every unit of the model. It cannot be extended to production unless the samples can be shown to represent production condition. It cannot be extended to other batches, whose consistency needs other evidence. And it cannot be extended to other models without a coverage argument. Sample representativeness is the core of this limitation: where submitted samples were selected, or were pilot builds, or were made by a process differing from production, the report's conclusion may diverge from production reality. When buying, therefore, check that delivered product matches the samples described in the report.
Submitted versus sampled. Submitted testing means the client sent the samples and the laboratory played no part in selecting them, so representativeness rests with the client. Sampled testing means the laboratory or a third party drew samples by rule, giving better representativeness. Reports state which applied, and that information bears on how representative the data are. Submitted testing does not mean the data are unreliable, but it does mean sample selection lay outside the laboratory's control.
What limited to the commissioned items means. The report covers only the items actually tested; those not tested are not addressed. It cannot be extended into a claim of full conformity: six items passing does not demonstrate conformity with every requirement of a standard. Check the item list against the standard's when using a report, to see what is missing. This is where scope overreach most often occurs: a report covering some items is used to demonstrate full conformity.
Standard and version limitations. Judgement follows a particular edition, and conclusions may change when the standard is revised. Version differences matter, since a new edition may add items or tighten requirements. And currency considerations arise after revision, with acceptance of reports to the old edition depending on the specific requirement. Watch the standard edition a report relies upon, particularly in the period after a revision.
Risks of overreach. Applying a report to other models without a coverage argument. Claiming conformity for untested items, beyond the report's scope. Citing data without their conditions, detaching them from the test conditions. Extracting parts selectively, quoting only favourable content. And using expired reports beyond the currency requirement. Each of these can cause problems during review, tendering or dispute, and each is avoidable in how the report is used.
Using reports correctly. Read the whole report, including the limitations. Check the applicable scope covering model, specification and configuration. Check item coverage against what needs demonstrating. Check conditions against actual use or requirements. Check currency against the requirement. Cite data with their conditions. And do not extrapolate beyond what the report states. Citing with conditions is simple and effective: stating a range in kilometres under a specified load, surface and temperature is accurate and hard to challenge, while a bare figure is neither.
A checklist for report users. On receiving a report, check the report number and issuing laboratory; whether the sample model and specification match the actual product; sample provenance, whether submitted or drawn; whether the items cover what needs demonstrating; whether the standard and edition apply; the measured data and their conditions; the report date and currency; and the content of the limiting statements. The checklist takes a few minutes and avoids most problems in using reports.
How we handle it. Our reports state sample provenance, the scope of items tested, the standard edition relied upon and the applicable limitations. Those statements are not disclaimers but a definition of what the report demonstrates. Items outside our accreditation scope are clearly annotated, so that reports cannot be cited beyond their scope. Where clients have questions about the scope of use, or need a statement for a particular purpose, we are glad to discuss it: clear definition benefits both sides.
If you have questions about using or checking a test report, get in touch: +86 132 4819 8029.