EuPhO-2022 Experimental Problem - Solution - Preliminary grading scheme

E1: Colour and temperature the light source and the measurement distance are not
given.
Theory
The infrared thermometer cannot be used to measure Experiment
the filament temperature for several reasons – the range
For the measurement of the power dependence of the
of the IR thermometer (stated on the instrument) only
temperature, we will read out the voltage and current
goes up to 500 °C. The filament is also too small to be the
from the power supply. To sample the expected curve
only thing measured. IR opacity of the glass bulb is also
of the T (P ) relationship, we must sample it sufficiently
not guaranteed. Therefore, the only way to measure the
well, especially at lower powers where the temperature
temperature is indirectly through the colour index, for
changes more quickly. We suggest sampling at least
which the relation to temperature is provided.
8 powers/temperatures to cover the relationship more
Wien’s displacement law suggests that at lower tem-
precisely and distinguish outliers from reliable mea-
peratures, the light will contain more red component
surements. For each power setting, we must measure
than green and blue, while at higher temperatures, the
the illuminance through the chosen filters by covering
green and then the blue will increase faster than red,
the sensor of the light meter with a filter. Covering the
leading to increasing ratios G/R, B/R and B/G. We
light meter filters all the light, including the light re-
must, however, consider, which pair of filters will be the
flected from the walls and the floor, leading to a better
most suitable choice.
measurement. Placing the filter next to the light source
The values measured through different filters depend also introduces the risk of burning the filter. Planning
on the spectral response of each filter, including its over- ahead, we can simultaneously measure the illuminances
all opacity. It also depends on the sensitivity of the light without a filter, needed in Task 2.
meter to each wavelength. Instead of theoretical predic-
Each colour index is then converted to a tempera-
tions, we are given reference measurements at known
ture by reading out from the calibration graph. We can
temperatures. If we plot the ratios for all three pairs, we
also estimate the relationship by employing the Stefan-
observe that B/G is the least suitable, as it changes much
Boltzman law if we neglect other losses and the contri-
less with temperature, compared to the other two. B/R
bution of the ambient temperature:
and G/R are comparable, but the blue filter has lower
transmittance, which will lead to lower accuracy (higher √
T ∝ P. (1)
4

relative error).
Any pair of filters is a valid choice to proceed with the According to measurements with multiple light bulbs in
measurements, but will affect the end accuracy. Averag- different environments, the fit is
ing the results is also an option, but including B/G com- √
bination may still reduce the accuracy of the end result. T = (1220 KW−1/4 ± 20 KW−1/4 ) P , (2)
4

To use the plot for converting the colour index to tem-

perature, we need a trend line. A linear trend is enough which is used as a baseline for determining the RMS of
to cover most of the range, except at lower tempera- students’ measurements.
tures, where the relationship tapers off. We can extend The background illuminance must be measured
the range by combining two trends across the range, or through all filters – it is most likely zero, but a good
to draw a smooth curve by hand. Zig-zag interpolation experimentalist must check, and if significant, it must
is less suitable due to scatter in the reference measure- be subtracted from measurements. This is also a way
ments. for us to detect if they left their desk lamp on – if the
background differs significantly from the rest of the
Figure 1: Three choices of the colour index based on contestants.
each pair of filters. Linear fits for the top part of the The distance between the light source and the light me-
range are shown, and a smooth curve. ter should be short enough to enable accurate measure-
ments at lower powers. Distance can also be different for
Colour Index
1.4
different power ranges, but care must be taken, as the ef-
G/R fect of the finite size of the filament may play a role, as
B/R
1.2
B/G well as the changing reflections from surroundings and
1 from the top of the light stand.
0.8

