Technology and Economics of High-Efficiency c-Si PV · 2016-04-06 · c-Si PV industry back on cost...

Technology and Economics of High-Efficiency c-Si PV

Doug Rose, SunPower Corp.

Presented at Silicon Valley PV Society meeting, 9/9/09

© 2009 SunPower Corp.

Safe Harbor StatementThis presentation contains forward-looking statements within the meaning of the Private Securities

Litigation Reform Act of 1995. Forward-looking statements are statements that do not represent historical

facts and may be based on underlying assumptions. The company uses words and phrases such as

"expects," “believes,” “plans,” “anticipates,” "continue," "growing," "will," to identify forward-looking

statements in this presentation, including forward-looking statements regarding: (a) our plans and

expectations regarding our cost reduction roadmap, (b) cell manufacturing ramp plan, (c) financial

forecasts, (d) future government award funding, (e) future solar and traditional electricity rates, and (f)

future percentage allocation of SunPower solar panels within our systems business. Such forward-looking

statements are based on information available to the company as of the date of this release and involve a

number of risks and uncertainties, some beyond the company's control, that could cause actual results to

differ materially from those anticipated by these forward-looking statements, including risks and

uncertainties such as: (i) the company's ability to obtain and maintain an adequate supply of raw materials

and components, as well as the price it pays for such; (ii) general business and economic conditions,

including seasonality of the industry; (iii) growth trends in the solar power industry; (iv) the continuation of

governmental and related economic incentives promoting the use of solar power; (v) the improved

availability of third-party financing arrangements for the company's customers; (vi) construction difficulties

or potential delays, including permitting and transmission access and upgrades; (vii) the company's ability

to ramp new production lines and realize expected manufacturing efficiencies; (viii) manufacturing

difficulties that could arise; (ix) the success of the company's ongoing research and development efforts to

compete with other companies and competing technologies; and (x) other risks described in the

company's Annual Report on Form 10-K for the year ended December 28, 2008, and other filings with the

Securities and Exchange Commission. These forward-looking statements should not be relied upon as

representing the company's views as of any subsequent date, and the company is under no obligation to,

and expressly disclaims any responsibility to, update or alter its forward-looking statements, whether as a

result of new information, future events or otherwise.


2

SunPower

Over 550 systems on 4 continents

Over 600 dealers and growing rapidly

Over 85 patents and 25 years of R&D

2008 Revenue of $1.4 bil

3

Commercial Power Plants

Largest commercial install

base in North America

Largest solar power plants

in North America

Residential

Largest residential install base

in North America

5,000 Employees; 100% solar

Over 400 MW/yr production rate


4

Roof Integrated Systems

PowerGuard®

T10 Roof Tile

SunPower Trackers

T0 Tracker

SunPower Product Families

Fixed Tilt Systems

SunTile®

T20 TrackerT5 Roof Tile

225 W 230 W 315 W

Panels

> 22% Efficiency SunPower Solar Cell


Highest efficiency mass-produced

cells and modules in the world

Patented all-back-

contact cell

Global PV Market

0

500

1000

1500

2000

2500

3000

Germany Spain Japan USA RO Europe RO World

2007

2008

5

= 2.82 GW

= 5.95 GW

Source 2007 data: Solarbuzz, 2008

Source 2008 data: Solarbuzz, 2009


Global Annual PV Market Outlook


6

(MW)

Source: EPIA, 2009

0.01

0.1

1

10

100

1000

1965 1970 1975 1980 1985 1990 1995 2000 2005 2010

$2

00

2B

Year

Semic.

