GCSE物理必做实验 比热容 电阻 密度 波长
GCSE 物理的实验能力不再单独考，十项必做实验的读数、控制变量和误差分析全部塞进了笔试。下面按 AQA 8463 的顺序把这十项过一遍，每项给出关键公式、最容易扣分的操作细节，以及考场上通常从哪个角度提问。
Reformed GCSE Physics has no separate practical paper. The ten required practicals are marked through written questions, so technique, readings and error analysis count as Physics content rather than as laboratory housekeeping.
AQA 8463 的十项必做实验来自哪里
AQA 的 8463 大纲把十项必做实验写进正文，每项对应 DfE 在 2015 年公布的科学科目内容里列出的仪器与技术条目。这批内容 2016 年首次教学，2018 年首次考试，之后每年卷面上都会出现。Ofqual 要求实验相关题目至少占全部分数的 15%，所以这十项不是课堂表演，是实打实的分。
各考试局的清单并不一样。Edexcel 把核心实验归成八项，OCR Gateway 也是八项，只有 AQA 排到十项。做真题前先确认手里的清单属于哪个考试局，混着刷会漏项，也会把别家没要求的步骤当重点背。
There is no separate practical exam in reformed GCSE Physics. AQA 8463 lists ten required practicals, each mapped to the apparatus and techniques set out in the DfE subject content published in 2015, taught from 2016 and examined from 2018. Ofqual sets a floor of 15% of the marks for practical work. Other boards organise the same skills differently: Edexcel and OCR Gateway each run eight core practicals. Check which list your specification uses before starting past papers, because mixing boards hides gaps.
实验一 比热容 用 IVt 代替焦耳计
给金属块通电加热，用 E=IVt 算输入的电能，再用 c=E/(mΔT) 求比热容。铝大约 900 J/(kg·°C)，铜大约 385。实测值几乎总是偏高，原因有两个：热量散到了空气里，温度计本身也吸走一部分热。
读数顺序要固定下来：先记起始温度，通电同时开始计时，中途同时读电流和电压，断电立刻读末温。电流会随元件升温略微变化，只读一次就代入会带来偏差，间隔半分钟读三次取平均更稳。金属块外面包一层绝缘材料能明显压低散热，答案里写上这句往往就是一个给分点。
Heat the block electrically, take the input energy as current times voltage times time, then divide by mass and temperature rise. Aluminium sits near 900 J/(kg °C) and copper near 385, while measured values come out high because energy leaks to the room and the thermometer absorbs some of it. Reading current more than once and averaging removes drift as the element warms. Wrapping the block in insulation is worth a mark on its own, since it attacks the largest source of error rather than the smallest.
实验二 隔热材料 冷却曲线的斜率
同样的烧杯、同样体积的热水、同样的起始温度，只换包裹材料，每两分钟记一次温度，最后比较冷却曲线的斜率。曲线比单个读数可靠，因为它把整段时间的散热都摊开了，一次读数抖动不会直接决定结论。
控制变量写全才有分：水量、起始温度、容器、盖子、室温、包裹层数。常见的错误是把不同材料包在同一个烧杯上连着做，残留余热会让后测的材料显得更好。每换一种材料都得重新倒水、重新等温度稳定。
Identical beakers of hot water are wrapped in different materials and logged every two minutes, then the cooling curves are compared. A curve uses the whole cooling history, so one shaky reading cannot decide the conclusion. Reusing a beaker without letting it return to room temperature flatters whichever material is tested second. Fresh water and a fresh starting temperature for every trial is the only version of this practical that produces usable data.
实验三 电阻 导线长度与串并联
这项实验有两个方向：改变镍铬丝的长度看电阻怎么变，以及在串并联电路里测总电阻。R=V/I 要用很多次，接法不能错，电流表串联、电压表并联在待测元件两端。长度翻倍电阻大致翻倍，因为电阻与长度成正比、与横截面积成反比。
串联总电阻等于各电阻相加，并联总电阻比任何一个支路都小。测导线电阻时接触电阻会混进读数，用鳄鱼夹每次夹在相同位置、接线柱拧紧，能把这项误差压到可以忽略。
Resistance work covers wire length plus the totals of series and parallel networks. The ammeter goes in series with the component and the voltmeter across it, then R equals V divided by I. Doubling the length of nichrome roughly doubles resistance, and a parallel network always reads lower than its smallest branch. Contact resistance at the clips adds to every measurement, so clamping at the same point each time keeps the error consistent instead of random.
