结论：断电抱闸是设计意图，能不能做到要验证
电动轮椅普遍采用电磁制动器：通电时释放，断电时靠弹簧抱紧。这个设计的意图很明确——任何导致断电的故障，都会让制动自动生效，属于失效安全设计。
但设计意图与实际表现之间有差距。断电之后抱闸需要时间；弹簧会疲劳；摩擦面会磨损；控制回路可能在不该断电时断电，也可能在该断电时没断。这些都需要验证。
需要验证的故障场景
| 场景 | 期望行为 | 验证方法 |
|---|---|---|
| 主电源切断 | 立即抱闸并保持 | 行驶中切断主电源 |
| 控制器失效 | 输出切断，抱闸生效 | 模拟控制器异常 |
| 制动器线路断开 | 抱闸生效 | 断开制动器供电线 |
| 电池电量耗尽 | 平缓停车后抱闸 | 低电量条件下持续运行 |
| 急停操作 | 立即抱闸 | 行驶中触发急停 |
| 手动释放后恢复 | 恢复正常制动 | 操作释放杆后复位 |
关键是所有场景都要在行驶状态下验证，而不是静止时。静止时断电，制动器抱住了，说明不了问题；行驶中断电，车辆在多长距离内停下、有没有溜车，才是要考察的。
响应时间的测量
从断电指令发出到制动力完全建立，有一段时间。这段时间内车辆仍在滑行。
测量方法是记录断电时刻与车辆完全停止之间的过程，同时区分两部分：制动器机械响应的时间，和车辆从减速到停止的时间。前者是制动器本身的特性，后者还受整车质量和速度影响。
响应时间偏长时，要区分是电路延迟还是机械延迟。 电路延迟可以通过改变控制逻辑改善，机械延迟取决于弹簧力和衔铁行程，要改结构。
保持力随磨损的变化
电磁制动器的保持力依赖摩擦面状态。使用中摩擦面会磨损，间隙增大，保持力下降。
验证这一点需要耐久试验：按预期的制动次数做循环，之后重新测量保持力。对比耐久前后的数值，才知道设计余量够不够。
这里有个容易忽略的因素：电动轮椅的制动器在正常使用中动作频繁——每次停车都是一次动作。所以循环次数的估算应当基于实际使用频次，这个数字通常比直觉大。
另外，摩擦面如果沾染油污（比如附近部件的润滑脂渗出），保持力会明显下降。建议在验证中包含污染条件，或者在设计上保证摩擦面与润滑部位隔离。
手动释放机构的验证
多数电动轮椅设有手动释放装置，用于断电时推行。这个机构带来额外的验证需求：
释放后制动完全解除，推行阻力在可接受范围；恢复时制动能可靠重新接合；释放状态下车辆不能被电驱动，这是重要的安全联锁——如果在释放状态下还能通电行驶，制动就失效了；释放状态有明确的提示，避免使用者忘记复位。
最后一条在实际中很重要。有的使用者或护理者释放后忘记复位，车辆在无制动状态下使用，坡道上会直接溜车。
与整车安全逻辑的配合
制动器不是独立工作的，它与控制器的安全逻辑配合。需要验证的配合点包括：
控制器检测到异常时，是否正确切断制动器供电；制动器状态是否有反馈给控制器，控制器能否知道制动是否真的生效；在制动生效期间，电机输出是否被禁止，避免电机与制动器对抗。
最后一项如果没做好，会导致电机和制动器互相较劲，结果是发热、磨损加剧，严重时烧毁。
数据怎么用于维护建议
耐久数据可以转成维护指引：制动器的检查周期、需要检查什么、保持力下降到什么程度应当更换。
对使用者而言，可以给一个简单的自检方法，比如在坡道上停稳后观察是否溜车。把专业指标转成使用者能操作的检查动作，比在说明书里写一个力值有用得多。
制动器选型时的考虑
电磁制动器多为外购件，选型时除了额定制动力矩，还应当关注几项：
断电到抱闸的响应时间；额定动作次数及其测试条件；在你的安装方式下的实际散热条件；手动释放接口的形式与可靠性；以及供应商能否提供失效模式的说明。
额定动作次数的测试条件尤其要问清楚——供应商给的数字可能是在空载或者理想散热条件下测的，与你的实际工况差距可能很大。
与坡道驻车的关系
制动器的保持力直接决定坡道驻车能力，而这是使用者日常会遇到的场景：在坡道上停下来等人、开门、接电话。
验证时除了测能不能保持，还应当测保持的时间——短时保持与长时间保持是不同的考察。有些制动器在持续受力下会缓慢滑移，短时间看不出来，停十几分钟就明显了。
制动器的日常检查
制动性能会随使用衰减，而使用者通常不会主动检查。产品设计上可以考虑加入便于检查的特征，比如摩擦片磨损指示、或者在显示界面提供制动自检功能。
如果没有这些特征，说明书里至少要给出简单可操作的检查方法和频次建议。检查方法要使用者做得到——要求测量制动力矩不现实，在坡道上观察是否溜车才是可操作的。
我们的做法
做制动器验证时，我们会分别记录响应时间、制动距离和保持力，并在耐久前后各测一次保持力。只测新件状态的制动距离，反映不出这个部件的真实可靠性。
对于安全要求较高的产品，建议把故障注入类的验证做全：逐个模拟前面列的故障场景，记录每种情况下的实际行为。这组数据在风险管理文件里能直接用作控制措施有效性的证据。
有需要可以把制动系统方案和控制逻辑说明发过来一起设计验证，或者直接联系：132 4819 8029。检测能力见服务介绍，产品分类见电动轮椅检测与代步车检测，标准信息见标准查询。
English version
Conclusion. Powered wheelchairs generally use electromagnetic brakes that release when energised and clamp under spring force when de-energised. The design intent is clear: any fault causing loss of power makes braking take effect automatically, which is fail-safe behaviour. Intent and actual behaviour differ, however. Clamping takes time after power is lost, springs fatigue, friction surfaces wear, and the control circuit may de-energise when it should not or fail to when it should. All of this needs verification.
Fault scenarios to verify. Loss of main power should produce immediate clamping and holding, verified by cutting main power while moving. Controller failure should cut output and engage the brake, verified by simulating controller malfunction. A break in the brake supply line should engage the brake. Battery depletion should produce a controlled stop followed by clamping, verified by running at low state of charge. An emergency stop input should clamp immediately. And restoration after manual release should return normal braking.
The critical point is that every scenario must be verified while the chair is moving rather than stationary. Cutting power at rest and observing that the brake holds proves little. What matters is the distance travelled after power is lost while moving, and whether the chair creeps afterwards.
