📚 Practical Work in Physics | 物理实验操作
Practical work is at the heart of A-Level Physics. It is not merely about following a set of instructions but about developing the skills to think like a scientist: asking questions, designing procedures, taking careful measurements, analysing data, and evaluating limitations. In the Cambridge International (CIE) syllabus, practical skills are assessed through Paper 3 (Advanced Practical Skills) and Paper 5 (Planning, Analysis and Evaluation). This article provides a comprehensive guide to the core competencies required, from planning an experiment to presenting and interpreting results, with an emphasis on uncertainties, graphing, and improvement strategies.
实验操作是 A-Level 物理的核心。它不仅仅是按照指令操作，更是培养像科学家一样思考的能力：提出问题、设计步骤、仔细测量、分析数据并评估局限性。在剑桥国际 (CIE) 考纲中，实验技能通过 Paper 3（高级实验技能）和 Paper 5（规划、分析与评估）进行考核。本文全面梳理了所需的核心能力，涵盖从实验规划到结果呈现与解读的各个方面，重点包括不确定度、作图以及改进策略。
1. Planning an Experiment | 规划实验
A well-planned experiment begins with a clear identification of the independent, dependent, and control variables. You must state how each variable will be measured or kept constant, specify the range and interval of values for the independent variable, and justify these choices. The plan should include a labelled diagram of the apparatus, a step-by-step method, and a description of how to obtain reliable data – for instance, by repeating measurements and calculating a mean. Consideration must also be given to safety precautions and any potential hazards.
一个精心规划的实验首先需要明确自变量、因变量和控制变量。你需要说明每个变量如何测量或保持恒定，指定自变量的取值区间和间隔，并说明理由。方案中应有带标注的装置示意图、分步操作步骤，并描述如何获得可靠数据——例如，通过重复测量并计算平均值。还需考虑安全措施及任何潜在危险。
When planning, always ask: "What relationship am I testing?" and "What graph will I plot to verify it?" For example, if investigating the period T of a pendulum, you might plan to vary length L, measure T for several lengths, and then plot T² against L, expecting a straight line through the origin whose gradient gives 4π²/g. The plan should detail how to measure L with a metre rule and T with a stopwatch using the fiducial mark technique to reduce reaction-time errors.
在规划时，始终要问自己：“我要验证什么关系？”以及“我将绘制什么图像来验证？”例如，若研究单摆的周期 T，你可以计划改变摆长 L，测量不同长度下的 T，然后绘制 T²-L 图，期望得到一条通过原点的直线，其斜率等于 4π²/g。方案应详述如何使用米尺测量 L，以及如何使用停表和基准标记法测量 T 以减小反应时间误差。
2. Measuring Instruments and Reading | 测量工具与读数
uncertainty linked to the instrument used. For analogue instruments such as a metre rule or a moving-coil voltmeter, the absolute uncertainty is usually taken as half the smallest scale division. For digital instruments, such as a digital multimeter or electronic balance, the uncertainty is typically the smallest digit or the manufacturer' t significant digit. You must be able to read instruments to the appropriate precision, using a vernier scale or a micrometer screw gauge where sub-millimetre accuracy is required.
每一项测量都涉及与所用仪器相关的不确定度。对于模拟式仪器，如米尺或动圈式电压表，绝对不确定度通常取最小分度值的一半。对于数字仪器，如数字万用表或电子天平，不确定度通常为最后一位有效数字的 ±1 或者制造商给出的精度。你必须能够以适当的精度读取仪器读数，当需要亚毫米精度时，应使用游标卡尺或螺旋测微器。
uler would be read to the nearest mm, but in practice you can estimate to the nearest half‑ meaning the true mass lies between 12.34 g and 12.36 g. Always record all the digits provided by a digital instrument, including trailing zeros, as they indicate precision.