实验四 I-V 特性 灯丝 二极管 固定电阻
给固定电阻、灯丝灯泡和二极管分别加变化的电压，记录电流画 I-V 图。固定电阻得到过原点的直线；灯丝电流越大温度越高，电阻随之上升，曲线越到后面越平；二极管正向大约 0.6 到 0.7 伏才开始导通，反向几乎没有电流。
很多学生以为灯丝不遵守欧姆定律，其实欧姆定律的前提是温度不变，灯丝恰好不满足这个前提。答案写"灯丝温度升高导致电阻增大"比写"不符合欧姆定律"准确得多。反向那一小段电流很小但不是绝对零，写成完全为零会丢分。
Varying the voltage across three components separates their behaviour on one graph. A fixed resistor gives a straight line through the origin. A filament lamp curves because a hotter filament has higher resistance, so current grows more slowly at high voltage. A diode stays effectively off until about 0.6 to 0.7 V in forward bias, and the small reverse leakage is tiny but never exactly zero. The lamp does obey Ohm's law at any fixed temperature; the experiment simply keeps changing that temperature.
实验五 密度 三条测量路线
规则固体的体积用尺子量边长再算；不规则固体用排水法，尤里卡罐溢出的水体积就是物体体积；液体用天平称空量筒和装液后的质量差。公式是 ρ=m/V，单位换算最容易错，1 g/cm³ 等于 1000 kg/m³，答题前先看清题目要哪一种。
顺序上要先称质量再浸水，否则石头带着水出来质量偏大。量筒读数看凹液面最低处，视线与液面齐平。物体如果浮起来，用细针压下去，或者换密度更小的液体，别让一半体积露在液面以上。
Density has three routes. Regular solids are measured with a ruler and a balance, irregular solids go into a displacement can where the overflow equals the object volume, and liquids are weighed before and after filling a cylinder. Unit conversion trips people up, since one gram per cubic centimetre is one thousand kilograms per cubic metre. Weigh first and immerse second, otherwise water clinging to the sample inflates the mass. A floating sample needs a thin pin to hold it under, or the volume read will be too small.
实验六 力与伸长 胡克定律的上限
挂砝码拉弹簧，用尺子量伸长量，画力-伸长图，斜率就是弹簧常数。F=kx 只在弹性限度内成立，加到某一点后曲线开始弯，那一段不能再当直线拟合。这一步最容易丢分的是答案里没写"未超过弹性限度"。
读数时眼睛与刻度齐平，避免视差。用整串砝码依次加、再依次减，看两条路径是否重合，能判断弹簧有没有被拉过头。伸长量是现长减原长，不是总长，把总长直接画进图里会让截距不为零。
Loading a spring with masses and plotting force against extension gives a straight line whose gradient is the spring constant. The relationship stops holding at the limit of proportionality, where the graph bends and a straight fit becomes invalid. Measuring extension as current length minus original length keeps the line through the origin, while plotting total length shifts the intercept. Loading and then unloading reveals whether the spring has been overstretched, because the two paths stop overlapping.
实验七 加速度 F=ma 与控制变量
小车、滑轮加吊挂砝码，两个光电门分别测挡光时间算出速度，再用 a=(v-u)/t 求加速度。改变拉力时总质量必须保持不变，做法是把砝码从小车挪到吊盘上，而不是另外再挂新的。
F=ma 里的 F 是合力。轨道没调水平或者摩擦没补偿，图线截距就不在原点，这也是一道常考的误差分析题。两个光电门的间距越大，测速的相对误差越小，但加速度的取样区间也随之变长，选间距时要看轨道长度够不够。
A trolley, a pulley and hanging masses drive the acceleration practical. Two light gates time a card, and the change in velocity over the interval gives acceleration. When the pulling force is varied, total mass must stay constant, so masses move from trolley to holder rather than being added. An unlevelled track or uncompensated friction shows up as a non-zero intercept on the force-acceleration graph, which is exactly the error question examiners like to ask.
