DESIGN PROJECT, EGEE 437, SPRING 2017 1

Solar Design for LEED Gold Proposed Garage,

Ferguson Township
Luke Meyers Eric Podwojski Nicholas Dobson Joel Galtieri

I. Executive Summary cially plausible, some constraints had to be altered. For

instance, the clients request to have onsite battery storage
T ECHNOLOGICAL advances are making solar power
an increasingly affordable form of renewable energy.
This means society is getting closer to having a clean,
would make the project costly. Financial analysis of the
new garage with different constraints is performed later in
this document. Due to many constraints and clients needs
pollution free source of electricity at a cheaper cost than
we were able to make recommendations on whether the
most fossil fuels. Tax incentives are also being put into
client should pursue a solar energy system on their new
place that are making PV systems even more economi-
building. With an estimated payback of 20 years, we rec-
cally attractive. According to UCSUSA By early 2014,
ommend Canadian Solar panels with a power output of
the United States had more than 480,000 solar systems
36 kW to meet the demands of the winter months. In the
installed, adding up to 13,400 megawatts (MW), enough
summer months, the panels will produce almost double the
to power some 2.4 million typical U.S. households. This
amount of energy than in the winter, which can be used to
rapidly growing trend is positively affecting both commer-
power the other pre-existing buildings around the Town-
cial and residential energy needs. The client, Ferguson
ships land. A battery is not recommended to use due to
Township is seeking to install a solar energy system on
cost that amplifies the payback period.
top of a LEED Gold certified garage building with the
possibility of onsite battery storage. To meet the clients
needs, we designed a roof-mounted photovoltaic power sys- II. Introduction
tem for the garage. This garage is going to act as a mainte-
nance garage as well as office space, which the client would
like to be fully powered by solar energy with an electri-
F ERGUSON Township put out a request for an engi-
neering firm to plan and build a new LEED Gold
garage on the site in early February 2017. A firm has not
cal storage system utilizing the grid as a back-up when
been selected as of when this document was written, how-
necessary. The client consists of Ferguson Townships lo-
ever a complete solar analysis of the site has been done so
cal government leaders. Ferguson Township is located in
when an engineering firm is selected to take on the project,
central Pennsylvania, a diverse climate region with four
they can use the data in this document to incorporate a
seasons. The site is located at 3147 Research Dr, State
solar energy conversion device into their building plans.
College, PA 16801 and the building is specifically at lati-
tude:40.762645 and longitude: -77.873797. Policy concern-
ing renewable energy is constantly changing on both the A. LEED
federal and state level. Pennsylvania has several incen-
Leadership in Energy and Environmental Design
tives to progress the use of solar energy. Several incentives
(LEED) is a sustainable building design certification that
used in this project to lower the cost of the energy system
can be obtained through the U.S. Green Building Coun-
are discussed further in the document. Using electricity
cil (USGBC). The certification is obtained by meeting
bills from the existing buildings on the site, it is estimated
requirements set forth by the USGBC. Different levels
that the new LEED Gold garage will use 2,200 kWh per
of LEED certification can be obtained by meeting more
month. To properly size the buildings load, we decided on
sustainable building requirements. With respect to this
using 102 photovoltaic panels arranged with 17 panels in
project, the client is seeking to obtain LEED Gold certifi-
series and 6 panels in parallel dependent on the inverter
cation. Installing a solar energy conversion device on the
we chose. Fortunately, there are not any obstructions to
proposed garage will help earn the LEED Gold certifica-
cause shade if the new garage is built higher than the ex-
tion.
isting garages surrounding it. To make this project finan-

L. Meyers is with the Department Department of Energy and

III. Proposed Site Plan
Mineral Engineering, The Pennsylvania State University, University
Park, PA 16802. e-mail: [email protected] The proposed garage will be approximately 9,000 square
E. Podwojski is with the Department of Energy and Mineral En- feet located on the Townships property– See Fig 1. The
gineering, The Pennsylvania State University, University Park, PA new garage will connect to garages directly next to it
16802. e-mail: [email protected]
N. Dobson is with the Department of Energy and Mineral En- that already exist– See Fig The plan is to have the ex-
gineering, The Pennsylvania State University, University Park, PA isting garages connected during construction phase. As
16802. e-mail: [email protected] long as the roofs of the existing buildings do not cast a
J. Galtieri is with the Department of Energy and Mineral Engineer-
ing, The Pennsylvania State University, University Park, PA 16802. large shadow on the LEED garage roof, there should not
e-mail: jwg5510psu.edu be much of an issue.
2 DESIGN PROJECT, EGEE 437, SPRING 2017

solar power to electricity. Potential-induced degradation

caused by stray currents can affect performance. Light-
induced degradation is due to stabilization of a panel after
installation. Pay back period is how long it will take for
the technology to pay for itself. Lastly, the real world per-
formance of products. All of this applies to solar panels
however there are some other factors that should be re-
searched before choosing inverters. You have to know the
AC/DC ratio, output power, input power, and the type of
inverter. Inverter type varies on whether the PV system
is connected to the grid or not as well as if a battery for
energy storage will be used.