Marking scheme
CI

0.6

0.4
General guidelines for marking in all tasks
0.2
• Granularity for marks is 0.1 p.
0 • measurements/results given with inappropriate num-
1600 1700 1800 1900 2000 2100 2200 2300 2400 2500 2600
ber of significant figures may get deducted 0.1 p. (rule
T [K]
of no propagating error applies)
• A simple numerical error resulting from a typo is pun-
Using the absolute values from the table instead of
ished by 0.1 p unless the grading scheme explicitly says
ratios is not correct, as the intrinsic luminous flux of
otherwise.
EuPhO-2022 Experimental Problem - Solution - Preliminary grading scheme

the brightness of the light source, so absolute values can-

Figure 2: Dependence of the temperature on the not be used to determine the temperature. Points for A2
power with a superimposed prediction from the Stefan- may nevertheless be given.
Boltzman law. Shown is a comparison measured at two
distances and using two combinations of filters. We see Measurements Points
the trends are comparable and remain within ±100 K, B1 Data collection † 1.2
and RMS is around ±40 K based on multiple measure- Measured U , I, E through 2 or 3 fil- 0.1n
ments. ters at n ≤ 8 points (for 1 filter only
3/4n and rounded, for measurement
Filament temperature
2600
of U and I or RG(B) only, which typ-
G/R, r=6cm ically happens with measurements
B/R, r=6cm
2400 G/R, r=10cm with IR thermometer, max. 0.3)
Compute P
B/R, r=10cm
T4 ~ P 0.1
Compute color indices or convert di-
2200
0.1
rectly to temperature (from graph or
T [K]

2000
trend in A2)
1800 No points below 5 W −0.1
1600
No points above 16 W −0.1
Measured RGB background 0.1
1400 Determined background constancy 0.1
(e.g. measured at the beginning and
2 4 6 8 10 12 14 16 18 20 22
P [W]
the end)
B2 Temperature plot ‡ 0.9
• Errors in theoretical derivations which cause dimen- Determine temperatures 0.3
sionally wrong results are punished by at least 50% of Plot of data 0.3
the marks for the derivation unless the grading scheme Add best fit curve 0.1
explicitly says otherwise. Proper axis labels & ticks 0.1
• Propagating errors are not punished repeatedly un- Proper size of graph 0.1
less they either lead to considerable simplifications or Each point determined/drawn incor- −0.1
wrong results whose validity can easily be checked rectly
later. Used G/B index −0.2
• Negative points cannot decrease the score under the B3 Result quality ‡ 0.4
same section (A1,A2,...) below zero. Nonlinearity of the relationship is 0.1
Calibration Points visible
A1 Plotting 1.0 Nonmonotonous relationship −0.1
Compute color indices for sufficient 0.3 RMS from best fit within 40 K 0.3
number of data points over range RMS from best fit within 80 K or 0.2
Plot of color indices 0.5 RMS from best fit within 120 K or 0.1
Proper axis labels & ticks 0.1 Total on Measurements 2.5
Sufficient size of graph for precise 0.1 † Presentation of data: If U and I are directly mul-
readout (≥ 12 page) tiplied and only P values are presented, no marks are
Each point computed/drawn incor- −0.1 deducted.
rectly Background: Points for background are only granted
A2 Trend line 0.5 if the background is subtracted from the measurements,
Smooth trend curve or a composite 0.5 or can be reasonably neglected.
of linear trends
‡ IR measurement: If temperatures are “determined”
Single linear trend line (generates or 0.3
from IR measurements or any other method unfit to de-
outliers at some temperatures)
termine the temperature, no marks are given for B2. The
Zig zag connected points (=used in- or 0.1
same holds for B3, because presence of nonlinearity is
terpolation for readout), or point-
not an indicator of quality for meaningless data.
wise connected curve
Total on Calibration 1.5 To determine the RMS at B3, we compare it to Eq. 2 and
take the root of the mean squared deviation. Reasonably
This section is only for the calibration data (tables, exclude any outlying measurements at very low powers
graphs) based on the given table. Plots made based on where we expect large deviations. The RMS calculations
IR measurements, get zero points. can be handled by the auxiliary Excel file.
Full points are given for a single colour index graph,
or for multiple colour indices which may be on the same
plot or on the separate plots.
Absolute values plotted: At most 0.2 points for A1 (if
axes and ticks are done correctly, see the table above),
if they plotted the absolute values through one filter or
differences of values of several filters, instead of colour
indices. The illuminance depends on the distance and
EuPhO-2022 Experimental Problem - Solution - Preliminary grading scheme

E2: Efficacy which turns out to match the experiment well.