例如，具有毫米刻度的米尺的绝对不确定度为 ±0.5 mm。一把 30 cm 的尺子可读数至最接近的毫米，但实际上你可以估读到半毫米，因此读数不确定度为 ±0.5 mm。一台显示 12.35 g（保留两位小数）的数字天平的不确定度为 ±0.01 g，意味着真实质量介于 12.34 g 和 12.36 g 之间。务必记录数字仪器给出的所有数字，包括末尾的零，因为它们标示了精度。
3. Uncertainties and Errors | 不确定度与误差
h the true value probably lies. The percentage uncertainty is calculated as (Δx / x) × 100% olute uncertainties add; for multiplication and division, percentage uncertainties add; when a quantity is raised to a power n, the percentage uncertainty is multiplied by n.
不确定度量化了测量结果的存疑程度。绝对不确定度 Δx 表明了真值可能所在的范围 x ± Δx。百分数不确定度的计算公式为 (Δx / x) × 100%。当组合测量值时，计算量的不确定度按以下规则传播：加减运算时，绝对不确定度相加；乘除运算时，百分数不确定度相加；当某个量被乘 n 次方时，其百分数不确定度乘以 n。
Consider an object' / V, where m = 24.0 ± 0.1 g and V = 8.0 ± 0.3 cm³. The percentage uncertainty in m is 0.42%, in V is 3.75%, giving a total percentage uncertainty in ρ of about 4.2% e the uncertainty to one significant figure and round the value to the same decimal place.
假设某物体的密度 ρ = m / V，其中 m = 24.0 ± 0.1 g，V = 8.0 ± 0.3 cm³。质量的百分数不确定度为 0.42%，体积的为 3.75%，因此 ρ 的总百分数不确定度约为 4.2%。计算出的密度为 3.0 g cm⁻³，绝对不确定度为 0.13 g cm⁻³，所以 ρ = 3.0 ± 0.1 g cm⁻³。你始终应将不确定度保留一位有效数字，并将数值四舍五入到相同的小数位。
4. Data Recording and Tables | 数据记录与表格
Data should be recorded in a neat, logically organized table with clear headings. Each column heading must include the quantity and its unit, separated by a forward slash, e.g., 'Length L / cm' or 'Potential difference V / V'. In a well-designed table, the independent variable appears in the first column, and subsequent columns contain repeated readings and the mean value. All raw data should be recorded to the precision of the instrument, and calculated quantities must be given to the correct number of significant figures.
数据应记录在清晰、逻辑分明的表格中，并配有明确的表头。每一个列标题都必须包含物理量和单位，用斜杠分隔，例如“长度 L / cm”或“电势差 V / V”。在设计良好的表格中，自变量位于第一列，后续各列依次为重复读数以及平均值。所有原始数据都应记录到仪器的精度，计算得出的物理量必须给出正确的有效数字位数。
For instance, when measuring the time for a ball to fall from different heights, a table might look like this:
| Height h / m | Time t₁ / s | Time t₂ / s | Time t₃ / s | Mean |
| 1.00 | 0.45 | 0.47 | 0.46 | 0.46 |
Notice the use of repeated readings to identify anomalies and improve reliability. Anomalous results should be circled but not removed unless a clear mistake has been made; if omitted, a reason must be stated.
例如，在测量小球从不同高度下落的时间时，表格可能如上所示。注意重复读数的运用，可以识别异常值并提高可靠性。异常结果应圈出，除非明确发生了操作失误，否则不应删除；若需省略，则必须说明理由。
5. Graph Plotting and Analysis | 作图与分析
A graph is the most powerful tool for revealing the relationship between variables. Axes must be labelled with the quantity and unit, scales should be chosen so that the plotted points occupy more than half the graph paper in each direction, and points should be plotted with small crosses or encircled dots. Error bars representing the absolute uncertainty of each point should be added if the question demands it or if the uncertainties are large enough to matter.