B. Solar Panels: Monocrystalline vs. Polycrystalline

The first step in choosing a solar panel to be used for
Fig. 1. Ferguson Township’s provided layout and site a photovoltaic system is deciding between Monocrystalline
and Polycrystalline. Monocrystalline solar panels are built
out of a single continuous crystal structure and is the more
developed of the two since it has been in use longer. Mono
panels are also more efficient in converting electricity per
area by producing a higher amount of power per square
foot when used in arrays. They also have a smaller degra-
dation of power output due to falling temperatures when
compared to poly. Another benefit is the long life span
usually coming with around a 25 year warranty for a panel
depending on manufacturer. The disadvantage for mono
panels is that they are the most expensive. This is due to
manufacturers that still produce mono to target the more
premium side of the market. Polycrystalline panels are
also made from a silicone crystal structure however they
are allowed to cool when being processed, which gives them
the graininess and distinctive edges on the cells. The main
advantage for poly panels is that they are cheaper due to
Fig. 2. Ferguson Township’s Site Location
smaller production costs. They are also more tolerant to
heat which means they will fail less in extreme tempera-
tures. The disadvantages to poly however are simply the
IV. Literature Review advantages of mono explained earlier. The next area to
look at is price.
A. Choosing Technology
Cost is just one common factor when it comes to choos- C. Batteries: Lithium Ion vs. Lead-Acid
ing technology to be used in a photovoltaic system. The There are some big differences when comparing lithium
price of different technology is dependant on power out- ion batteries vs lead-acid batteries. First, lithium ion
put, physical size, material quality, brand, and durabil- batteries have 2000-5000 life cycles depending on how well
ity. Durability can be safely assumed based upon the war- you take care of it when compared to 300-500 life cycles
ranty period product feedback. Price will also vary with for lead-acid batteries. Secondly, lithium ion batteries
the number of units being purchased. Usually the cost are a much more clean technology and are more environ-
is cheaper per unit when buying in bulk. To decide on mentally safer than a lead-acid battery. Also, lithium-ion
the quantity of the product, you have to find how much batteries are nearly 100 percent efficient in both charge
power output you will need and also how much area you and discharge. Lead acid batteries inefficiency leads to a
have to install the panels specifically. One could also look loss of 15 amps while charging and rapid discharging drops
into what products meet criteria for government incen- voltage quickly and reduces the batteries capacity.The
tives which can in turn lower costs. There is also a set lithium ion battery is a better choice for our project
of other factors to look into. Panel tolerance shows how because of how our client wants use to utilize LEED and
much the power output could differ from the specifications. if lithium ion batteries are safer for the environment and
A bad tolerance could mean a 200 Watt panel might actu- are a cleaner technology than its the best battery for use
ally produce 194 Watts. Temperature coefficient explains to employ.
how heat can negatively affect panel operation. Conver-
sion efficiency is basically how efficient a panel converts
Luke Meyers: SOLAR DESIGN FOR LEED GOLD PROPOSED GARAGE 3

D. Comparing Two Specific Solar Companies

Solar World
• Largest US manufacturer for solar panels
• Produced solar panels for 40 years in both Germany
and US
• 25-year linear performance guarantee
• 20-year product workmanship warranty
Canadian Solar
• Based in Canada
• 16-year-old company and only getting bigger
• Have state-of-the-art manufacturing facilities in
Canada, China and Vietnam
• Leading Global solar company
• panels have high system energy yield at low irradiance
and low NOCT