For the incandescent bulb, a similar assumption can
Theory be made based on a thin filament model. The different
orientation of the symmetry axis leads to a different re-
Light sources do not radiate in all directions equally. The sult: R π/2
angular distribution of luminous flux Φ (luminous in- cos2 θ dθ π
C̃W = R0π/2 = ≈ 0.79. (10)
tensity) must be integrated over the solid angle. A light cos θ dθ 4
0
meter at distance r to the light source, oriented so that
the light falls on it perpendicularly, measures the illu- These approximations can be used to a good effect but
minance E of a certain part of the imagined integration are not required in the experimental task.
sphere surrounding the light source:
I
Φ = E(Ω)r2 dΩ. (3)
Experiment
To measure the angular dependence, a suitable distance
The LED only shines the light into a hemisphere, and to the light source must be chosen. Too far, and the sig-
has cylindrical symmetry around the direction straight nal becomes weak and any background could become
ahead, so we can simplify the expression, noticeable. It is advisable to measure the angular depen-
Z π/2 dence at the highest power in order to improve the sig-
ΦLED = 2π E(θ)r sin θ dθ,
2
(4) nal to background ratio. Measurement can also be per-
0 formed through one of the filters.
For the incandescent bulb, the finite size of the fila-
and for the incandescent bulb, the symmetry axis is per-
ment becomes an issue if we measure too close to the
pendicular to the direction straight ahead, and shines
bulb. This becomes noticeable at distances lower than
into full solid angle:
10 cm. This was not an issue for the colour index mea-
Z π/2 surement, but it matters for the absolute flux estimation.
ΦW = 4π E(θ)r2 cos θ dθ. (5)
0
Figure 3: Angular profile of the incandescent light bulb,
The integrals will have to be evaluated numerically – measured at r = 15 cm, P = 20.6 W in increments 10°.
it can be done by using the trapezoidal or the Simpson Vertical lines are the division angles for formula (6). We
method, or by using the formula for a spherical segment obtain C = 10.01 (C̃ = 0.80).
area given in the hint:
W profile
X
Φ = 2πr2 E(θi+1/2 )(cos θi − cos θi+1 ) (6) 1
E(θ)/E(0)

i
0.8
and equivalent (but with sin ⇐⇒ cos) for the incandes-
cent bulb. Here, choosing evaluation points in the mid-
E(θ)/E(0)

0.6
dles of intervals is better than choosing one of the edge
points. However, the exception are the “edge” measure- 0.4

ments, where the measurement point is actually in the

middle of the interval – the point straight ahead for the 0.2

LED is in the middle of the spherical cap. The same goes 0

for the “poles” of the incandescent light bulb. 0 10 20 30 40 50 60 70 80 90
θ[°]
The ratio between the head-on measured illuminance
and the luminous flux, can be expressed as
To describe the inflection point in the light distribution
Φ = Cr2 E(0), (7) well, we will need at least 5 measurements in the θ ∈
[0, π/2] interval. We can either rotate the light source on
or, more intuitively, as a correction factor to the isotropic the spot, or position the light meter at different angles in
source: relation to the stationary light source.
Φ = {4π, 2π}C̃r2 E(0). (8) For the light distribution left-right symmetry can be as-
sumed, or, alternatively, the entire θ ∈ [−90°, 90°] range
can be measured, allowing to take into account asymme-
Analytical estimates tries and an angular offset in the light distribution. The
straight ahead measurement is centered in a symmetric
One possible pathway is to estimate these factors with- band, which needs care so it is not double-counted in
out measurements, using reasonable assumptions about case only half of the range is integrated and then dou-
the light distribution. The LED can be assumed a planar bled.
emitter, with a cosine distribution of luminous flux:
With the conversion factors known, the luminous ef-
R π/2 ficacy can be determined by measuring the frontal illu-
cos θ sin θ dθ 1
C̃LED = 0
R π/2 = (9) minance at powers that cover the entire range from the
0
sin θ dθ 2
lowest detectable illuminance to the maximum allowed
EuPhO-2022 Experimental Problem - Solution - Preliminary grading scheme

