Numerical simulation of wind effects on a stand-alone ground mounted photovoltaic (PV) system

• "Critical wind directions according to this study are: 0° for maximum drag, 180° for maximum uplift and 45° and 135° are for maximum overturning moments."
• "For 0° and 180° wind directions, symmetry in both mean Cp distributions as well as the wake structures, about the stream-wise midline of the panel, is observed."
• Mean Cp values range from -1.5 to 1.5 when loading at an attack angle of 0 and 180 degrees respectively.
• "Critical wind directions according to this study are: 0° for maximum drag, 180° for maximum uplift and 45° and 135° are for maximum overturning moments."

Solar Ready Guidelines for Solar Domestic Hot Water and Photovoltaic Systems

• When designing the roof structure to accommodate a solar system, an additional design dead load of 0.24 kPa (5 psf) accommodates the weight of SDHW collectors and/or solar PV modules as well as all mounting hardware and internal fluids for the majority of CSA certified systems when they are mounted in parallel to the roof surface.

Wind Loading on Full-Scale Solar Panels

Abstract: "Wind load governs the design of supporting structures of solar panels and constitutes approximately fifty percent of the total cost. There are various test scale related issues while testing solar panels (small structures) in boundary layer wind tunnel laboratories meant for tall buildings (large structures). Emergence of large testing facilities, however, is enabling testing full-scale solar panels. In this thesis an extensive experimental program is conducted at WindEEE Dome using full-scale solar panels and finite element modelling. The experimental program includes: (i) high resolution pressure tests to understand the sensitivity of pressure taps density and distribution; (ii) force balance test to determine the reactions of the solar panel under wind loading accounting for aeroelastic effects and validate pressure test results; (iii) finite element modelling to assess the internal stress of the solar rack elements and improvement of the rack cross section. Study of pressure tap layout and resolution illustrated that a fairly high density resolution is required to capture all the aerodynamic features of pressure on the solar panel surfaces. It is found that the uplift force obtained from the force balance tests are larger compared to the pressure taps as result of the dynamic effects of wind loading. The finite element analysis of solar racks was performed using the experimentally established wind loading data, which includes all dynamic features of the forces obtained from the force balances. It is concluded that the solar rack cross sections can be structurally optimized and there is a possibility to save in aluminum elements up to 40%."

• "The National Building Code of Canada (NBCC 2010) National Building Code of Canada (NBCC, 2010) currently does not provide specific guideline for determination of wind loading on ground-mounted solar panels. In current practice, the designers of solar panel select the pressure coefficients based on their interpretation from the code by using similar geometries outlined in NBCC 2010. Due to the lack of a unique applicable standard for wind load estimation on solar panels in Canada, most of the designers refer to ASCE 7-10 which specifies pressure coefficients of mono-slope free roofs for different elevations with respect to the ground levels."
• "However, when the panel is ground mounted, the minimum and maximum peak pressure coefficients were occurred at 30 degree and 135 degree wind angles of attack, respectively. The 45 degree inclination angle of the panel resulted in both maximum and minimum peak pressures"
• "The inclination of the solar panels is set at 25° as typically used for southern Ontario latitude, for high energy product during summer seasons."
• Drag force coefficient obtained from ASCE CpCg = 1.95, experimental analysis = 1.75 at 0 degree wind angle of attack.
• Aluminum frame for racking system: Aluminum 6061-T6. Density = 2700 kg/m3, Yield Strength = 270 MPa, Shear Strength = 200 MPa. Resistance factor = 0.9
• As per NBCC, London Ontario specified wind speed for a 50 year return period is V=28.9 m/s. Wind speed used in analysis is V=15.88m/s.

