-
10 - 1
The mechanical integration of photovoltaic arrays requires an
understanding of the site conditions, the physical and electrical
characteristics of PV modules chosen, the desired electrical output
for the array, and the mounting system and structural attachments.
It also involves considerations for the installation, maintenance
and accessibility of equipment, and architectural integration. The
objective is to produce the least-cost mechanical installation that
is safe, secure, appealing and appropriate for the application.
References:Photovoltaic Systems, Chap. 10Minimum Design Loads
for Buildings and Other Structures, ASCE 7Wind Load Calculations
for PV Arrays; Stephen Barkaszi, FSEC & Colleen O’Brien, BEW
Engineering: www.solarabcs.org/wind/
Mounting hardware manufacturers websites:Unirac:
www.unirac.comProfessional Solar Products: www.prosolar.comIron
Ridge: www.ironridge.comDirect Power & Water:
www.dpwsolar.com
10 - 2
Reference: Photovoltaic Systems, p. 255
-
10 - 3
PV arrays are constructed from building blocks of individual PV
modules, panels and subarrays that form a mechanically and
electrically integrated DC power generation unit. The mechanical
and electrical layout and installation of PV arrays involves many
interrelated considerations and tradeoffs that are affected by the
system design, the equipment used and the site conditions.
Reference: Photovoltaic Systems, p. 255-260
10 - 4
Reference: Photovoltaic Systems, p. 255-260
-
10 - 5
Higher operating temperature reduce array voltage, power output
and energy production, and accelerate degradation of modules and
their performance over many years. Mounting system designs have a
strong effect on average and peak array operating temperatures.
Rack mounted arrays have the greatest passive cooling and lowest
operating temperatures, with temperature rise coefficients from 15
to 25 °C/kW/m2. Direct mounts have the highest operating
temperatures, with temperature rise coefficients of 35 to 40
°C/kW/m2. Standoff mounts have moderate operating temperatures,
depending on the standoff height.
Reference: Photovoltaic Systems, p. 257-258
10 - 6
Preferably, PV modules in source circuits are installed in a
single row or rack, with each module adjacent to another, with the
module junction boxes aligned on the same sides to facilitate
wiring connections. Note that PV module connector leads are only a
certain length and additional cabling and connectors may be
required for non-standard installations.
Reference: Photovoltaic Systems, p. 255-260
-
10 - 7
Reference: Photovoltaic Systems, p. 268, 279-280
10 - 8
Building codes require all roofing system and building
penetrations to be properly flashed and weathersealed. There are a
variety of weathersealing methods used, depending on the type of
penetrations or attachments used, with some requiring sealants and
others not. Weather sealants, where required, should maintain
flexibility over 30-50 year life and expected temperature extremes,
readily adhere to roofing and construction materials, and have UV
inhibitors to resist degradation from sunlight. Butyl rubber,
elastomeric and polyurethane based sealants have long life and
readily adhere to dry roofing materials, including metal, wood and
asphalt shingles. Silicone, latex and acrylic sealants are
generally not suitable for the application.
Reference: Photovoltaic Systems, p. 279-280
-
10 - 9
Reference: Photovoltaic Systems, p. 256-260
10 - 10
Reference: Photovoltaic Systems, p. 256-260
-
10 - 11
PV arrays can be mounted on the ground or attached to buildings
or other structures using a variety of methods. PV array mounting
orientations can also be classified as fixed-tilt, adjustable or
sun-tracking mounts. Ground-mounted designs include racks, poles
mounts and sun-tracking arrays. Common building mounts include
standoff mounts and rack mounts that can be retrofitted to existing
rooftops. Building-integrated PV (BIPV) arrays, including direct
mounts and integral mounts are integrated with building components
and cladding materials such as windows, awnings and roofing
tiles.