图像是揭示变量间关系最有力的工具。坐标轴必须标注物理量和单位，刻度尺应选择得当，使得各方向上的数据点占据坐标纸一半以上的面积，数据点用小叉号或加圈的点表示。如果题目要求或不确定度足够显著，应添加代表每个点绝对不确定度的误差线。
A line of best fit is drawn to pass through as many error bars as possible and to balance the scattering of points. This line may be straight or curved, but in A-Level Physics most experiments are designed to yield a straight line. The 'worst acceptable line' – the steepest or shallowest line that still passes through the error bars – is used to find the uncertainty in the gradient and y‑intercept. Always calculate the gradient using a large triangle drawn on the line of best fit, not by using plotted data points directly.
最佳拟合线的绘制应尽可能多地穿过误差线，并使散点的分布达到平衡。这条线可以是直线也可以是曲线，但在 A-Level 物理中，大多数实验均设计成能得到一条直线。“最差可接受线”——即穿过误差线的最陡或最缓的直线——用于求取斜率和 y 轴截距的不确定度。务必使用在最佳拟合线上所画的大三角形来计算斜率，而不是直接使用原始数据点。
6. Gradient and Intercept | 斜率与截距
two well-separated points (x₁, y₁) and (x₂, y₂) on the line of best fit. The gradient is:
m = (y₂ − y₁) / (x₂ − x₁)
The y-intercept c is read directly from the point where the line crosses the y‑axis. Both m and c carry units and physical significance. For example, in a graph of V against I for a fixed resistor, the gradient corresponds to resistance R, and the intercept is ideally zero.
直线图 y = mx + c 的斜率 m 是通过在最佳拟合线上选取两个相距较远的点 (x₁, y₁) 和 (x₂, y₂) 计算得出的。斜率公式如上。y 轴截距 c 可直接从直线与 y 轴的交点读取。m 和 c 都带有单位并具有物理意义。例如，在固定电阻的 V-I 图中，斜率对应于电阻 R，截距理想情况下为零。
To find the uncertainty in the gradient, plot the worst acceptable line, calculate its gradient mw, then:
Δm = |mbest − mw|
The percentage uncertainty in the gradient is Δm / m × 100% d from the gradient, such as the acceleration due to gravity g from a pendulum experiment.
要计算斜率的不确定度，可画出最差可接受线，算出其斜率 mw，然后按上述公式计算。斜率的百分数不确定度为 Δm / m × 100%。该不确定度随后可用来估算由斜率导出的任何物理量的不确定度，例如单摆实验中的重力加速度 g。
7. Systematic and Random Errors | 系统误差与随机误差
Random errors cause readings to be scattered around the true value. They can be reduced by taking many readings and averaging, and they affect precision. Systematic errors shift all readings in one direction, either by a constant amount or by a constant percentage, and they affect accuracy. Common systematic errors include zero errors on instruments, incorrect calibration, and parallax error if the eye is not positioned perpendicularly to the scale.
随机误差导致读数在真值周围离散分布。通过多次读数取平均值可以减小随机误差，它们影响精密度。系统误差会使所有读数朝同一方向偏移，可能是常量偏移或按恒定百分比偏移，它们影响准确度。常见的系统误差包括仪器的零点误差、校准不当，以及眼睛未与刻度垂直所产生的视差。
In an experiment using a stretched spring, if the metre rule is placed with its zero end 1 cm above the unstretched position, all extensions will be 1 cm too small – a systematic error. This error cannot be eliminated by averaging; it must be corrected by re-zeroing or by subtracting the offset. Random errors might arise from timing oscillations by hand: sometimes you start the stopwatch early, sometimes late, giving scatter in the period values.