Measuring response time. There is an interval between the de-energising command and full braking force, during which the chair continues to travel. Measure from the moment of de-energisation to a complete stop, distinguishing the mechanical response of the brake from the deceleration of the vehicle, since the former is a property of the brake and the latter also depends on mass and speed. Where response is slow, distinguish circuit delay from mechanical delay: circuit delay can be improved through control logic, while mechanical delay depends on spring force and armature travel and requires a structural change.
Holding force and wear. Holding capability depends on the condition of the friction surfaces, which wear with use, increasing clearance and reducing holding force. Verifying this requires endurance testing to the expected number of brake applications followed by remeasurement of holding force, because the comparison before and after is what reveals whether design margin is adequate. An easily overlooked factor is that a powered wheelchair brake operates frequently, once per stop, so the cycle estimate should come from actual use frequency and is usually larger than intuition suggests. Contamination also matters: grease migrating from adjacent components onto the friction surface markedly reduces holding force, so verification should include a contaminated condition or the design should isolate the friction surface from lubricated parts.
Manual release mechanisms. Most powered chairs include a manual release for pushing when unpowered, which adds verification requirements. Braking must fully release with acceptable pushing resistance. Braking must re-engage reliably on restoration. Electric drive must be inhibited while released, which is an important interlock, since drivable operation without braking means braking has effectively failed. And the released state must be clearly indicated, because users and carers do forget to reset it, leaving the chair in use without braking and free to run away on a slope.
Integration with vehicle safety logic. The brake works together with controller safety logic. Verify that the controller correctly removes brake power when it detects a fault; that brake state is fed back so the controller knows whether braking is actually engaged; and that motor output is inhibited while braking is engaged. Without the last, motor and brake work against each other, producing heat and accelerated wear and in severe cases damage.
Turning data into maintenance guidance. Endurance data support inspection intervals, what to inspect, and the point at which reduced holding force warrants replacement. For users, provide a simple self-check such as stopping on a slope and observing whether the chair creeps. Translating a technical figure into an action a user can perform is considerably more useful than printing a force value in the manual.
How we handle it. We record response time, stopping distance and holding force separately, and measure holding force both before and after endurance. Measuring stopping distance on a new unit alone does not reflect the real reliability of this component. For products with higher safety requirements we suggest running the full set of fault injections and recording actual behaviour in each case; that dataset serves directly as evidence of control effectiveness in the risk management file.
Send us the braking system design and control logic and we will design the verification. Phone or WeChat: +86 132 4819 8029.