The Energy Performance of In-roof PV

R
viridian
solar
systems, it is becoming ever more important that the solar
industry can give credible advice on their performance.
The Energy
Performance
of In-roof PV
Image: Elliotts Premier Roofing
January 2014
The performance penalty for roof integrated
solar PV panels is almost certainly less than
you thought it was.
Photovoltaic panels installed integrated with the roof
covering will have less ventilation of the rear of the panel,
run at a higher temperature and so deliver less electricity
than the same panel installed above the roof with an air
gap behind. But by how much?
Energy yield from a solar pv module drops as the
operating temperature rises, for example EN61215 tests
found that the Clearline PV15 modules used in the study
have a power-temperature coefficient of -0.509%/˚C.
Figure 1 shows how the power output varies with
operating temperature.
Power - Temperature Curve
Clearline PV15 Photovoltaic Module
300
Power Output (W)
250
200
A review of the literature reveals little published work in
this area. For this reason engineers at Viridian Solar and
Enphase collaborated with researchers at Cambridge
University to conduct an experiment to assess the
temperature response of solar PV modules across a
range of installation situations. An annual energy yield
for each installation type was then predicted using
climate data and the power-temperature coefficient.
Method
A test rig was constructed with five Clearline PV15
(240Wp) monocrystalline solar photovoltaic panels
installed in a range of typical roofing situations. See
figure 2b.
Free-standing
The panel was installed free-standing on a
framework of aluminium angle. Consequently the
panel rear was completely ventilated.
Above tiled roof
The panel was installed above a roof of Marley
Modern interlocking concrete tiles in grey. The
panel was fixed to a pair of P1000T Unistrut beams
of 21mm x 41mm section. The beam was supported
on four roof anchors. The gap between the panel
(rear of laminate) and the tiles below ranged from
170mm to 190mm. See figure 3b overleaf.
In tiled roof, cold roof construction
The solar panel was integrated with a roof covering
of concrete interlocking tiles using a VAT flashing
kit. The roofing membrane below the tile battens
was draped by 20mm between the rafters. See
figure 4b.
In tiled roof, warm roof construction
The roof detailing was as for the cold roof , but
with 100 mm of rigid polyurethane insulation board
tightly fitted between rafters leaving a ventilation
gap of 25mm below the top face of the rafter. See
figure 5b.
150
100
50
0
0
10
20
30
40
Panel Temperature (°C)
50
60
70
Irradiation: 1,000 W/m 2
Figure 1 Power curve for modules used in the study
However, it is not clear how much of the cooling comes
from rear ventilation compared to heat losses from the
top face. In any case, depending on the details of the
roof build-up, integrated panels may be cooled by air
movement in the batten space behind the tiles.
More and more customers for solar PV are seeking
products that enhance the aesthetics and resale value
of their property. With greater interest in roof integrated
subscribe
Get it first - new briefings direct
to your inbox (max one/month)
In shingle roof, without ventilation
A layer of 18 mm thick plywood sheeting was
applied over the rafters and the solar panel fixed
directly to this, completely enclosing the rear of the
panel. Black ashphalt shingles were applied to the
roof and the panel weathered with a VAS flashing
kit. See figure 6b.
All panels were installed at a tilt angle of 35 degrees
and facing due south. The walls of the structure were
boarded to enclose the area behind the roof. The batten
...continues
R
viridian
solar
Free Standing
Figure 2 Test rig
Weather
Station
18.0
Free
Standing
Delta T from Ambient (degrees C)
16.0
14.0
y = 0.0143x
R² = 0.869
12.0
10.0
8.0
6.0
4.0
2.0
Above
Roof
Cold
Roof
Shingle
Roof
Warm
Roof
0.0
200
400
600
800
1000
1200
-2.0
Irradiance (W/m2)
Above Pitched Roof
32.0
30.0
Tile
PV module
Tile batten
Rail
Roof anchor
Delta T from Ambient (degrees C)
180mm
28.0
26.0
24.0
y = 0.0245x
R² = 0.9229
22.0
20.0
18.0
16.0
14.0
12.0
10.0
8.0
6.0
4.0
2.0
Rafter
Figure 3
Roof build up - above roof
0.0
-2.0
0
200
400
-4.0
600
800
1000
1200
Irradiance (W/m2)
Integrated in Pitched Cold Roof
40.0
38.0
Tile
Delta T from Ambient (degrees C)
36.0
Flashing
PV module
Tile batten
Membrane
Ventilated
batten space
34.0
32.0
30.0
28.0
26.0
24.0
22.0
20.0
y = 0.0333x
R² = 0.9529
18.0
16.0
14.0
12.0
10.0
8.0
6.0
4.0
Rafter
2.0
25
m
m
0.0
Figure 4
Roof build up - cold roof
-2.0
0
200
400
600
800
1000
1200
-4.0
Irradiance (W/m2)
Integrated in Pitched Warm Roof
38.0
36.0
34.0
25mm
Flashing
PV module
Tile batten
Ventilated
batten space
Membrane
Delta T from Ambient (degrees C)
Tile
32.0
30.0
28.0
26.0
24.0
y = 0.0326x
R² = 0.9282
22.0
20.0
18.0
16.0
14.0
12.0
10.0
8.0
6.0
4.0
2.0
25
m
m
Insulation
0.0
-2.0 0
Figure 5
Roof build up - warm roof
200
400
-4.0
600
800
1000
1200
Irradiance (W/m2)
Integrated in Pitched Shingle Roof
44.0
42.0
40.0
Delta T from Ambient (degrees C)
38.0
Ashphalt shingle
Flashing
PV module
Plywood
sheet
36.0
34.0
32.0
30.0
y = 0.036x
R² = 0.9388
28.0
26.0
24.0
22.0
20.0
18.0
16.0
14.0
12.0
10.0
8.0
6.0
4.0
2.0
0.0
Rafter
-2.0
0
200
400
600
800
-4.0
Figure 6
Roof build up - shingle roof
Irradiance (W/m2)
Figure 7 Derived temperature Curves
subscribe
Get it first - new briefings direct
to your inbox (max one/month)
1000
1200
R
viridian
solar
space behind the tiles was sealed with expanding foam
at the free edges of the tiles at verge and ridge, but left
open at eaves.
the results show a progression from lower to higher
gradient as the installation method reduces the level of
back ventilation.