在使用拉伸弹簧的实验中，若米尺的零端位于未拉伸位置上方 1 cm 处，则所有伸长量都会偏小 1 cm——这就是系统误差。此类误差无法通过取平均值消除，必须通过重新调零或减去偏移量来修正。随机误差可能源于手动对振荡进行计时：有时你过早按下停表，有时过晚，从而造成周期值的离散。
8. Improving Precision and Accuracy | 提高精密度与准确度
To improve precision, you should use instruments with a finer resolution, take multiple readings and calculate a mean, or use data-logging sensors that record rapidly and eliminate human reaction time. For example, replacing a stopwatch with a light gate connected to a timer dramatically reduces the uncertainty in the measurement of time intervals. Repeating the experiment under the same conditions and checking the spread of results also gives confidence in the precision.
要提高精密度，应使用分辨率更高的仪器，多次读数并求平均，或采用能快速记录并消除人为反应时间的数据采集传感器。例如，用连接计时器的光门代替停表，可显著减小时间间隔测量的不确定度。在相同条件下重复实验并检查结果的离散程度，同样能增强对精密度的信心。
Accuracy improvements target systematic errors. Zero errors should be checked before each set of measurements and corrected. Parallax errors can be avoided by placing the eye directly in front of the pointer, or by using a mirror behind the scale. Calibrating an instrument against a known standard, such as checking a voltmeter against a calibrated source, can also improve accuracy. In electrical experiments, using short, thick connecting wires minimises the internal resistance effect, which often introduces systematic error.
提高准确度则针对系统误差。在每组测量之前应检查并修正零点误差。通过将眼睛正对指针，或在刻度后方设置镜面可避免视差。对照已知标准校准仪器，例如用校准源检查电压表，也能提高准确度。在电学实验中，使用短而粗的导线可以减小常引入系统误差的内阻效应。
9. Safety in Practical Work | 实验安全
Safety considerations are a mandatory part of any practical plan. You should identify specific hazards associated with the apparatus or materials – for instance, hot objects, heavy masses, sharp electrodes, or high voltages – and state practical precautions to minimise risk. These precautions must be relevant and not generic: instead of "wear safety goggles", say "wear safety goggles to protect eyes from a possible spring snapping under tension".
安全考量是任何实验方案中必不可少的一部分。你应该指出与装置或材料相关的具体危险——例如高温物体、重物、尖锐的电极或高电压——并说明降低风险的实际预防措施。这些措施必须具有针对性而非泛泛而谈：与其说“佩戴护目镜”，不如说“佩戴护目镜以防弹簧在张力下断裂伤及眼睛”。
In a free‑fall experiment, a heavy metal ball dropped from a height poses a crush injury risk; a sand tray should be placed beneath to catch the ball safely. In an electrical circuit, a fuse or a circuit breaker should be included, and the voltage supply should be kept low. When heating water, a Bunsen burner poses a fire and burn hazard; tie back long hair, use a heat‑resistant mat, and allow apparatus to cool before handling. Always perform a risk assessment before beginning.
在自由落体实验中，从高处落下的重型金属球有压伤风险，应在下方放置沙盘以安全承接落球。在电路中应设置保险丝或断路器，并保持电源电压较低。加热水时，本生灯有火灾和烫伤危险；应将长发束起，使用耐热垫，并在操作前让装置冷却。每次实验前都应进行风险评估。
10. Common Experiments and Their Pitfalls | 常见实验及其陷阱
Certain classic experiments appear repeatedly in CIE practical exams. Understanding their common pitfalls helps you avoid losing marks. The pendulum experiment requires measuring from the point of suspension to the centre of mass of the bob; a common mistake is measuring only to the top of the bob, introducing a systematic error. For oscillations, time at least 10 periods to reduce the impact of reaction time, and count from zero – that is, "zero, one, two...".