V. Technological Constraints
A. System Losses
Losses occur in any PV system. Our system in particu-
lar, contains a string of panels connected to either several Fig. 3. System Losses
or a single inverter, then further connected to a battery
bank ( Dependent upon Simulation ran).The system losses
can be summarized in Fig. 3.
The task of the inverter is taking the Direct Current
(DC),created initially by the panels during photogenera-
tion, and convert it to Alternating Current (AC)– typi-
cally appliances and machines run off of AC current. In
doing so there is roughly some amount of energy lost. Of
the two inverters we simulated there was varying losses (A
loss of 1.8 percent for the 36kW inverter and a 2.43 per-
cent loss of energy for the 5200W converter). The decline
of efficiency occurs from small circuitry losses and moving
mechanical parts.The Panel’s soiling losses are caused from
the estimated layer of dust and pollen that accumulate on
the receiver’s surface. During the summer drought months
Fig. 4. Winter Months Optimum Tilt and Azimuth
and the pollen heavy fall and spring are the times when
this loss is at its highest. The remainder of the losses in
Fig. 3 are an accumulation of circuitry and wiring losses
B. Resource Losses and Optimizations
in the system. Be aware that the addition of a battery sys-
tem may be useful to store the daytime captured energy, By utilizing Skelion (Program Extension for Solar De-
but with this comes additional losses when AC energy is sign), the exact location and directional orientation of the
stored and then later dispatched. As a final, the panels proposed building’s site plan can be evaluated. This is
themselves have an efficiency of converting the incoming known as the Azimuth of the collector. This parameter is
irradiation into usable solar power. As advised before, the needed for estimating the amount of solar resource able to
installation of low efficiency panels is not the best choice. be captured– compared to that of the ideal condition. Sim-
The panels selected for review ranged from 16.17 to 18.20 ilarly the tilt of a collector is used to determine the amount
percent efficient. Price is usually a correspondent to ef- of usable irradiation absorbed by the collector. Depending
ficiency, so, the choice of panel would need to be made on the orientation of the roof and the building there is a
upon that parameter. The panel of best fit for this lo- certain percentage of collection available– seen in figure 4
cation was the Canadian Solar CS6U-335 mono. It was and 5. The Optimum location is within the red sector. As
noticed that no matter what panel was chosen the amount marked by the blue line, being the proposed realistic tilt of
of Dollars/kWh did not change substantially. The advised the roof, ( 14.04o ) and the black line being the building’s
choice was the Canadian Solar CS6U-335 mono technol- proposed location azimuth (144.71o , North = 0o and South
ogy with an efficiency of 17.3994 percent and a price of = 180o , East reducing toward North), this shows the loca-
70cents
kW h .This panel gave a close estimate to the demand of tion of our collectors being placed directly in the plane of
electricity for the winter months, as well as a good instal- the roof. Electric demand for the garages is highest in the
lation price– more economic concerns are given later in the winter months, so an additional tilt could be given to the
report. panels upon installation The Panels could be given an ad-
4 DESIGN PROJECT, EGEE 437, SPRING 2017

Fig. 5. Summer Months Optimum Tilt and Azimuth

Fig. 7. Tracking System Variances

California, there is a larger amount of irradiance– espe-

cially in the winter months. This is where the investment
of a tracking system or mounting manipulations make
for a favorable difference ( both economically and solar
harvesting wise). The main goal for us designers is to
minimize these losses and maximize the amount of solar
in this locale, simultaneously we look to minimize the cost
of unnecessary installations. The exact value of a tracking
system of those mentioned is not absolute because it
would vary on the amount of panels you would need as
well as the desired orientation. What we do know is that
operation and maintenance cost are going to increase
significantly compared to just laying panels out flat on
the desired tilt of the roof.

C. Modeling Limitations
One parameter that was unavailable to model was the
affect of snow losses for the PV system. This may not have
Fig. 6. Global Horizontal Irradiance Averaged for Each Month been much of a concern in the recent years with mild win-
ters. This is not the case for a 25 years evaluation period.
There is bound to be some winters with large storms. Since
ditional tilt when installed but as seen in the Optimization our main concern is the winter supply of energy for the
diagrams the losses do not improve by much, at most the garage This would be an issue to evaluate more in depth
winter months you may gain an additional 2-4 percent. As before installing a large array on the roof. Certain param-
seen in fig. 6 there is a large difference between the poten- eters that were found we believe disrupted the accuracy of
tials of each month. The materials going in to making the our simulations. Specifically, the simulation of distributing
pitch of the roof or adding mounts to the system would our electricity to other parts of the location and reducing
most likely cost more than the additional 4-6 percent of the overall bill of our location. The cost of our modules
irradiance that is gained from changing directional param- and inverters were found through a retail market supplier.
eters. This would mean for larger prices than those you through
Similarly to optimizing direction, the idea of using a track- an installation company like Solar City. This could have af-
ing systems over a fixed one is used to harvest a greater fected the payback period of our project significantly. Also,
amount of the sun’s rays. A tracking azimuth (follows the inflation rate of fuel prices and electricity prices was
the azimuth of the sun across the sky),Seasonal tilt (tilts unknown. Arbitrarily the default inflation value of money
according to the declination of the sun), tracking axis 1 was used, but this does not model the realistic decline of
and 2(tracks north and south or North,South, East,and the fossil fuel reserves. The use of batteries in a simula-
West) systems were all tested in parametric simulations tion was complicated. I am not sure if the simulation was
and the results are shown in figure 7 with runs 1,2,3,4,5 ran properly for modeling them. there are specific that
being Fixed, 1 axis, 2 axis, azimuth, and seasonal respec- are needed for a proper simulation, such as dispatch rates
tively. and exact charge rates. Shading was another limitation,
When you are in a location such as Arizona or Southern described however in it’s own section.
Luke Meyers: SOLAR DESIGN FOR LEED GOLD PROPOSED GARAGE 5