Reference: Photovoltaic Systems, p. 260-267
10 - 12
Fixed-tilt PV arrays are non-movable structures that position
the PV array in a constant orientation. Fixed PV arrays installed
in northern latitudes are tilted up from the horizontal and
oriented toward the south, and away from shading obstructions to
maximize the solar energy received. Most PV arrays installed on
buildings and ground mounts are fixed-tilt PV arrays.
Reference: Photovoltaic Systems, p. 260-267
-
10 - 13
Adjustable-tilt PV arrays use mounting structures with removable
fasteners, telescoping legs or other manual means to allow for
seasonal adjustments of the array tilt angle. Adjusting the tilt
angle of PV arrays twice per year, around the time of the equinoxes
in the spring and fall, can marginally improve system output, but
is generally not practiced for most installations.
Suggested Exercise: Compare the differences for your location
for the average peak sun hours on a fixed south-facing latitude
tilt surface compared to an adjustable south-facing surface
adjusted to latitude - 15° tilt on April 1st, and adjusted to
latitude + 15° tilt on October 1st.
Reference: Photovoltaic Systems, p. 261
10 - 14
Sun-tracking arrays use mounting structures that automatically
and continually move the array surface array to follow the sun’s
position throughout the day. Sun-tracking arrays are characterized
by their tracking mode and whether they track the sun on one or two
axes. Most single-axis trackers are designed to move the array
surface east to west on a north-south tracking axis that is tilted
from the horizontal.
Sun-tracking arrays can receive up to 20-30% more solar
radiation than fixed south-facing arrays. Single-axis trackers do
not point exactly to the sun at all times, but generally receive
20% or more solar radiation than received on south-facing
fixed-tilt surfaces. Marginal benefits are achieved in going from
single to two-axis tracking. Point-focus concentrating PV modules
require two-axis sun-tracking to capture the direct beam solar
radiation component, while linear-focus concentrators can use
single-axis tracking.
Reference: Photovoltaic Systems, p. 266-267
-
10 - 15
Sun-tracking arrays can be controlled by passive of active
means. Passive means use solar heating of working fluids in the
tracker internal structure to create weigh shift and move the
tracker, or to pressurize piston actuators to move the structure.
Active tracking methods use electromechanical drives or stepper
motors that are controlled by smaller PV modules attached to the
array, or by external power sources and microprocessors.
Reference: Photovoltaic Systems, p. 266-267
10 - 16
The tradeoff for sun-tracking arrays involves considering the
initial costs and recurring maintenance, versus the cost of
additional modules on a fixed array to achieve similar energy
performance. Tracking PV arrays are usually installed on the ground
as opposed to on top of buildings due the large structural loads at
the foundations. Sun tracking arrays also require larger surface
areas and sufficient spacing between individual trackers to avoid
one tracker shading an adjacent one. Some tracker designs use a
backtracking approach to limit the tracker movement early in the
morning and late in the afternoon and sacrifice some solar energy
gain, in order to permit closer spacing of individual trackers and
avoid shading.
Reference: Photovoltaic Systems, p. 266-267
-
10 - 17
Active sun-tracking arrays use hydraulic pistons or motors to
drive the tracking mechanism.
Reference: Photovoltaic Systems, p. 266-267
10 - 18
Ground-mounted arrays are detached from buildings, and usually
permit the greatest flexibility in mounting and orienting the
array. Types of ground mounts include racks, poles and sun-tracking
mounts. Ground mounts require anchoring to foundations such as
concrete, setting poles directly in the soil, or by
self-ballasting. The site conditions and the methods and materials
specified by the mounting system designer manufacturer dictate the
best installation practices.
Since ground-mounted arrays are typically at lower elevations,
shading from nearby trees, fences, buildings and other obstructions
may be a concern. Ground-mounted PV arrays generally require
restricted access by fencing or elevating the array to reduce
safety hazards.
Reference: Photovoltaic Systems, p. 264-265
-
10 - 19
Reference: Photovoltaic Systems, p. 261-265
10 - 20
Rack-mounted arrays are commonly used on the ground, buildings
and other structures, and offer the greatest flexibility in
mounting the array at specific tilt angles. Small rack-mounted
arrays can be installed on poles, and larger racks can be installed
in multiple rows for larger arrays.