某些经典实验在 CIE 实验考试中反复出现。了解其常见陷阱有助于避免失分。单摆实验需要从悬挂点测量到摆球质心；常见的错误是只量到摆球顶端，从而引入系统误差。对于振荡，至少要计时 10 个周期，以减小反应时间的影响，并且从零开始计数——即“零、一、二……”。
When using a micrometer screw gauge, always check for zero error by closing the jaws gently and reading the sleeve; correct subsequent readings accordingly. In electricity experiments with a potentiometer or a metre‑bridge, ensure tight connections and clean contacts, as loose connections introduce erratic resistance. In heat experiments, stir well and wait for the thermometer reading to steady before recording temperature, and always insulate the container to reduce heat exchange with the surroundings.
使用螺旋测微器时，务必检查零点误差，方法是轻合量爪并读取套筒刻度，然后据此修正后续读数。在使用电位差计或滑线电桥的电学实验中，确保连接牢固、触点清洁，因为松动的连接会引入不稳定的电阻。热学实验中，要搅拌均匀并等待温度计读数稳定后再记录，同时务必对容器进行保温，以减少与周围环境的热交换。
11. Combining Uncertainties from a Graph | 结合图像的不确定度
antity that is obtained by combining the gradient and the intercept, or by using the graph' plot T² against L. The gradient m = 4π²/g, so g = 4π²/m. The uncertainty in g is found by:
Δg/g = Δm/m
because percentage uncertainties in a power relationship multiply accordingly. If Δm/m = 2%, then Δg/g = 2% and you can quote g with its absolute uncertainty.
通常你需要求出由斜率和截距组合而成，或利用图像趋势得到的某个量的不确定度。例如，在利用单摆测量 g 的实验中，你绘制 T²-L 图。斜率 m = 4π²/g，因此 g = 4π²/m。g 的不确定度可由上式给出。这是因为百分数不确定度在乘方关系中是直接传递的。若 Δm/m = 2%，则 Δg/g = 2%，进而可给出带有绝对不确定度的 g 值。
If the graph's y-intercept is expected to be zero but is not, you can determine whether the non‑zero value is significant by comparing it with its uncertainty. If |c| h zero. If not, there may be a systematic error, such as a misreading of the initial position, which should be discussed in the evaluation.
若图像的 y 轴截距预期为零但实际不为零，可以通过比较截距与其不确定度来判断该非零值是否显著。如果 |c| ≤ Δc，则截距与零一致。若不一致，则可能存在系统误差，例如初始位置读数有误，这应当在评估部分加以讨论。
12. Evaluation and Conclusion | 评估与结论
The final part of any practical write‑up is the evaluation. Here you should comment on the reliability of your results by comparing percentage uncertainties of measured quantities, identify the largest source of uncertainty, and suggest realistic improvements. A good evaluation does not simply say "use a more accurate instrument"; it explains which instrument, why it reduces the dominant uncertainty, and what effect that would have on the final result.
任何实验报告的最后一部分都是评估。在评估中，你应当通过比较各测量量的百分数不确定度来评论结果的可靠性，找出最大的不确定度来源，并提出切实可行的改进建议。一份好的评估不会简单地说“使用更精确的仪器”，而是说明具体使用哪种仪器，为什么它能减小主要的不确定度，以及这对最终结果会产生什么影响。
You should also state whether the experimental relationship proposed has been verified. If, for example, a straight line through the origin was predicted and your line of best fit passes comfortably through the origin within experimental uncertainty, you can conclude that the data supports the relationship. Any anomalous points should be discussed, and if they were discarded, justification must be provided. Finally, compare your derived value with the accepted value, if known, and calculate the percentage difference to assess accuracy.
你还需要说明所提出的实验关系是否得到验证。例如，假如预测的是一条通过原点的直线，而你的最佳拟合线在实验不确定度范围内很好地通过了原点，那么便可得出结论：数据支持该关系式。所有异常点都应讨论，若已删除，则必须提供理由。最后，如果已知公认值，可将你的推导值与公认值进行比较，计算百分数差值以评估准确度。
Published by TutorHao | Physics Revision Series | aleveler.com
更多咨询请联系16621398022（同微信）
或截图后用「扫一扫 → 相册」选择本图