Photovoltaic Ready Guidelines

• "To prepare for a photovoltaic system, one PV conduit of at least 3.8 cm (1-1/2") nominal diameter constructed of rigid or flexible metal conduit, electrical metallic tubing (EMT), rigid PVC conduit, or liquid-tight flexible conduit (as per Section 12 of the Canadian Electrical Code Part 1 concerning "raceways") should be installed."
• "In installations with any bends or elbows greater than 45 degrees, a nylon pull-rope (6 mm (¼") diameter or larger) should be installed in the conduit to facilitate installation of conductors at a future date."
• "When designing the roof structure to accommodate a photovoltaic system, an additional design dead load of at least 0.17 kPa (3.5 psf) accommodates the weight of solar PV modules as well as all mounting hardware for CSA certified components when they are mounted in parallel to the roof surface. Systems mounted at an angle to the roof surface (i.e., rack mounted systems) and ballasted systems may incur additional loads beyond the 0.17 kPa (3.5 psf) dead load."
• "Snow management devices such as snow clips and fences are available for attaching to PV arrays to slow and breakup the release of snow. These attachments are typically limited to regions with snow loads up to 2.39 kPa (50 psf)."
• "It should also be noted that systems mounted at low angles (generally 15° (4/12 pitch) or less) will not shed snow as well as systems mounted at slightly steeper angles and will thus not perform as well in winter months."

Auger and Ground Screw Application Guidelines: Sunmodo

• "Ground augers in the poorest of soils easily provide 3000 to 10000 pounds of pullout strength, far more than the 1500 to 2500 pounds required for PV arrays."
• "However, with smaller PV systems one may not need to spend money on a soil engineering analysis and the cost to permit the design separately."
• "If the ground is too rocky, other options such as post and concrete, ballasted arrays, or rock anchors may be a better alternative."
• "However, in soft, loamy soils a ground screw will not provide big pullout values compared to an auger."

The Soils of Middlesex County: Ministry of Agriculture and Food

• London Ontario USDS Soil classifications: Sandy loam, loam at depths of 30cm, silty clay >30cm.
• On sheet 3 of Soil Survey Report No. 56, London Ontario is classified as "Not mapped - Significant areas of land which has been disturbed, modified, or permanently withdrawn from agricultural use".

National Building Code of Canada 2015: Specified Snow Loads

• "The specific weight of old snow generally ranges from 2 to 5 kN/m3, and it is usually assumed in Canada that 1 kN/m3 is the average for new snow. Average specific weights of the seasonal snow pack have been derived for different regions across the country(6) and an appropriate value has been assigned to each weather station. Typically, the values average 2.01 kN/m3 east of the continental divide (except for 2.94 kN/m3 north of the treeline), and range from 2.55 to 4.21 kN/m3 west of the divide."
• Page 581: Specified Snow Load in London, ON is 1.9 KPa. Max Snow Load in Ontario = 3.4 KPa.
• "In no case shall the specified snow load be less than 1 kPa"
• "Snow removal by mechanical, thermal, manual or other means shall not be used as a rationale to reduce design snow loads"
• Associated rain load in London, ON = 0.4 kPa
• Hourly Wind Pressure in London, ON = 0.47 kPa

National Building Code of Canada 2015: Specified Wind Loads

• Minimum Specified Roof Surface concentrated load = 1.3 kN
• For a 4/12 pitch, in London ON, in open terrain, minimum wind load = 0.68 KPa
• Critical load combinations of solar panel supporting structure (Ultimate Limit State): Case 1) (1.25 or 0.9)Dead + 1.4Wind (+ 0.5Snow Companion load). Case 2) (1.25 or 0.9)Dead + 1.5Snow (+ 0.4Wind Companion load).
• CpCg at a 0/180 degree contact angle = -2.0
• London Ontario's 50 year return wind speed = 97.07 km/h

National Building Code of Canada 2015: Foundations & Footings

• Page 810: Minimum depth of foundation for silts, clays, or soils not clearly defined is to be 1.2 m, or below the depth of the frost penetration, whichever is more. No limit for coarse grained soils that have "good soil drainage".
• Page 820: Footings shall rest on undisturbed soil, rock, or compacted granular fill.
• Page 819: Wood or concrete footings not subject to surcharge on stable soils with an allowable bearing pressure of 75 kPa or greater.

Page 1257: The requirements for clay soils or soils not clearly defined are intended to apply to those soils that are subject to significant volume changes with changes in moisture content.

Quikrete Concrete Mix Data Sheet: ASTM Compliance

• Slump test complies with ASTM C143
• Unit Weight complies with ASTM C138
• 28 day Compressive Strength: 27.5 MPa. Complies with ASTM C39
• Table provides ideal amount of water required to pass slump test

LG 410W NeON2 BiFacial Solar Panel

• Withstands a front load of 5.4 kPa and 4.0 kPa back load. Almost a Factor of Safety of 2.