Figure 4: Angular profile of the LED, measured at r = Figure 5: Efficacy of both light sources depending on the
10 cm, P = 1.33 W in increments 10°. Vertical lines are input power.
the division angles for formula (6). We obtain C = 2.63 LED efficacy
(C̃ = 0.42). 400

LED profile

E(θ)/E(0) 350
1

η[lm/W]
0.8
300
E(θ)/E(0)

0.6
250
0.4

0.2 200
0 0.2 0.4 0.6 0.8 1 1.2
P[W]
0
0 10 20 30 40 50 60 70 80 90
W efficacy
θ[°]
25

C C̃ 20
W 10.01 0.80
LED 2.63 0.42 15
η[lm/W]

Table 1: Example values of the conversion factor be- 10

tween the frontally measured illuminance and the lu-
minous flux for both light sources. The values will 5
vary within some wider distribution because of varying
light sources and other errors, which is indicated by the 0
brackets in the grading sheet. 2 4 6 8 10 12 14 16 18 20 22
P[W]

power. For the incandescent bulb, this measurement

can be done simultaneously with Task 1 for better time
Theoretical background Points
efficiency.
A1 Eq. (3) or equivalent 0.5
It is not required to measure at the same distance as Dependence on r−2 0.2
the angular dependence. Multiple distances may also be Dependence on angle (noticing 0.3
used. anisotropy)
We have to avoid placing any additional objects near A2 Take into account cylindrical sym- 2×0.1
the light source to avoid introducing more reflected or metry (each)
blocked light – such as placing the light source directly Choose right symmetry axis for light 2×0.1
on the white paper, or having other obstructions such as sources (each)
the black paper screen or any filters too close to the light Correct factors of 2π and 4π (LED and 0.1
bulb. bulb, respectively)
To plot the efficacy, we divide the Φ obtained from eq. Assume area of the sensor comes −0.2
(3) for each of the light sources, with P = U I read out into calculations
from the power supply. Total on Theory 1.0
The result shows that the efficacy of the incandescent 2× means one for each light source (bulb, LED).
light starts out at zero at low powers and increasing with
power, as its temperature increases. The LED has the
highest efficacy at lowest powers, then it drops off at
higher powers, mostly due to increased temperature of
the light emitting junction.
At the lowest settable currents, the readout on the
power source is no longer reliable – for example, LED
may glow slightly even at 0 A. The pole at the origin can
be attributed to this source of error.

Marking scheme
The basic equations could be stated in a separate section
of the solution, or spread over different parts of the so-
lution.
EuPhO-2022 Experimental Problem - Solution - Preliminary grading scheme