Reference: Photovoltaic Systems, p. 261-265
-
10 - 21
Reference: Photovoltaic Systems, p. 261-265
10 - 22
Some array structures may be installed as a monolithic unit by a
crane.
Reference: Photovoltaic Systems, p. 261-265
-
10 - 23
Rack-mounted arrays can be elevated on poles above harms way for
safety and protection.
Reference: Photovoltaic Systems, p. 261-265
10 - 24
Pole-mounted arrays employ either fixed, adjustable, or
sun-tracking arrays installed on a rigid metal pipe or wooden pole.
Pole-mounted designs allow the arrays to be elevated to protect
from harm and to avoid shading, and most allow the array azimuth
angle to be rotated for optimal orientation. Due to the large
foundation loads, pole mounts are usually installed on the ground
and not on buildings, and have limitations on the size of array
they can support based on he size of the pole and foundation.
Reference: Photovoltaic Systems, p. 265-266
-
10 - 25
Reference: Photovoltaic Systems, p. 265-266
10 - 26
Reference: Photovoltaic Systems, p. 265-266
-
10 - 27
Depending on the site soil conditions, poles may be set directly
into holes and compacted, or may require reinforced concrete
foundations. Many types of standard utility and light poles are
rated for wind loads, and their ability to support heavy equipment.
Foundations used for pole mounts must have sufficient mass and any
attachments must have appropriate strength to counter the forces
attempting to blow the pole over during high wind load
conditions.
Reference: Photovoltaic Systems, p. 265-266
10 - 28
Roof-mounted PV arrays may use standoff mounts, rack mounts and
building-integrated PV arrays. Rooftops often have large areas of
unused space, and are popular locations for installing PV arrays.
Rooftop locations provide higher elevations that help avoid shading
and offer additional protection and safety for the array. Most
roof-mounted PV arrays use fixed-tilt support structures that are
retrofitted to existing rooftops. Roof mounts may also be
classified according to the type of roof structure or roof covering
the array attaches to, such as sloped or flat roofs, or asphalt
shingle, metal, tiles or composition roofing materials. For
practical and structural considerations, roof-mounts generally do
not use movable sun tracking arrays or pole mounts.
Reference: Photovoltaic Systems, p. 260-263
-
10 - 29
Standoff-mounted arrays are the most common way PV arrays are
attached to sloped rooftops. Standoff mounts typically locate the
PV modules 3 to 5 inches above and parallel to the roof plane. They
are not usually tilted at a different angle than the roof surface.
This is because the added complexity and costs required to install
mounting structures obliquely with respect to the roof surface do
not usually justify the marginal increases in solar energy gain and
system performance. Several manufacturers provide standard mounting
hardware for standoff arrays that meet the structural loads for
most applications.
Reference: Photovoltaic Systems, p. 265-266
10 - 30
A self-ballasted PV array is a type of rack mount that relies on
the weight of a the PV modules, support structure and additional
ballast material to secure the array. Self-ballasted arrays are
intended to reduce or eliminate direct structural connections to a
building or foundation, thereby avoiding additional labor and
weathersealing concerns. Typical ballast materials include sand and
concrete blocks installed in trays at the bottom of the racks.
Self-ballasted arrays usually require additional restraints in
seismic and high wind load regions.
Reference: Photovoltaic Systems, p. 261
-
10 - 31
Building-integrated PV (BIPV) arrays replace conventional
building cladding, where PV modules are integrated into roofing
materials, glazing, awnings and other architectural features.
Integral mounts and direct mounts are types of BIPV arrays. The
advantage of BIPV arrays is that the PV array replaces conventional
building materials, saving on materials and construction costs.
Most BIPV arrays use custom designed modules and require special
installation procedures.