实验八 水波 波长与 v=fλ
用波纹水槽或振动弦测波长和频率，再用 v=fλ 求波速。水槽里用频闪灯把波纹冻住，量若干个波长的总距离再除以个数，比只量一个波长准得多。频率从信号发生器上直接读，不用另外测。
弦上的波用驱动频率和驻立波节点间距求波长，两个相邻节点之间是半个波长。波长越长，同一频率下波速越大，这个关系在比较不同水深的水槽时特别明显。
Ripple tanks and vibrating strings both test v equals f times wavelength. A stroboscope freezes the tank pattern so several wavelengths can be measured at once and divided by the count, which beats measuring a single crest. On a string the distance between adjacent nodes is half a wavelength. Deeper water supports a longer wavelength at the same driving frequency, a comparison examiners set up with two tanks side by side.
实验九 光的反射与折射 sin i 对 sin r
把光射进玻璃块，记录入射角和折射角，画 sin i 对 sin r 的图，斜率就是折射率，玻璃大约 1.5。角度一律从法线量起，量成与表面夹角就会全线偏。反射部分简单，入射角等于反射角，换成粗糙表面则得到漫反射。
插针法要求针竖直插，前后两根针从眼睛看过去重合，眼睛位置偏一点角度就读偏。玻璃块两个表面要平行，出射光线与入射光线平行但有侧移，侧移量随入射角增大而变大，这也是判断玻璃块是否合格的一个办法。
Directing a ray into a glass block and recording both angles gives sin i against sin r, whose gradient is the refractive index, near 1.5 for glass. Angles are measured from the normal and never from the surface. In the pin method the pins must be vertical and viewed so that successive pins appear aligned, or the angle drifts. A parallel-sided block sends the emerging ray sideways without changing its direction, and the shift grows with the angle of incidence.
实验十 红外辐射 莱斯利立方体
莱斯利立方体四个面做成不同颜色和材质，装满热水后用红外探测器逐个测。探测器到每一个面的距离必须一样，只改表面性质。黑色粗糙面发射和吸收红外都最强，银色光亮面最弱，白色居中。
距离不一致是这项实验最大的系统误差，读数随距离下降得很快，稍挪一点就能盖过不同表面之间的真实差别。立方体里要换同温的水，四个面温度一致，否则测到的是水温差不是表面差异。
A Leslie cube with differently finished faces is filled with hot water and read face by face with an infrared detector. Matt black emits and absorbs most strongly, shiny silver least, white in between. Detector distance dominates the error budget, because the reading falls quickly with distance and a small shift can outweigh the real difference between surfaces. The cube needs water at one temperature throughout, or the comparison measures cooling rather than surface finish.
实验题怎么被考 4 到 6 分的三段式
实验题通常 4 到 6 分，评分点大致分三块：读数或计算、控制变量、误差来源与改进。误差要写到具体机制，"热量散失到环境""视差""接触电阻"这类说法才给分，写"人为误差"等于没写。
数学技能在物理卷里分量不轻，AQA 8463 要求物理部分的数学技能约占 30%，单位换算、有效数字、图表斜率都在评分点上。画图时坐标轴要标物理量和单位，斜率要带单位，这些细节每年都有人丢。
Practical questions usually carry four to six marks, split between readings or calculations, controlled variables, and a named source of error with an improvement. Vague answers about human error score nothing, while heat loss to the surroundings or parallax earns the mark. Mathematics is weighted heavily in AQA 8463 physics, so unit conversions, significant figures and the units attached to a gradient all count. Labelling axes with both quantity and unit is free precision and costs nothing.
Cambridge International AS ALevel Physics Coursebook 正版剑桥国际AS ALevel考试物理教材带线上账号 英文原版¥557.0📱 去京东购买京东精选
需要一对一辅导？Need One-on-One Tutoring?
A-Level / GCSE / IB / KS2-KS3 全科辅导
16621398022 同微信
Visit aleveler.com
或截图后用「扫一扫 → 相册」选择本图