Extreme 50 year return wind speeds from the USAF data set

• 50 Year Wind Speed of Western Africa is about 15 m/s
• For v = 15 m/s, Cd = 1.0, Air Pressure = 1.2 kg/m^3, Wind Pressure = 0.27 kPa

Sign Support Manual - Ministry of Transportation

• Division 6 - Timber Design Supports
• "Signs with 2 posts were designed using corresponding tributary sign area. For signs with 3 or 4 posts, the distribution of wind reactions on the posts were analysed using continuous beam theory. The governing limiting Hmax is based on the maximum wind load acting on either the exterior or interior post."
• "For design purposes, the post height used in flexure calculations, was 200mm below ground surface, to allow for post flexibility within the soil."
• "Timber shall be Coastal Douglas Fir or Jack Pine No.1 Grade, S4S, in accordance with CSA Standard 086 for beams and stringers, NLGA grading rules.

(ii) All wood shall be pressure preservative treated in accordance with OPSS 1601."

Connector and splice bolts and nuts shall be in accordance with ASTM Specification A325M. Footing bolts shall be in accordance with ASTM Specification A307M. (vi) Pressed steel shear plates shall be in accordance with CSA Standard 086. (vii) All steel including nuts and bolts shall be hot dip galvanized.

• "The indicated footing depths (given in Figure 6.5.3) are the absolute minimum for each post size based on a passive earth pressure of 68 kPa (1400 psf) at SLS."
• As per the design tables on Page 200, the maximum height above the ground to the centre of the sign for a 1200x6000 sign using 4 6x6s is 2620mm for a local wind pressure of 465 Pa. For 3 posts and a 1200x1800 sign, Hmax = 2040mm.
• The 25-year hourly mean wind pressure in London Ontario is 455 kPa.

Vertical bifacial solar farms: Physics, design, and global optimization

• Assumes a module height of 1.2 m.
• "The optimally tilted farm yields 32% more energy than the vertical farm."
• " Unfortunately, close to equator (within latitudes), yield-optimized monofacial farms have a row-spacing less than 0.25 m, which could be difficult to install and maintain. For example, the row-spacing in the farm is required to be 2 m or higher [47]. Therefore, the 'yield-optimized' comparison of the farms close to equator (as discussed in the preceding section) may not be practical. Therefore, next we compare the energy yield of the farms with fixed 2 m row-spacing."

Wind Loading on Solar Panels at Different Inclination Angles

• Experimentally, Cd for 90 degree attack angles = 2.0
• For 90 degree attack angles, modules spaced a distance X away with an array depth D having a ratio of X/D = 2.0 will have 25% more drag than a system with X/D = 1.0.

City of London - Fence By-Law

Fences and walls are subject to the following regulations:

  (a)   Fences and walls on all lots of record in all R Residential Districts which enclose property shall not exceed six feet (6') in height, measured from the surface of the ground, and shall not extend beyond the required minimum front yard setback line.

  (b)   Fences and walls in all R Residential Districts, on recorded lots having a lot area in excess of 20,000 square feet and a frontage of at least one hundred feet (100'), or acreage not included within the boundaries of a recorded plat, are excluded from these regulations.                                                               

  (c)   Permanent fences and walls on lots of record shall not contain barb wire, electric current or charge of electricity.

  (d)   Fences and walls enclosing public institutions, public uses, parks, playgrounds or public landscaped areas, shall not exceed eight feet (8') in height, measured from the surface of the ground, and shall not obstruct vision to an extent greater than fifty percent (50%) of their total area.

  (e)   All fences and walls shall comply with the building code of Richland County as it applies to fence and wall installation and materials.

• A fence is defined as, "a railing, wall, line of posts, wire, gate, boards, pickets or other similar substances, used to enclose or divide in whole or in part a yard or other land, to establish a property boundary, or to provide privacy; and includes any hedge or grouping of shrubs used for the same purpose located in corner visibility triangle or driveway visibility triangle".
• Since vertical systems are not used to divide or enclose a yard to establish a property boundary, it is not necessary to follow this By-Law.

Government of Canada: Beaufort Wind Scale Categories

• 100 km/h winds corresponds to some structural damage occurring, and trees can uproot.
• Hurricanes are classified as winds upwards of 118 km/h.