D. Shading

This site has an optimal no shading feature. Fig 1 shows

the pitch of the other garage roofs are sloping towards the
centralized pavement area. The optimal direction and tilt
of the LEED garage is the opposite of the other preexisting
garages. This would put the garage just about due south
with a tilt appropriate for the panels. The location of this
garage will have no shading problems from existing struc-
tures or trees, as long as the pitch of the roof points south
East. as for features on the roof, we were at a disadvan-
tage with not knowing of any vents or high profile objects
on the roof. If there are any after construction they would
have to be evaluated for possible shading on the panels. Fig. 8. Estimated Load for the months of the year, this was done
Shading only hurts the performance of solar panels. The using the calculation in the Loading Estimate Section
roof has a substantial amount of surface area available for
manipulation of solar panel orientation, so minor shadings B. Client Demands
should not be an issue.
As a way to simulate the installation of possible bathroom The client also stressed that the townships seeks to be
vents (bathrooms are in the design plan). We installed a more independent and sustainable. Ferguson Township in-
1.5 meter venter on the roof of the garage. Through Ske- sisted on not selling energy back to the grid to power more
lion, using the Geo-location, we were able to map out the on sight equipment and buildings. This decision lowers the
path of the sun and show the shading on panels that sur- ability for the photovoltaic system to pay itself off sooner,
round the Vent. Through the Minimum Shading analysis, however if the client has the appropriate funds, then it
the shadow of the vent was given a perimeter of shading will be a worthy decision to become more independent.
area through the course of the year. This perimeter as well The client also suggested the need for onsite battery stor-
as the surrounding shading losses are shown in pictures at age to charge their equipment and use for backup power in
the end of the document. The path of the sun moves east case of grid failure. Some of the equipment includes truck
to west over the course of the day as well as an aphelion batteries, power tools, and small electric vehicles. Appro-
pattern through the course of the year– due to the elliptical priately sized batteries for the system are needed to ensure
orbit of earth around the sun coupled with the 23.5 degree all the clients charging needs can be met. Unfortunately,
tilt of the earth. These shading losses were not considered incorporating batteries into the design makes the cost of
in our simulations simply because the usable area of the the project dramatically higher.
roof is much greater than the small area of shading caused C. Loading Estimates
by a 1.5 meter object. This was done to give awareness to
the affect of a small item in a PV system. As mentioned, the new building will be much more en-
ergy efficient than the pre existing garages already onsite.
From previous electric bills obtained from the client, the
VI. Socioeconomic Analysis new buildings electricity consumption was estimated. The
analysis in this report includes a solar energy system sized
A. Carbon Prevention for a normal building like the rest already onsite as well
as a more energy efficient building estimated to use 25
The client is in need of a LEED Gold building with con- percent less energy. The system was sized to meet the en-
siderations of solar energy energy incorporated into the ergy needs of the winter months when the garages use the
building with a possibility of onsite battery storage. The most amount of energy, according to the previous electric-
building is still in the design phase, so the analysis in this ity bills. The loading estimates for a given winter month
report includes several simulations of solar design. Fergu- is 2,200 kWh. Sizing the system to produce 2,200 kWh in
son Township is seeking to reduce their carbon footprint. the winter increases cost dramatically. During the warmer
This is why the township is calling for a sustainable build- summer months, the buildings load is cut in half, but the
ing with renewable energy generation and storage. The solar array produces twice the amount of energy. If a bat-
client asked for a shadow price of carbon calculation to tery is incorporated into the system, the township can ex-
evaluate how much the township will be reducing their port the unused energy into the battery for use during the
carbon footprint. For the large simulation performed, the night. An energy efficient LEED building will cut costs of
solar array will produce about 40,000 kWh over a year. the 2,200 kWh sized solar energy system as the building
At a conversion of 0.55 kg of CO2 emitted per kWh, the load will be reduced.
client will prevent 22,000 kg or 24.3 tons of carbon from Load estimate Calculation for The Billing Period of Jan
going into the atmosphere every year [7]. Being a LEED 20th to Feb 20th:
building, the carbon prevention will be even higher as the
building is built to be more energy efficient. kW hconsumed = 6, 388kW h
6 DESIGN PROJECT, EGEE 437, SPRING 2017