Reference: Photovoltaic Systems, p. 261
10 - 32
A direct mount is a fixed array mounting system where the PV
modules are attached flush to an existing roof surface or decking.
Special PV modules and custom array designs are typically required
for direct mount applications. Direct-mounted arrays also
experience high operating temperatures.
Reference: Photovoltaic Systems, p. 261
-
10 - 33
Integral-mounted arrays are a type of BIPV array where PV
modules replace conventional building cladding, such as roofing and
window systems. Integral mounts are custom designs using special
installation and weathersealing procedures.
Reference: Photovoltaic Systems, p. 261
10 - 34
Mobile PV arrays require custom designs and special attachments
to protect the array from shock and vibrations.
Reference: Photovoltaic Systems, p. 260-263
-
10 - 35
How not to install PV arrays!
10 - 36
The attachment of PV arrays to buildings and other structures is
governed by the standard ASCE 7, Minimum Design Loads for Buildings
and Other Structures, which is adopted into most building codes
throughout the U.S. However, it does not specifically address the
installation of roof-mounted arrays. PV system installers and
designers are responsible for ensuring that the design and
structural attachments of PV arrays meet all anticipated loads, and
that allowable loads on existing structures and mounting systems
are not exceeded.
References:Photovoltaic Systems, p. 268-279Minimum Design Loads
for Buildings and Other Structures, ASCE 7
-
10 - 37
PV arrays must be designed and secured to withstand the maximum
possible mechanical loads.
Dead loads (D) are static loads due to the weight of the array
and mounting hardware, typically about 5 pounds per square foot
(psf) for most PV arrays. Self-ballasted arrays can have
substantially higher dead loads.Live loads (L) are loads from
temporary equipment and personnel during maintenance activities.
Generally these loads are small for PV installations, on the order
of 3 psf. Typical flat and pitched roofs must be designed for a
minimum uniformly distributed live load of 20 psf. All roofs
subject to maintenance workers must be designed for a minimum
concentrated point load of 300 lbs.Wind loads (W) are typically the
highest of all loads experienced by PV arrays, and the only
uplifting force. Snow loads (S) can be up to and greater than 20
psf in northern climates.Hydrostatic Loads (H) are due to the
lateral pressure of the earth (soil) or ground water pressure on a
structure.Seismic (earthquake) loads (E) are based on region and
seismic design category.
References:Photovoltaic Systems, p. 268-279Minimum Design Loads
for Buildings and Other Structures, ASCE 7
10 - 38
Allowable Stress Design is method to determine the design loads
for structural materials based on the maximum allowable elastic
stress limits for the structural materials used. Consequently, it
includes a factor of safety for unfactored loads. Allowable stress
design considers various load combinations, and the most
unfavorable loading condition is used for structural design.
Reference: Minimum Design Loads for Buildings and Other
Structures, ASCE 7, Chap. 2
-
10 - 39
For common standoff-mounted PV arrays installed just above and
parallel to sloped roofs, the mounting methods are quite similar
among different PV modules and mounting system suppliers.
Key points of the structural evaluation include:•PV module
allowable loads and required position of deflection support and
attachments.•PV module attachments to underlying beams or rails
(machine screws or clamps).•Allowable deflections in beams or
rails•Point attachments to structure
Reference: Photovoltaic Systems, p. 268-279
10 - 40
The Main Wind-Force Resisting System (MWFRS) consists of
structural elements that provide support and stability for the
structure, like walls. beams, trusses, roof deck, etc. The PV array
attachments to a structure may be considered part of the MWFRS.
Components and cladding are building envelope materials such as
roof coverings and glazing that are not part of the MWFRS. PV
modules and attachments to the immediate support structure are
considered components and cladding. Different calculations also
apply to wind load calculations for the MWFRS and for components
and cladding. Higher wind loads generally apply to components and
cladding where the load is distributed over a smaller area than for
the MWFRS. A conservative approach evaluates all PV array elements
as components and cladding.