PV

Historical Semiconductor and PV Module Annual Sales

7

PV uses more

silicon than the

IC industry


8

Area comparison of PV to Semiconductor now

Semiconductor 2008: 5.2 km2 of chips

Photovoltaics 2008: 42.5 km2 of PV

http://www.semi.org/en/MarketInfo/SiliconShipmentStatistics/index.htm

Source: SEMI.org

mil in2 km2

Q1 2163 1.4

Q2 2303 1.5Q3 2243 1.4Q4 1428 0.9

2008 8137 5.2

Semiconductor details

PV details

Assumed average of 14% efficiency

Assumed Solarbuzz 2008 value of 5.95GW

(is market #, production was higher; other

market estimates as high as 7GW)

ICs

PV

...


c-Si PV industry back on cost learning-curve

9

By most estimates, PV LCOE without incentives using

these modules will be lower than peaking natural gas

LCOE. Best-in class thin film and high efficiency PV

power plants will give even lower LCOE.Last 4 data points are forecast by the Prometheus Institute

Here, PV is competing

for baseload (in the 7 to

10 years it takes to build a

nuclear plant, PV installed

over that same time frame

will be lower cost)

1

10

100

1 10 100 1000 10000 100000

Mo

du

le A

SP

(2

00

8$)

Cumulative Production (MW)

1979

$33/W

2008

$3.17/W

Silicon

Shortage

81% Progress

Ratio

2012

$1.40/W

Large decreases in

balance of system

cost during this time

period

Mo

du

le A

vera

ge S

ale

s P

rice (

AS

P),

2008$


10

Solar PV Power Plants Are Cost Competitive

0

0 50 75 100 150 200 250 300 350 400

Levelized Cost ($/MWh)

Renewables

Conventional

$87 - 196

$129 - 206

$57 - 113

$216- 334

$69-96

Gas Peaking

Gas Combined

Cycle

Wind

Solar Thermal

Solar PV

LCOE by Resource $/MWh: 2009 - 2012

Prices include 30% federal incentive

Source: Lazard Capital Markets 3/18/2009


Source: CEC PIER-funded study by GE Energy, July 2006 (Temporal Pattern: July 2003 Average Day)

Resource Generation Profiles

Source: Hal LaFlash, PG&E

11


12

2050 View 450ppm / 80% CO2 Reductions by 2050

PV Moderate Growth Case

PV and other other renewables and energy efficiency can easily reach the goal.

2040: What is needed from PV: 2000 TWH/yr

What is possible from PV: 5000 TWH/yr (Moderate Growth case)

2000 TWH/yr

5000 TWH/yr

TW

H/y

r

DPV: Distributed PV

CPV: Central PV

Sources: McKenzie Report, 2007 for starting points and energy efficiency; AWEA for wind; internal SunPower calculations for DPV, CPV, CSP


Diversity is needed for maximum success of PV

Energy market > 1 trillion $ / yr

Diverse applications require diverse product characteristics

Competition drives cost reduction

Multiple approaches will give greater total volume (reduced impacts

from material limits, exponential scaling limits, etc.)

13


OVERVIEWEconomics of High Efficiency PV

15

Spend a little more in cell processing…

Wafer Solar Cell Solar Panel SystemIngotPolysilicon

…to deliver savings across the value chain

How can high efficiency cells be cost effective?

more W/g more W/$ more W/m2

High Efficiency and the Value Chain

15


more W/$

16

Wafer Solar Cell Solar Panel SystemIngotPolysilicon

Value Chain for conventional Si:

* Value chain distribution percentages are for new Centrotherm 347MW turnkey plant as reported in Dec. 2008 Photon International, with 30%

GM added to all steps and system costs of $1.65/W (including margin), using average of U.S. and China locations and $1.32/euro exchange rate.

10% 6% 7% 12% 43%21%

Rough percentages for conventional c-Si*:

With high

efficiency:

Lower $/g Si Lower $/W module conversion and installation

Invest here to get saving across the whole value chain


Example: Lower area-related costs

Reduce materials costs

– Less module area: Glass, silicon,

encapsulant, frames

– Less system materials: Wiring,

mounting

17


Reduce shipping costs (even

with high density shipping of full system)

305 Wp per tile

22 tiles per pallet

14 pallets per truck

93-kWp per truck

Reduce installation costs(even with an easy to install, non-

penetrating mount)

High efficiency can:

Levelized cost of energy: An LCOE Equation

18

Initial investment Area related costs

Grid interconnection costs

Project related costs

Is the present value of the

benefit over the financed

life of the project asset.