f t2total = 26, 180.327f t2 the final design proposition we decided to model the Cana-
dian Solar Panel CS6U-335 mono with a single Solentria
6, 388kW h kW h
= .244 2 Inverter PVI-36kW TL 480 VAC as well as the Solar World
26, 180.327f t2 ft SW 350 XL mono with the same inverter. These two were
kW h chosen because we are not constrained with space and wish
.244 ∗ 9, 000f t2 = 2, 196.0kW h to have a panel that works best in low irradiance condi-
f t2
tions.
Note: This value calculated was rounded up to make for
an over estimated energy output. Also, since this building A. Design Goal
is LEED is is most likely going to be designed to consume
less energy than a normal maintenance garage. The goal when designing for rooftop solar for the LEED
Gold garage was to maximize the amount of power ex-
D. Incentives tracted during the winter months. The roof top has more
To help lower costs of the solar energy system, differ- than enough room for panels and minimal shading. The
ent federal and state incentives were taken into account. idea of maximizing for the winter and having excess dur-
One of the biggest federal incentives driving renewable en- ing the summer for storage or other on site locations was
ergy projects across the country is the Investment Tax issued.
Credit (ITC). The ITC is a 30 percent tax credit for so-
lar systems on residential or commercial properties [8]. As B. Panel: Canadian Solar CS6U-335W mono
long as the photovoltaic panels are designed and installed To achieve the 36 kW of power needed, 102 panels are
between 2017 and 2019 then the project will be able to being implemented. To meet the 36 kW of power and
utilize the full 30 percent tax credit. The utility com- 480 volts of DC current, the modules are placed with 17
pany providing power to Ferguson Township, West Penn panels in series and 6 rows of parallel series of panels.
Power participates in many state wide incentives for re-
newable projects as well. One of the utility companies
Note: Voltage increases when panels are placed in series
incentives is a 25,000 dollar grant program for non profit
and current increases when panels are placed in parallel.
entities (https://energy.gov/savings/west-penn-power-sef-
Voltage stays the same in parallel and current stays the
grant-program). West Penn Power also offers net metering
same in series. The limiting factor to the orientation of
for electricity sales back to the grid, however the grant
the solar circuitry is the inverter. It has a maximum DC
program and the net metering incentive cannot be used
voltage so the panels have to be placed accordingly to the
together. The client also prefers to not use net metering
limit the inverter can handle.
to stay independent from the grid. Through simulations,
the 25,000 dollar grant was deemed to be more financially
worth it to utilize. The cost of the Canadian solar panel system is 23,919
dollars at year of installation plus the cost of installation
E. Electricity Rate Structure itself would end in a net price of about 95,000 dollars.
With certain incentives and grants the price I simulated
Ferguson Township municipal buildings include offices
was 74,527 dollars. The surface area of the roof allows for
and garages. Using previous electric bills, it is calculated
freedom in placement of panels. As long as the circuits
that the township pays about 8.3 cents per kWh on all the
configuration match that of the inverter’s limits.
buildings and garages. This estimation includes the elec-
tricity rate of 6.07 cents per kWh plus extra costs including
C. Panel: SW 350 XL mono
the inflation rate of fuel of 2.5 percent. The inflation rate
of fuel will make the average price the township pays for A similar kW output was desired and because of this the
electricity to increase, however the new onsite electricity same amount, 102 panels placed in same 17 series and 6
generation from the solar system will save the township parallel, was needed. This is due to the similar range of
from increased electricity prices over the years. max power voltages, Canadian solar has one of 37.8V and
Solar world has a Vmp of 38.4 V. Similar to the Canadian
VII. Design Criteria Solar simulation the 36 kW inverter was used. The price
The Size, cost, and production of kWh are all results for the SW panels are a little higher giving a net cost at
that were dependent upon many factors. The panel, the year of installation of 31,773 dollars with a total installa-
inverter, and the desired wattage can all be changed by tion cost of 86,301 dollars. The Solar World panels had
a small amount and the entire system is changed. Three a better review than Canadian Solar according to the lit-
Solar World panels were tested each of a different material erature review section, but a simple analysis was done to
and watt rating as well as two Solectria inverters.These show which panel to model with batteries.
technologies were simulated according to the lowest cost. This break down gave that the amount of power pro-
The appropriate combination of the two were used to re- duced from both simulations compared to the reduced
duce Capital cost for the system. The final outputs were price from grants and incentives; resulting in the Cana-
summarized by S.A.M.(Systems Advisory Model) and as dian Solar being the best choice to do a modeling scenario.
Luke Meyers: SOLAR DESIGN FOR LEED GOLD PROPOSED GARAGE 7