Reference: Minimum Design Loads for Buildings and Other
Structures, ASCE 7, Chap. 6
-
10 - 41
Wind loads are by far the most significant concern for mounting
PV arrays for most applications. Most PV modules are listed to
handle wind loads of 2400 Pa (50 psf), and some are tested for
loads up to 5400 Pa (112 psf). Generally, PV modules must be
supported in certain positions to achieve maximum load capability.
Refer to specific PV module manufacturer’s installation instruction
for allowable mounting configurations and maximum loads.
Three methods can be used to determine the design wind loads for
buildings and other structures; a simplified procedure, an
analytical procedure, and a wind tunnel procedure. The simplified
procedure applies to many residential and commercial buildings.
References:Photovoltaic Systems, p. 268-279ASCE 7, Chap. 6
10 - 42
The design loads for components and cladding can be computed
using the simplified method if certain conditions are met. The
building must be enclosed and regular-shaped, and must have a flat
roof or a gable roof with slope no more than 45 degrees, or a hip
roof sloped no more than 27 degrees. The mean building height must
be no more than 60 ft and the building or site must not have
unusual characteristics or wind response.
References:Photovoltaic Systems, p. 268-279ASCE 7, Chap. 6
-
10 - 43
Reference: ASCE 7, Chap. 6
10 - 44
Basic wind speed maps show the maximum design wind speed by
location.
References: Photovoltaic Systems, p. 272ASCE 7, Chap. 6
-
10 - 45
Buildings and other structures are classified in categories
based on their consequences of failure. Buildings and structures
classified in higher categories must be designed for greater
loads.
Reference: ASCE 7-10, Table 1-1
10 - 46
Reference: ASCE 7, Chap. 6
-
10 - 47
Most buildings in urban and suburban areas fall under exposure
B.
Reference: ASCE 7, Chap. 6
10 - 48
Adjustments factors increase wind loads for building heights
above 30 feet and for exposure category.
Reference: ASCE 7, Figure 6-3
-
10 - 49
Reference: ASCE 7, Chap. 6
10 - 50
Pressure coefficient zones define areas of a roof with higher
wind loads. The width of the pressure coefficient zone is a minimum
of 3 feet, or 10% of the smallest horizontal dimension or 0.4h,
whichever is greatest.
Reference: ASCE 7, Figure 6-3
-
10 - 51
Net Design Wind Pressures for components and cladding are
determined from the basic wind speed, roof type and slope, pressure
zone and effective wind area. These net pressures are defined for
Exposure B, for a mean roof height of 30 feet, for importance
factor = 1 and for topographic factor = 1. The effective wind area
is the area tributary to a single set of supports. Higher wind
pressures are given for smaller effective wind areas. The given
pressure coefficients are applied as downforce (positive) and
uplift (negative) forces perpendicular to the building surfaces.
The positive and negative design wind pressures used for load
calculations must not be less than +10 psf and -10 psf,
respectively.
Reference: ASCE 7, Figure 6-3
10 - 52
Standoff PV arrays attached close to and parallel with sloped
roof surfaces can be considered components and cladding, and design
loads can usually be evaluated using the simplified method.
Reference: ASCE 7, Chap. 6
-
10 - 53
Most PV modules and pre-engineered mounting systems are designed
to support the loads determined in the example.
Reference: ASCE 7, Chap. 6
10 - 54
PV array mounting system designs and all components must be able
to withstand the maximum forces expected in any given application.
Oftentimes, independent engineering or test results may be required
to certify PV array structural designs for local building code
compliance. The critical design area is usually the point
attachments of the array mounting system to a structure.
Reference: Photovoltaic Systems, p. 268-279
-
10 - 55
PV module specifications give the maximum mechanical loads that
the module can support using specified supports and
attachments.
Reference: Photovoltaic Systems, p. 268-279
10 - 56
Most standard flat-plate PV modules are glass laminates enclosed
in an aluminum frame. The frame provides mechanical support for the
laminate, and a means to structurally attach the module to a
mounting system and for electrical grounding. PV modules are either
bolted with fasteners or clamped to supporting rails or beams.