Annual Costs Annual system operating and

maintenance costs ( inverter

maintenance, panel cleaning,

monitoring..)

Depreciation Tax/

other Public BenefitSystem Residual Value Present value of the end of life

asset value is deducted from

the total life cycle cost in the

LCOE calculation.

System energy production

First year energy generation

(kWh/kwp) then degrading output

over the system life based on an

annual performance degradation

rate

n = the system’s financing term

(which will determine the

duration of cash flows)


For additional information on LCOE see “Minimizing utility-scale PV power plant levelized cost of energy

using high capacity factor configurations” by Matt Campbell, SunPower Corp. (in the fourth print edition of

Photovoltaics International Journal, and available on SunPower web site).

The LCOE Sensitivity to Input Variables


19

A PV Power Plant with the same installed price and first year performance

can yield LCOE values of a tremendous range

Case 1 Case 2 Case 3

System Price 100% 100% 100%

kWh/kWp 100% 100% 100%

Annual Degradation 1.0% 0.5% 0.3%

System Life 15 25 40

Annual O&M $/kWh 0.030$ 0.010$ 0.005$

Discount Rate 9% 7% 5%

LCOE $/kWh 0.23$ 0.13$ 0.09$

Same installed Price($/Wp) but different LCOE ($/kWh)

LCOE: System Cost vs. Capacity Factor


20

20%

22%

24%

26%

28%

30%

32%

34%

36%

38%

40%

$2.00 $2.50 $3.00 $3.50 $4.00

Capacity Factor

System Price - $ / Wp

Sample Range of Equivalent LCOE Values

LCOE Equivalence

$3.41/W at 33% CF is equivalent to $2.50/W at 24.2% CF

21

Higher Capacity Factor

Mid-to-high-efficiency modules

enable cost-effective tracking

Captures up to 30% more

sunlight than fixed tilt systems

Tracking has higher capacity

factor (so reduces the $/kwh costs

of the electrical BOS costs)

Better matching of energy

production with summer load

(time of day and total)

0%

20%

40%

60%

80%

100%

120%

12

:00

:00

AM

1:0

0:0

0 A

M

2:0

0:0

0 A

M

3:0

0:0

0 A

M

4:0

0:0

0 A

M

5:0

0:0

0 A

M

6:0

0:0

0 A

M

7:0

0:0

0 A

M

8:0

0:0

0 A

M

9:0

0:0

0 A

M

10

:00

:00

AM

11

:00

:00

AM

12

:00

:00

PM

1:0

0:0

0 P

M

2:0

0:0

0 P

M

3:0

0:0

0 P

M

4:0

0:0

0 P

M

5:0

0:0

0 P

M

6:0

0:0

0 P

M

7:0

0:0

0 P

M

8:0

0:0

0 P

M

9:0

0:0

0 P

M

10

:00

:00

PM

11

:00

:00

PM

Fixed Tilt CF - Mojave

T0 Tracker - Mojave

Cal ISO Load 7/15/08

Peak CA Summer

Load

PV Power Plant Output vs.

Summer Utility Demand

Curve

Load

T0 tracker

Fixed tilt


0%

20%

40%

60%

80%

100%

120%

12

:00

:00

AM

1:0

0:0

0 A

M

2:0

0:0

0 A

M

3:0

0:0

0 A

M

4:0

0:0

0 A

M

5:0

0:0

0 A

M

6:0

0:0

0 A

M

7:0

0:0

0 A

M

8:0

0:0

0 A

M

9:0

0:0

0 A

M

10

:00

:00

AM

11

:00

:00

AM

12

:00

:00

PM

1:0

0:0

0 P

M

2:0

0:0

0 P

M

3:0

0:0

0 P

M

4:0

0:0

0 P

M

5:0

0:0

0 P

M

6:0

0:0

0 P

M

7:0

0:0

0 P

M

8:0

0:0

0 P

M

9:0

0:0

0 P

M

10

:00

:00

PM

11

:00

:00

PM

Fixed Tilt CF - Mojave

T0 Tracker - Mojave

Cal ISO Load 7/15/08

Peak CA Summer

Load

20.0%

22.0%

24.0%

26.0%

28.0%

30.0%

32.0%

34.0%

36.0%

38.0%

40.0%

Fixed Tilt T20 Tracker T0 Tracker

Annual CF

Summer CF

Summertime capacity factor with SunPower modules

and T0 tracker in Las Vegas is ~39%

Major contributors to LCOE:

Capital cost

Module $/W

Area-related BOS

Electrical BOS

Project-related $

– Efficiency & GCR

– Watts / project


22

Capacity Factor

Environmental conditions

– Total solar radiation

– Diffuse/direct & spectrum

– Ambient temperature,

wind, soiling conditions

Mounting

– Tracking vs. fixed

– Tilt angle

– GCR and shading

Performance

– Response to solar

radiation conditions

– Operating temp & temp

coef; Soiling

– Degradation rate

– System availability

Cost of capital

Perception of risk

Average financing

conditions

O & M

Reliability x number

of components

Operation (cleaning, …)Example:

For area-constrained

applications, high efficiency

allows more Watts for the

project (which has value to the

customer and further reduces

$/W by amortizing fixed costs,

like sales and permitting costs,

across more watts)

23

Simplified LCOE

Panel + Balance of Plant + O&M Costs

Sunlight Collection x Conversion Efficiency≈LCOE

Balance of Plant

High efficiency reduces materials (modules,

wiring, mounting, etc), installation, and

shipping costs

Low cost 1-axis tracking where applicable

Integrated value chain engineering

Lower $/W fixed costs for area-constrained

O&M Costs

High efficiency reduces O&M costs (e.g.

less to clean and upkeep)

Grouped trackers reduce number of

motor / controls

Experience & Closed loop learning

Sunlight collection

1-axis tracker adds up to 30% energy

(this reduces the $/kwh costs of modules,

inverters, wiring, monitoring, etc. by up to

23%)

Conversion efficiency

Higher panel power density

Superior performance in high

temperature, low light, and range of

spectral conditions

Low degradation / year


(eliminated discount rate, converted O&M to capital expense,…)

Some of the levers used by SunPower (plus experience and reliability to decrease financing costs):

Use of LCOE

Do NOT compare calculated LCOE to current average electricity prices!

LCOE should only be used to compare to the LCOE of alternatives, preferably after

correcting for differences in the value of the energy, value of assured price, and impact on

the country

– Why only compare PV LCOE to LCOE of alternatives instead of current electric rates?

Traditional fuel prices are expected to increase from general inflation, scarcity, and internalization of

environmental impact. General inflation will also increase non-fuel operating costs. This impacts the LCOE of

traditional fuel energy sources, but does not affect current electric rates.

LCOE discounts the value of future energy generation (so a high expected inflation rate will increase the

discount rate, thus increasing the LCOE). But, with high inflation, you would want to put in a hard asset like PV

with very little costs other than the up-front costs.

– Why correct for value of the energy?

Energy value at peak times is significantly higher than at off times (and PV matches the load curve well)

If possible, should also correct for other value/impact to the customer (e.g., carport shade, assured power)

– Why correct for value of assurance of the future cost of energy?

LCOE of PV reflects a firm guarantee at a known cost (the hard asset of the PV array backs that guarantee to

the first order, with the risk premiums included in the financing providing a backstop should the array

underperform). In contrast, current conventional energy prices include no guarantee of future prices.

Can correct for this difference by adding the cost of a guarantee on the alternatives prices over the same

contract length, backed by multiple layers of insurance.

– Why correct for value of impact on the country?

The differences in non-internalized environmental impact, job creation, and the value of accelerated cost

reduction of PV on future costs can be significant.

If you include subsidies for one technology, include them for the others (e.g., insurance for nuclear industry).

24


SunPower Central Station Competes with MPR

$0.10

$0.12

$0.14

$0.16

$0.18

$0.20

2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020

25

Source: 2008 CA MPR, 25 year contract: CPUC Resolution E-4214 December 18, 2008

1) Time of Delivery Multipliers vary by utility

2008 Market Price Referent by Contract Start Year ($/kWh)


CELLS: TECHNOLOGY

Basic solar cell operation

V

I

V

I

Diode Solar cell in light

Sunlight creates current in the

opposite direction as the

applied voltage.