Angular measurement Points Efficacy measurement Points

B1 Incandescent measurement 0.9 C1 Incandescent measurement 0.7
Measured at n ≤ 6 or more angles be- 0.1n Measured U , I, E for incandescent 0.1n
tween 0 and π/2 light at n ≤ 7 points
Specified auxiliary data (power, dis- 0.2 No measurements above 16 W −0.1
tance) C2 Plotting bulb efficacy 0.8
Measured full angular range 0.1 Convert and plot points n ≤ 6 (if con- 0.1n
[−90°, 90°] (e.g. by rotating light) verted into flux, not efficacy max 0.3)
Maximum illuminance below 100 lx −0.2 Missing axis labels −0.1
Measured closer than 10 cm −0.1 Deviation from a monotonous shape −0.1
B2 Integration procedure 0.7 † Values within RMS< 0.2 0.2
Used the hint 0.7 Values within RMS< 0.5 or 0.1
Values at the edges of intervals −0.1 C3 LED measurement 0.7
Double counting of the equator −0.1 Measured U , I, E for LED at n ≤ 7 0.1n
Used trapezoidal or similar rule or 0.7 points
Incorrect treatment of edge values −0.1 Fewer than 2 measurements above −0.2
Double counting of the equator −0.1 0.2 W
Analytically derived eq. (10) or sim- or 0.7 Fewer than 2 measurements below −0.1
ilar 0.2 W
Averaged values without weights or 0.3 C4 Plotting LED efficacy 0.8
B3 C result accuracy for Tungsten 0.3 Convert and plot points n ≤ 6 (if con- 0.1n
Calculated C or equivalent 0.1 verted into flux, not efficacy max 0.3)
Value of C ∈ [9.7, 10.3] 0.2 Missing axis labels −0.1
Value of C ∈ [9.4, 10.6] or 0.1 Deviation from a monotonous shape −0.1
B4 LED angular measurement 0.9 ‡ Values within RMS< 5 0.2
Measured at n ≤ 6 or more angles be- 0.1n Values within RMS< 10 or 0.1
tween 0 and π/2 Total on Efficacy 3.0
Specified auxiliary data (power, dis- 0.2 † Vary the C factor within the range [9.4, 10.6] to check
tance) for a better match. This helps remove propagation error
Measured full angular range 0.1 from a badly determined C and additionally, any small
[−90°, 90°] (e.g. by rotating light) variations in power output between light bulbs.
Maximum illuminance below 100 lx −0.2
‡ Vary the C factor within the range [2.6, 3.4] to check
B5 Integration same as B2 0.7
if there is a better match. The LEDs may vary in light
B6 C result accuracy for LED 0.3
distribution and absolute intensity, so the trend cannot
Calculated C or equivalent 0.1
be matched 1:1. Exclude also the lowest power outliers
Value of C ∈ [2.8, 3.2] 0.2
from the assessment, as the singularity can give a high
Value of C ∈ [2.6, 3.4] or 0.1
RMS without being significant to the LED itself. It de-
B7 Background measured 0.1 pends on the voltage bias of the power source and resis-
Background constancy check (multi- 0.1 tivity of junctions.
ple measurements)
RMS comparison: To make the comparison of the
Total on Angular 4.0
graphed solutions to the reference less subjective, we do
Plotting the angular dependence is not required for a root mean square comparison: average square devi-
the procedure, but it counts as 0.2 points if the integra- ation from an empirical model curve based on a larger
tion procedure was otherwise not performed. number of measurements in a controlled environment
If a matching analytical value for both conversions are (Fig. 5). For the light bulb efficacy, we use
obtained, but angular dependence is not measured, the
contestant can get all marks except for the measurement η(P ) = ln(1 + exp(1.189P − 4.632)) (11)
(1.0 out of 1.9 for each light source).
which is just a linear function smoothly flattened at the
If a comparison of analytical and experimental esti- bottom.
mate are done, or if an analytical model is fitted to the
For the LED, we use
experimental data, the procedure is correct, and up to
additional +0.3 points per light source may be given to η(P ) = 2.56/P − 23.78P + 259.56 (12)
compensate points lost for steps that merit points that
may not be necessary for the method used. where possible intensity variations can be compensated
by allowing C variation in C2/C4. Variations in the 1/P
part (due to different offsets in power supply readout)
can be compensated by excluding the low-power mea-
surements from the RMS calculations in a reasonable
way.
All these are in base units without prefixes (omitted
for clarity).
EuPhO-2022 Experimental Problem - Solution - Preliminary grading scheme

If angular dependence is ignored use this table as a

shortcut for grading
A1 dependence on r−2 only 0.2
A2 Correct factor of 2π or 4π only 0.1
B1- not applicable 0.0
B6
B7 Background measured 0.1
Background constancy check (multi- 0.1
ple measurements)
C1 No change 0.7
C2 Rescale with correct C to check RMS 0.8
C3 No change 0.7
C4 Rescale with correct C to check RMS 0.8
EuPhO-2022 Experimental Problem - Solution - Preliminary grading scheme