Follow the PV module manufacturer’s installation instructions for
the allowable mounting points to meet the maximum loads.
Suggested Exercise: Review PV module installation instructions
for approved mounting and attachment methods.
Reference: Photovoltaic Systems, p. 268-279
-
10 - 57
Commonly, PV modules are installed across two aluminum rails
that act as structural beams. The rails are then bolted to the
underlying point attachments at specific locations along the rail
that define the rail spans. PV arrays installed in higher wind
regions require stronger rails, or smaller spacing between rail
attachments to avoid excessive rail and module deflections.
In common sloped rooftop applications, the rails are usually
laid out with the length in an east-west direction across the roof,
which permits variable width attachments to the underlying roof
structural members, such as rafters or trusses. As the spacing
between rafters or trusses is usually fixed, this may constrain the
installation of rails up and down the roof slope (in a north-south
direction). This is because PV modules require the support rails to
be located at certain points on the module frame to support the
specified mechanical loads. Refer to mounting hardware
manufacturer’s data on maximum allowable loads and deflection on
module support beams.
Suggested Exercise: Review some mounting system manufacturer’s
websites and tools to determine the appropriate mounting hardware
and support structures using any PV module for any size PV array
for various applications.
References: Photovoltaic Systems, p. 268-279
10 - 58
Point attachments connect the array assembly to the underlying
structure (building or ground) at specified intervals. Point
attachments produce concentrated loads on a structure or
foundation.
Reference: Photovoltaic Systems, p. 268-279
-
10 - 59
Reference: Photovoltaic Systems, p. 268-279
10 - 60
Point attachments are usually made directly to structural
members like trusses, not to the roof decking. Additional blocking
between trusses may be required for some installations.
Reference: Photovoltaic Systems, p. 268-279
-
10 - 61
Array installations on metal roofs use special clamps that crimp
to the metal roof seams, and rely on adequate attachment of the
metal roofing to the underlying structure. These attachments do not
penetrate the roof surface, and therefore avoid problems with
leakage. Installations on slate or tile roofs use special
attachments that direct the mechanical loads on the array directly
to the roof structure, not on the tiles.
Reference: Photovoltaic Systems, p. 268-279
10 - 62
Lag screws are commonly used to secure point attachments
directly to the tops of trusses or other structural members. The
screw diameter, thread embedment and species of lumber determine
the allowable withdrawal loads. A proper size pilot hole should be
drilled 60-70% of the screw shank diameter, unless self-drilling
SPAX lag screws are used.
The allowable withdrawal loads in the table are calculated for
lag screws in the side grain of seasoned wood, using the empirical
formula: P = 1800G3/2D3/4, where G is the specific gravity, D is
the screw diameter in inches, and P is the allowable withdrawal
load in pounds per inches of thread penetration depth, and includes
a factor of safety of 4.
References:Photovoltaic Systems, p. 274-275Marks’ Standard
Handbook for Mechanical Engineers, 8th Ed.
-
10 - 63
Reference: Photovoltaic Systems, p. 268-279
10 - 64
Lag screw load calculations are based on point loads, diameter
of the screw, thread embedment depth and type of lumber. Note that
lag screws longer than 1 inch are not threaded the full length of
the shank.
Reference: Photovoltaic Systems, p. 268-279
-
10 - 65
Many manufacturers offer universal hardware for installing all
types and sizes of flat-plate PV modules on racks, poles, trackers,
and on many types of roofs and other building surfaces. This
greatly improves standardization, and reduces design, materials,
and labor costs associated with installing PV arrays.
Reference: www.unirac.com
10 - 66
Reference: www.prosolar.com
-
10 - 67
Reference: Photovoltaic Systems, p. 281-284
10 - 68
Reference: Photovoltaic Systems, p. 281-284