CS6U-355 Mono Solectria PVI-36kW TL 480V

Panels 102 Vmmpmin (min. Power Point) 520 V
Dollars/WDC .70/W Vmmpmax (max. Power Point) 800 V
System Price per Unit Power 1.795/kWh Vdcmax (max DC voltage) 36,715.2 Wdc
TABLE I PInverter (max AC power) 36,000 Wac
Canadian Solar Panel Breakdown Fdc−ac (Ac/Dc ratio) 1.20
Ninverters 1
TABLE IV
Known of Desired Inverter
SW 350 XL- mono
Panels 102
Dollars/WDC .89/W
System Price per Unit Power 1.952/kWh 350V
Nseries = 37.8V =9
TABLE II 9∗46.1V ≤ 5, 338.07W so, Nparallel = 33,193W
9∗335W = 11.01 = 11
Solar World Panel Breakdown 9∗11∗335W
Ninverter = 1.20∗5,200W AC
= 5.3 = 5inverters
Note: the ideal ratio is 1.2 for DC to AC current, but for
this case it is 1.28.
Case 2: Solectria PVI- 36kW TL 480 V
D. Inverters
Vmid = 520V +800V
2 = 660V
This algorithm will leave the decision of inverter choice 660V
Nseries = 37.8V = 17.46 = 17
up to the designer. As long as the information on the in- 34,199W
17 ∗ 46.1V ≤ 36, 715.2W so, Nparallel = 17∗335W = 6.005 =
verter and panel of choice is known then this calculation 6
will give the appropriate circuit orientation of the panels. 6∗17∗335W
Ninverter = 1.20∗36,000W = .791 = 1inverters
The Superscript P is for the module and I is for the In- AC

verter. Arbitrarily we chose to look at two Solectria Inverters.

I I
I
Vmppmin + Vmppmax
Vmid = One with a low Voltage at 5200 WattsDC and one with a
2 higher voltage of 36 kWDC . The two were modeled with
I
P Vmid each of the panels tested. The unit price for a Solentria
Nseries = P
Vmp PVI 36 kW TL 480 V inverter is 4,411.76 dollars/unit and
the price for a Solentria PVI-5200 TL 208 V inverter was
if:
P P I P P 1,640 dollars/unit. When the load required is in need of
Nseries ∗ Voc ≥ Vdcmax , Nseries = NSeries −1
multiple inverters vs. having one with a greater voltage
if not then: and current capacity then after a certain point, it is more
P economically feasible to purchase one inverter at a higher
Nseries ∗ NPParallel ∗ Pmodule unit cost; rather than purchase multiple smaller inverters.
Ninverter =
Fdc−ac ∗ Pinverter In this specific case the 36 kW of 102 panels required one
P 36 kW inverter or four 5200 W inverters. The better choice
Parray
NPParallel = P
economically was to choose the single inverter.
NSeries ∗ Pmodule
Example Inverter Checks: E. Battery
Calculations: Referring to the Canadian Solar Panel and the 36 kW
Case 1:5200 W Inverter inverter with 102 panels.During the summer months there
is an excess of energy produced compared to the amount
200V +500V
Vmid = 2 = 350V demanded. Roughly 1,650 kWh to 1,850 kWh of extra
power is available during the months of July and June
respectively (estimates made from the Demand and pro-
Solectria PVI-5200TL 208V duction figures). A month has 30 days, so 60 to 33 kWh
Vmmpmin (min. Power Point) 200 V for each of those months of extra power per day. This is
Vmmpmax (max. Power Point) 500 V the value to which we are going to size the battery bank.
Vdcmax (max DC voltage) 5,338.07 Wdc Using 60 kWh, the Lithium ion: Nickel Cobalt Aluminum
PInverter (max AC power) 5,200 Wac Oxide battery was chosen. Compared to other batteries
Fdc−ac (Ac/Dc ratio) 1.20 this material required only three in a string. The Lithium
Ninverters 5 Ion was chosen due to the detailed literature review sec-
tion. The life time of the battery is much longer than
TABLE III the other Lead based batteries, and considering the LEED
Known of Desired Inverter project the Lithium Ion is less likely to pollute and is more
sustainable.
8 DESIGN PROJECT, EGEE 437, SPRING 2017