E3: Radiative heating Albedo

For the white plate, only a part of the incident flux is ab-
Theory sorbed, so we replace j by j(1 − a) if a is the albedo:
The plate receives a radiant flux density j, determined P
by the power P of the light source, and the distance r j = (1 − a) . (22)
Cr2
between the target and the light source. The light source
does not shine equal amounts of light in all directions, As a consequence, any slope measured for both plates
therefore we must use the correction factor C, derived in will be in the ratio (1 − a) to each other. This can be ex-
Task 2, to convert from the total radiant flux to forward pressed as a fraction of trend slopes, ratio of tempera-
radiant flux density. ture differences, or similar.
P
P = Cr2 j → j = . (13)
Cr2 Experiment
Not necessary, but also correct, is to (numerically)
integrate/average across the entire plate, j(πr2 ) = The radiant flux density can be varied in two main ways,
R
P Cr cos θ dA to take into account spatial variation of
−2 or a combination of both: by changing the distance, or by
C, r and θ (angle of incidence). changing the current through the light bulb. Both meth-
The incident flux density is dissipated to the environ- ods are acceptable, but varying the current also changes
ment directly, as well as by heat conduction through the the spectrum and the efficiency of the light bulb, so it
plate. Mark by TF the front temperature and TB the back may produce biased and nonlinear results. The students
temperature. Conservation of energy gives us the system should know that varying a single parameter is the cor-
of equations rect procedure.
The required measurements in this task are the front
λ
j = h(TF − T0 ) + (TF − TB ) (14) and back temperature at different powers, for black and
d white plate. It is essential to wait for equilibration,
λ
0 = h(TB − T0 ) + (TB − TF ). (15) which includes waiting the back temperature to stabi-
d lize. It is advisable to measure starting with the lowest
This system of equations leads to the following relations: flux density, because it will require the least equilibra-
tion time from the initial room temperature of the plate.
j = h(TF + TB − 2T0 ) (16)
The target should not be too close to the light source,
λ
j = (h + 2 )(TF − TB ). (17) not only because of the risk of burning, but also because
d close to the light bulb, the light is very nonuniformly dis-
Any linear combination of equations (14,15) also al- tributed across the plate. Increased convection rate due
lows determination of both h and λ. A particular linear to high temperature also starts deviating from the linear
combination that may be used is the isolation of individ- regime. Placing the target too far from the light source
ual temperatures: leads to a negligible heating and thus a very large rela-
! tive error in temperature differences, especially for the
1 1 1 white plate.
TF − T0 = + j (18)
2 h h + 2 λd In this task, the measurements are subject to many
! sources of errors: measuring from different distances
1 1 1
TB − T0 = − j. (19) and at different angles may include different propor-
2 h h + 2 λd tions of background or reflected IR radiation from the
light source (if the targeted area is still illuminated), if
In our system, 2 λd > h, but still in the same order of the measurement takes too long, the plate may start cool-
magnitude. Treating the slope of TF as 1/(2h) or the slope ing down (this is noticeable in a few seconds), air cur-
of TF −TB as (2λ/d)−1 is a reasonable approximation, but rents may increase convective heat dissipation, and the
still not theoretically correct. ambient temperature may also change during the mea-
surement (especially if the light source is placed too close
Error analysis to the wall, or if the power source’s fan exhaust is too
Errors should be propagated from the slope. For ex- close to the measurement setup). The errors are most
ample, if they obtain slopes k1 = 1/h and k2 = 1/(h + noticeable at low radiant flux and for the white plate,
2λ/d), they should propagate the errors. We should al- where increases in temperature are the smallest.
low both straight addition of error contributions of dif- For these reasons, it is advisable to take more than
ferent terms, or adding squared errors (independent er- one measurement per data point and average the re-
rors), e.g. sults, and to cover a sufficiently wide range to reduce the
slope error. At least 3 points are needed to draw a trend,
1 σ1
h= ± 2 (20) but 5 is better. With more points, it is easier to spot out-
k1 k1 liers and utilise the measurements which are least sub-
 
1
λ = d2 ( − ) ±
1 d σ1 σ2
+ 2 (21) jected to errors. Back and front temperatures are best
k2 k1 2 k12 k2 measured in pairs one after the other to reduce the er-
ror in the temperature difference signal due to changing
and analogously for other slope definitions.
conditions.
EuPhO-2022 Experimental Problem - Solution - Preliminary grading scheme