The modules were wired to an Enphase M215 microinverter ensuring the electrical independence of each
module.
It is interesting to observe that the temperature
response of the ‘cold roof’ and ‘warm roof’ installations
were almost identical. It seems that the presence of
the ventilated batten space behind the roof integrated
panel offers a similar level of cooling irrespective of the
presence of the insulation layer found in a warm roof
construction.
Thermocouple temperature sensors were fixed to the
solar panels at top centre (front and rear) and to the
rear centre of the module. A pyranometer recorded
the level of irradiation and a weather station recorded
the ambient temperature, rainfall and wind speed and
direction.
Over a period of 106 days from September 2013 to
December 2013 data was collected every minute to
a data logger. 152,640 data points were collected for
each module.
Analysis
The temperature of each solar module minus the
ambient temperature was plotted against the irradiation
to produce a characteristic temperature-irradiation
curve for each installation method.
Data was sorted to include only periods where:
irradiation was stable (+/- 10% for the previous 10
minutes), to reduce heat capacity effects.
the wind speed was less than 4 m/s
modules were not partially shaded (power output
was within a band of +/-20% of the average)
The resulting scatter plots for each installation type
are presented in Figure 7 and show good fit to a linear
curve with r-squared values between 0.93 and 0.95.
Impact on Annual Energy Yield
From Figure 8, it can be seen that at 1,000W/m2
(corresponding to bright, direct sunshine on the
modules) that the free standing module would be
expected to be around 15˚C above ambient, the aboveroof around 25˚C and the in-roof modules around 33˚.
If under these light conditions the ambient temperature
was 25˚C then based on the declared powertemperature coefficient the free standing module would
have an output of 221Wp, the above roof module would
have output 209Wp and the in-roof panel 200Wp.
However, considering only performance at high
irradiation exaggerates the performance differences
because the sky is not always clear and the sun is not
always directly above the modules. Figure 9 shows the
distribution of irradiation levels incident on the plane
of a solar panel on a south facing roof from a year of
climate data collected as part of an earlier study into the
energy performance of roof-integrated Clearline solar
thermal panels. It is clear that the a solar module only
experiences irradiation at such a high level for only a
small part of the year.
Hours at Irradiation Level
South Facing Surface at 35 degree Pitch Angle, Cambridge, UK
Temperature Characteristic
800
700
Shingle
Roof
40
35
Warm
Roof
Cold
Roof
30
25
Above
Roof
20
15
Freestanding
10
Number of Hours in Year
Temperature (°C above ambient)
45
600
500
400
300
200
100
0
5
0
0
200
400
600
800
1000
1200
April 2006 - March 2007
Irradiation on Plane of Solar Panel (W/m2)
2
Irradiation (W/m )
Figure 8 Temperature characteristic of different installations
Figure 8 compares best fit lines for each installation
type.
Lines with steeper gradient reach higher
temperatures as light levels increase. As expected,
Figure 9 Hours of sunlight by irradiation level, UK
To arrive at a more meaningful comparison between
the energy performance of the different installations this
climate data was used to predict the annual power yield
for each of the five installation methods.
...continues
subscribe
Get it first - new briefings direct
to your inbox (max one/month)
R
viridian
solar
The module temperature above ambient for each point in
the data set is calculated from the measured irradiation
using the temperature characteristic curves derived
earlier (figure 8). The module operating temperature
is simply calculated by adding the measured ambient
temperature at the same time interval. The module
instantaneous power output is calculated from the
measured irradiation level taking into account the
module temperature coefficient and the calculated
operating temperature.
The modest reduction in enegy yield may be a trade off
that many customers are willing to make in return for the
enhanced aesthetics of integrated solar installations.
The energy output for each three minute interval is
calculated and summed for the whole year.
Results
Figure 10 shows the relative annual energy output from
each of the five installation methods, with in-roof (cold
roof) set to 100%.
Annual Energy Yield
(Cold roof integrated = 100%)
110%
108%
106%
104%
102%
100%
98%
96%
94%
92%
Free Standing Above Roof
Cold Roof
Warm Roof
Shingle Roof
In-roof
Figure 10 Impact of installation on annual energy yield
It can be seen that for the climate file used, the freestanding PV module with open rear would generate
6.0% more energy than the in-roof module, while the
installation above the pitched roof produces only 3.1%
more.
There is an very small difference (0.1%) between the
cold roof build up and the warm roof build up for inroof installations, reflecting the very similar temperature
responses of the two construction methods.
The in-roof module fixed to plywood sheeting would
produce 1.1% less energy than the cold roof detail.
A belief that solar PV should always be installed with
the highest possible level of shade-side ventilation
is widespread in the industry. This study provides
evidence to temper that position and enable solar
professionals to offer advice to prospective customers
based on evidence rather than opinion.
subscribe
Get it first - new briefings direct
to your inbox (max one/month)
Viridian Solar
Atlas Building, 68 Stirling Way,
Papworth, Cambridge UK
CB23 3GY
Tel +44 (0)1480 831501
[email protected]
www.viridiansolar.co.uk