• Modules: CS6U-335 W- 102 panels, Area- 196.6 m2 ,

cost- 23,939.42
• Array: Series-17 panels, Parallel- 6 Sets
• Inverter: PVI-36kW 408V- 1 inverter, Cost- 4,411.76
• Lifetime Evaluation: 25 Years
• Orientation: Azimuth- 144.71o East of south, Tilt-
14.04
• Tracking System: None, Fixed to Plane of Roof
• Battery: Lithium ion: Nickel Cobalt Aluminum
Oxide, 400 V, 60 kWh Storage, 3 on site, Cost-
11,088.00/battery
• Financial: Mortgage, 25 years
• Incentives: West Penn power Sustainable programs-
25,000 Grant, West Penn loan Program, Investment
Fig. 9. Mono vs. Poly power production: Arizona
Tax Credit- 30 percent Electric Buy rate: .0838/kWh
(Averaged with all bill charges) Load: (See Socioeco-
nomics Section) LEED garage only w/ battery stor-
age.
The next biggest challenge was getting the project to
pay off. Since the electricity is not being sold back to the
grid then we had to find other ways to bring the payback
period down. Coupling the battery bank with the large PV
installment we are looking at a total of 108,656 dollars after
the West Penn Power Grant. Over the 25 years this project
will not pay off. The battery bank alone cost more than
the panels and the inverters combined. The economics ran
were as follows:
• Buying Price: .0838 dollars/kWh
• Savings Price: .0838 dollars/kWh
Fig. 10. Mono vs. Poly power production: State College, Pa. • Grant: 25,000 dollars year zero
• ITC: 30 percent off installation cost
• Buying Price Inflation: 2.5 percent
F. Poly. vs Mono. - 255W Panels
Limitation: In the simulation we accumulated the kWh
As seen in figure 8 and 9, run 1 and 2 being Mono and produced each month over the course of the year. At
Poly-silicon crystals respectively, there is no visible dif- the end of the year the amount left over is credited as
ference in the amount of irradiation collected. Different ’selling’ back, even though realistically the electricity
locations with a noticeable difference in irradiance were would just be used elsewhere on site. So, we modeled this
simulated simultaneously. There is however a difference in by selling the left over kWh at the end of each year as
the price per Watt. The Poly-silicon panels range around .0838 dollars/kWh, Figure 10 shows the amount of kWh
90 cents/W and mono-silicon ranges at 1.10 dollars/W. De- left over at the end of the year that is able to be used
pending on the brands of panels and the desired wattage we elsewhere in our site location. The problem with this is
would suggest picking a poly-silicon based, however there every year the price of electricity will change(most likely
is better economic parameters and higher energy output going up with fossil fuel constraints). This aspect we
with higher watt rated PV panels. This type of panel is could not model.The payback period would most likely
what we decided to model as our final simulation output. be quicker considering by year 25 the savings price of
electricity could be well above the 8.83 cents that was
VIII. Model Description modeled.
A. Economic Evaluation Since the price for a battery bank was so high We ran
a simulation with no battery bank. The results were as
After running parametric simulations with different PV
follows:
modules, inverters, tracking systems, and batteries. We
decided to simulate the Canadian Solar CS6U-335 W
Panel, with the Solectria 36 kW 480 V Inverter, the fixed • Payback Period: 20.1 years
mounting system, and the installation of Lithium Ion bat- • Net Present Value: 185 dollars
tery banks. The payback period as well as the alterna- • Net Capital Cost: 74,527 dollars
tive economic simulations were examined to suggest the Limitations: Similar to as described before, the ’saving’
best system to install for this particular site. First a brief price is not inflated with time. Payback period is most
overview of the parameters being ran. likely a lot lower.
Luke Meyers: SOLAR DESIGN FOR LEED GOLD PROPOSED GARAGE 9