Note: the ambient temperature T0 is an effective tem-

perature that combines air temperature and radiative Figure 6: Black plate measurements for eqs. (16,17). The
exchange with the surrounding walls, ceiling and other flux density j was varied by changing the distance r. The
objects. We do not need its value, we only need the slope of the first graph equals h−1 , so h = 10 Wm−2 K−1 .
slopes of the linear trends. Inexact T0 can lead to inac- Lowest and highest measurement were excluded from
curacies if used together with an assumption the linear the fit. The intercept is 2T0 .
relations go through the origin. T0 cannot be reliably de- The slope of the second graph equals (h + 2λ/d)−1 , so λ =
termined by measuring surrounding temperatures, but 0.072 Wm−1 K−1 . The intercept is reasonably close to 0.
it can be estimated by measuring the equilibrium tem- black plate

perature of the plate in the absence of the light source. 80

slope=0.0988 °C/(W/m-2)
75

The measurements of the front and back temperature 70

at different radiant fluxes, must be processed and plot- 65

TF+TB [°C]
ted to extract the necessary slopes. For the black plate, 60

two plots will be needed, based on equations (14,15), 55

equations (16,17), or any linearly independent pair. Lin- 50
ear regression gives us the slopes h and h + 2 λd (or their 45
reciprocals). T0 is best determined by the j = 0 intercept
40
of the trend line for (Eq. 16) or any equivalent plot, and 0 50 100 150 200 250 300 350 400
should match T0 determined by other methods. If mea- P/Cr2 [Wm-2]

sured correctly, the intercept of the trend line for (Eq. 17)
should be zero within the error margin. black plate
8
slope=0.0173 °C/(W/m-2)
7
It is possible to calculate the necessary slopes from
a measurement at a single input power (for each plate
6

color), if T0 is measured well. This can be done with- 5

TF-TB [°C]

out a graph. However, using multiple measurements de- 4

creases the impact of statistical errors and enables us to 3

better estimate the error, so a single measurement will 2

carry a significant error. 1

0
0 50 100 150 200 250 300 350 400
Albedo, as defined in the task text through irradiance P/Cr2 [Wm-2]
units, cannot be measured using a light meter, which
measures in photometric units. Additionally, light re-
flected from a white plate introduces additional geomet-
ric considerations and angular distribution of reflected Theoretical background Points
light, that cannot easily be taken into account. A1 Power to irradiance 0.4
Correct eq. (13) or equivalent 0.3
Realizing the same geometry from 0.1
The albedo can be estimated as a fraction of the cor- Task 2 applies (C or equivalent)
responding line slopes between the black and the white A2 Heat exchange balance 0.6
plate, taking any of the relations (14,15,16,17). This Correct each of eqs. (14,15) or equiv- 2×0.3
means that for the white plate, measuring only one side alent.
of the plate is enough to determine the albedo, assum- Each partially correct eq., e.g. as- or 0.1
ing h and λ remain the same. The difference slope or the sumed TB = T0 in (14) or missing 2
back temperature slope are the least suitable, as they in- or h in (17)
troduce a large relative error to the measurement due to A3 Albedo balance 0.2
a minimal increase in temperature. Correct eq. (22) or equivalent 0.2
Total on Theory 1.2
The assumption that the left hand side of equation
P /A = h(T − T0 ) distributes the full power of the light
source to the area of the plate, indicates a lack of under-
standing and merits 0 points for theory part.
If conduction is not considered at all a maximum of 0.1
points is given to A2.
Marking scheme

The basic equations could be stated in a separate section

of the solution, or spread over different parts of the so-
lution.
EuPhO-2022 Experimental Problem - Solution - Preliminary grading scheme

Dissipation parameters Points

Figure 7: White plate measurements for eqs. (16,17). C1 First of the two plots 0.8
The slope ratio with the black plate result is (1 − a) = n ≤ 5 correctly converted and drawn 0.1n
0.0254/0.0988 = 0.26 for the first graph. The second points
graph confirms this with a closely matching (1 − a) = Correct trend line 0.1
0.00465/0.0173 = 0.27. Correct slope readout 0.1
white plate Slope error estimate 0.1
70
slope=0.0254 °C/(W/m-2)
Intercept disagrees with expecta- −0.1
tions
65
Missing axis labels, ticks or unsuit- −0.1
able size
TF+TB [°C]