Wind Loading on Solar Panels at Different Inclination Angles

• Provides a comparison of drag coefficient results between CFD 2D - Re=1.2x10^5 and Experimental - Fage & Johansen.
• 0-15 Degrees = 0.25, 30 Degrees = 0.60, 40 Degrees = 1.00, 50 Degrees = 1.40, 60 Degrees = 1.75, 70 Degrees = 1.90, 80 Degrees = 2.00 90 Degrees = 2.10.

ASCE: MINIMUM DESIGN LOADS AND ASSOCIATED CRITERIA FOR BUILDINGS AND OTHER STRUCTURES (7-16)

• Includes a section for cantilever signs subject to gravity and lateral loads.
• 26.10.2 Provides the velocity pressure, q, which is composed of many factors such as topography, wind speed, terrain, etc.
• Provides a more accurate representation of gust accumulation upon cantilevers compared to assuming a roof structure in the NBCC.

Taylor, P.; Santo, H.; Choo, Y. s Current Blockage: Reduced Morison Forces on Space Frame Structures with High Hydrodynamic Area, and in Regular Waves and Current.

• Provides the true drag coefficient, Cd, of an open frame, perpendicular to the wind attack, to be 1.30.

IronRidge: Ground Mount for Open Fields

• Assuming a 1.43 kPa Snow load, by switching to 3" mechanical tubing, spans can be up to 8-12'.
• Uses inexpensive U-bolts for top cap rail connectors. Excellent for resisting uplift wind loads.
• Uses 340 MPa yield strength, 2" OD mechanical tubing.

IronRidge: Ground Screws

• A 3" or 4" ground screw designed to secure 2" or 3" mechanical pipe. Fastened using 3 galvanized hex bolts.
• Simple to install and remove, requires a hand pile driver.
• May require a soil survey report. Sedimentary/foliated rock, and sandy gravel may need consulting to verify soil bearing capacity.

DIY Solar Panel System

• Cost of racking on 5x5 array = $2508, using a racking system from ReadyRack
• 2x extended support members installed at end to place a 25th panel, only cost $22 USD for extension.

SunTurf Multi-purpose Ground Mount System

• Connections are U-clamped, allows for use of loose galvanized nuts and bolts as needed. More secure than U-bolt.
• Beam to joist connection consists of 2 sets of nuts and bolts installed from under the joist. Keeps bolts in tension rather than shear. Could be problematic for joist torsional loads.
• Bracing between columns to prevent buckling and enhance stability. Bracing turns system into truss elements.
• Helical Augers used for weak soils with high water tables, ground screws used for highly packed granular soils.
• Ground screws available in 63" and 80".
• Drawings stamped by P.Eng of Ontario. Max spacing between columns = 96". Max overhang = 34". Suggests a N/S column spacing of 75". Designed snow load = 1.8 KPa, governing wind loads = -1.49 and 2.21 KPa
• Max height of Southmost column = 3'9"
• Ground screws have a max compression load of 12.9 kN, and a max uplift of 10.67 kN.

DIY Solar Panel Ground Mount: Unistrut and PT Lumber

• A welded unistrut (Superstrut) frame is connected to a 2x6 using a a heavy duty gate hinge, allowing rotation to any angle. Hardware not galvanized.
• Unistrut diagonal member bracing may have questionable buckling capacity, could brace each member to each other.
• Lumber frame mimics a fence frame. 4x4 or 6x6 posts can be placed on ground using universal deck blocks. Overall very cost effective.

DIY Ground Mounted Solar Panels with Adjustable Angles

• Uses ASTM A153 Hot Dipped Galvanized Lag Screws, >150 ksi tensile strength.
• All deck hardware from Simpson Strong-Tie, all hot dipped galvanized and readily available at hardware stores.
• Must use D10 Galvanized Joist hanger nails for connections in shear, cannot use screws.
• Gate screws are zinc plated, and thus are problematic. Black coated gate hinges will last just as long as hot dipped galvanized hardware.
• Cannot keep lumber and panels on the ground due to weathering issues, must elevate.

Build An Inexpensive Solar Ground Mount

• Uses 7' and 2' posts (Cut from 10') and spaces 8'6" to achieve 30 degrees.
• 2x6x10 beams connected to posts with 2 lag screws at each post. 2x4 joists with unistruts rest on beams, but don't have hardware to resist lateral torsional buckling.