Fig. 13. Area Comparison of the Panels to Surface area Available,

Shading location and Shading area is outlined to show locations of
losses

available space. There is more than enough room to orient

the panels so they are not obstructed by shading and also
are in a location that makes for easy battery or power line
hook up. Figure 12 shows the l02 panels on the proposed
building with an area of shading that was evaluated.
Fig. 11. Monthly Load and Energy Production with Yearly kWh
excess viable for diverting to other on site locations
IX. Conclusions and Recommendations
In comparing the simulations that we ran, here are the
recommendations that we suggest. First, the tilt of the
roof is not proposed yet. If the client wishes to maximize
the winter irradiance collected then you should assign a
tilt to the roof over the realistic 14.04 degrees. Anything
over this would improve the efficiency by about 2-6 per-
cent. Eventually the tilt of the roof would make for difficult
maintenance and installation. The proposed roof of 14.04
degrees still gives a decent amount of winter irradiance and
during the summer the 14.04 degrees would give for 99 per-
cent of the available solar resource, thus maximizing the
amount to be diverted to either the battery bank or the
highly suggested of the other on-site locations. Second, if
Fig. 12. Evening Hours shading during the month of June, casts a you would decide to proceed with this project and wish
shadow on the lower panels over the solar azimuth sun path to change from our suggested Canadian Solar panel, the
choice between a poly and mono silicon based crystal ma-
terial would be the options.From our results we found that
B. Visual Models
the irradiance captured from the poly and mono panels
Now that the design of ours is deemed economically fea- were negligibly different.The difference in price would lead
sible we used the skelion program to model the look of us to suggest the ply based panels. Third, even though no
the 102 panels on the total surface area of our roof. Si- shading was needed to be done, the installment of bath-
multaneously, we ran a simulation of features on the roof rooms in the garage could potentially mean putting vents
that caused shading. We were limited to exact locations, or fans on the roof top. Due to the skelion shading analysis
heights, and other features of high profile objects on our an arbitrary height of an object of 1.5 meters was modeled
roof ( vents, HVAC units, etc). However, panels that are to cast a 5 meter diameter around the object. In this area
within a distance of 5.3 meters from a 1.5 meter object are any panels would experience some type of shading losses.
subjected to partial shading at some time of the day. These As a suggestion, if any objects are on the roof, since there is
percentages of shading over the course of the year are not an abundance of space available, refrain from putting them
relevant in the course of the entire year. Some objects cast around such objects to avoid unnecessary losses. Fourth,
larger shadows for longer periods of the day and at that for our client we recommend that they do not include a
point the panels would be better to be moved elsewhere. battery in their system. The price increase from a battery
Even if some minor causes of shading were to occur in system alone would cause the payback period to well over
some sections of the roof, there is such a vast surface area 25 years rather than our current 20.1 year payback period
available for installation that we could work around this and would not be worth it for the client unless required
in some way. When 102 panels is mentioned, it seems like for charging stations or something that requires a large
the area of the roof would be constrained.However, total- amount of energy. (Reference model description for exact
ing an area of 196.6 m2 compared to the 862.845 m2 of battery costs.) Instead, the excess energy can be utilized
10 DESIGN PROJECT, EGEE 437, SPRING 2017

in other buildings on the property such as in the offices

for air conditioning during the summer. The only reason a
battery bank would be suggested, is if the bank is coupled
to an electric charging station. Having a charging station
installed would not only decrease fuel costs further but also
increase the LEED rating significantly. In a final conclu-
sion this site was assessed and the output of the evaluated
parameters were not as favorable as we hoped. Consid-
ering the wishes of the client to become more sustainable
and reduce their carbon footprint then I would suggest the
installation of PV on the roof top of the proposed LEED
Gold maintenance garage.

References
[1] Pros and Cons of Monocrystalline vs. Polycrystalline Solar Pan-
els.” Solar Reviews — Consumer Reviews of Solar Companies,
Solar Panels, Solar Installers and Solar Contractors. N.p., 07 Apr.
2016. Web. 01 May 2017. http://www.solarreviews.com/solar-
energy/pros-and-cons-of-monocrystalline-vs-polycrystalline-
solar-panels/
[2] Choosing the best solar panels - cost is only one fac-
tor.” Energy Matters. N.p., n.d. Web. 01 May 2017.
http://www.energymatters.com.au/panels-modules/choosing-
solar-panels/
[3] Kays, Elizabeth. ”Guide to Solar Components: Choosing the
Right Inverter.” Slingshot Power. N.p., 29 July 2015. Web.
01 May 2017 http://www.slingshotpower.com/guide-to-solar-
components-choosing-the-right-inverter/
[4] Bean, Robert. ”Lead-acid vs Lithium Ion batteries .” LinkedIn, 2
Feb. 2016. Web. 1 May 2017. Robert Bean - CEO @ R and H Con-
sulting ¡https://www.linkedin.com/pulse/lead-acid-vs-lithium-
ion-batteries-robert-bean¿.
[5] Best Solar Panel Reviews of 2017.” Reviews.com. N.p., 10 Apr.
2017. Web. 01 May 2017. http://www.reviews.com/solar-panels/
[6] Inc, Canadian Solar. ”The Canadian Solar Difference.” Make
The Difference — Canadian Solar. N.p., n.d. Web. 01 May 2017.
http://www.canadiansolar.com/about.html
[7] How Much Carbon Does the Average Solar System Save”
Solar Market. N.p., 15 Dec. 2015. Web. 28 Apr. 2017.
¡https://www.solarmarket.com.au/carbon-reduction-solar-
panels¿.
[8] Solar Investment Tax Credit (ITC).” SEIA. N.p., n.d. Web.
29 Apr. 2017. ¡http://www.seia.org/policy/finance-tax/solar-
investment-tax-credit¿.
[9] Database of State Incentives for Renewables and Ef-
ficiency.” DSIRE. N.p., n.d. Web. 03 May 2017.
¡http://www.dsireusa.org/¿.
[10] SolarWorld.” SolarWorld — CivicSolar. N.p., n.d. Web. 03 May
2017. ¡https://www.civicsolar.com/manufacturers/solarworld¿.