60
C2 Second of the two plots 0.8
55
Same breakdown as C1
C3 Calculation of h 0.8
50 Correct algebraic relation to slopes 0.2
Numerical value within [10, 14] 0.3
45 Numerical value within [8, 16] or 0.2
0 100 200 300 400 500 600 700 800 900
P/Cr2 [Wm-2]
Numerical value within [6, 18] or 0.1
Correct error analysis 0.2
white plate
Error estimate < 1 (if error analysis 0.1
4.5 is reasonable)
slope=0.00465 °C/(W/m-2)
4 C4 Calculation of λ 0.8
3.5 Correct algebraic relation to slopes 0.2
3 Numerical value within [0.06, 0.08] 0.3
2.5
Numerical value within [0.05, 0.09] or 0.2
TF-TB [°C]

2
Numerical value within [0.04, 0.10] or 0.1
1.5
1
Correct error analysis 0.2
0.5
Error estimate < 0.01 (if error analy- 0.1
0 sis is reasonable)
-0.5 Total on Dissipation 3.2
0 100 200 300 400 500 600 700 800 900
P/Cr2 [Wm-2]
The values are in SI base units.

Measurements Points
B1 n ≤ 5 measurements of TF (black) 0.1n
B2 n ≤ 5 measurements of TB (black) 0.1n
B3 n ≤ 5 measurements of TF , TB or 0.1n
both (white) Without plotting: If the entire fitting process is done
B4 Measured the unchanging values 0.2 numerically without plotting, use equivalent concepts to
(distance if U, I varied, U, I if dis- the grading above – tables instead of plots, slope calcu-
tance varied) lations instead of trend lines, etc. As the plot is not re-
Estimated measurement errors (at 0.2 quired, a correct procedure can yield full points.
least separate for each plate color)
– the instrument precision is not a The error analysis in this case may consist of doing
valid error estimate the entire procedure (e.g. using a single measurement
Estimated measurement errors or 0.1 with T0 knowledge, or two points without background),
(common for all) with multiple measurement runs and doing statistics.
Measured by varying the current −0.2 Another option is propagating relative errors from the
(not distance) single measurement errors. The main criterion is, that
Total on Dissipation 1.9 the error source is statistical, not instrumental.
Determination of h and λ will require extraction of two
trend lines from two plots. Plotting on the same graph The point count includes the origin for the plot of the
counts as two, but the vertical axes must be labelled cor- temperature difference (eq. 17).
rectly. The trend lines will have a j = 0 intercept that
will be 0 in case of temperature difference, and related The slope error is the main source of error – distances
to ambient temperature otherwise. Using r−2 or P in- and powers can be considered accurate. Error estimate
stead of j as an axis is valid as long as the conversion is on the slope can be done based on point scatter (but not
done correctly at the slope readout. with fewer than 5 points), with or without taking into
account errorbars (if the students estimated them).

Error propagation: if a wrong value of C is used, re-

calculate with a suitable value and grade accordingly.
EuPhO-2022 Experimental Problem - Solution - Preliminary grading scheme

Albedo Points
D1 One or more plots 1.0
n ≤ 5 correctly converted and drawn 0.1n
points
Estimated individual measurement 0.1
error
Correct trend line(s) 0.2
Correct slope readout 0.1
Slope error estimate 0.1
Intercept disagrees with expecta- −0.1
tions
Missing axis labels −0.1
D2 Data processing 0.7
Correct algebraic expression for a 0.2
Numerical value a ∈ [0.65, 0.75] 0.2
Numerical value a ∈ [0.6, 0.8] or 0.1
Correct error analysis 0.2
Error estimate < 0.05 (if error analy- 0.1
sis is reasonable)
Total on Albedo 1.7
The possibility of measuring both temperatures allows
combinations where both sets of data can be used for
albedo estimation – by averaging two slope ratios, or
similar. This is also a valid approach.
In cases where only pointwise numerical evaluation
using several data points is employed a maximum of 0.5
points for D1 (0.3 conversion of data, 0.2 for error esti-
mates) and a maximum of 0.7 for D2 will be awarded.
For evaluation with one data point only a maximum of
0.2 points for D1 (0.1 for conversion of data, 0.1 for er-
ror estimate) and 0.5 for D2 (0.2 algebraic expression, 0.2
value, 0.1 error analysis) will be awarded.