Isc

Rsh

Rs

Simple equivalent circuit

27


SunPower vs. Conventional c-Si Cell

28

Lightly doped front diffusion

• Reduces recombination loss Texture + ARC

Backside Mirror

• Reduces back light

absorption & causes

light trapping

Localized Contacts

• Reduces contact

recombination loss

Backside Gridlines

• Eliminates shadowing

• High-coverage metal

reduces resistance loss

.

.

Passivating

SiO2 layer

• Reduces surface

recombination loss

FRONT

BACK

Silver Paste PadAluminum

paste

Texture + ARC GridlinesN-Type

diffused

junction

.

.


SunPower cell efficiency

29

2005 2006 2007 2008

Ce

ll E

ffic

ien

cy (

%)

20042003

20%

21%

22%

23%

24%

25%

20.6%

22.0%

23.4%

20.6%

22.4%

Laboratory Prototyping Results

Production median (Gen 1 & 2)

21.3%

29

Gen 3

Median production cell efficiency > 22.4%


Note Gen 2 distribution is

tighter than Gen 1 distribution

30

20.6

22.4

0.9

0.3

5.2

4.4

1.2 0.7

1.0 1.1

18

20

22

24

26

28

30

Gen 2 Gen 3

Effic

iency (

%)

Analysis of losses in Gen 1&2 Technologies

- -+P-MetalN-Metal

N-Diffusion P-Diffusion

N-Type bulk

FSF

ARC

Simulation of loss breakdowns in

Gen 1 and Gen 2 solar cells

Gen 2 Improvements

- thinner 190 --> 165 microns

- improved processes

- tighter pitch; smaller feature sizes

- improved edge and pad design

Gen 1 Gen 2

optical

resistive

recomb


CELLS: COST

Less silicon

Higher efficiency proportionately lowers silicon usage – (i.e., 20% efficiency cell uses half the silicon as a 10% cell).

SunPower’s back-contact architecture allows thinner wafers with no

loss in efficiency

32

Conversion

Efficiency %

Cell Thickness (microns)

Cost reduction

SunPower solar cell efficiency improves as wafer thickness decreases versus

conventional solar cells which become less efficient on thinner wafers.

Some changes to

raise efficiency, for

example a BSF, can

reduce the loss from

thickness reduction

in conventional cells


Novel wafering

(i.e., taking full advantage of ability high-efficiency cells to

work well with thin wafers)

Cleaved wafers (e.g., SiGen)

– From SiGen.net:

Thickness: “20µm – 150µm”

“Superior mechanical strength: 10X stronger”

Jet or laser sawing (e.g., Fraunhofer)

Others (electrical discharge, thermal expansion delta, etc.)

33

Kerf-free 50 μm c-Si wafer


Other cost reduction opportunities for high effic. cells

Capital Cost Reduction

– Equipment for high efficiency cell manufacture mostly at first or second

generation. Plenty of opportunity for:

Value engineering; component standardization

Tool availability improvement

Throughput optimization and line balancing

Improved automation (better yield, lower labor content)

Consumables Reduction

– Process optimization

– Water and power conservation

Process simplification

– Step elimination

Increased scale

– Plant size

34

(in addition to efficiency increase, thickness reduction, and lower ingot and wafering costs)


Phase 1

ENERGY / RATED WATT

36

The cell features which make SunPower cells high efficiency also lead to more energy per rated watt

Superior temperature performance

– Power coefficient of -0.38%/C vs. -0.5%/C for conventional silicon provides up to

~4% energy/W advantage (from high Voc vs. Eg)

– Also, runs cooler (from higher efficiency and better IR reflectance)

Superior light capture

– Retains performance at high-angle illumination (from front texture)

– Single-junction with wide spectral response (nearly 100% spectral response from

0.4-1.1µm)

Better low-light performance (from high shunt resistance)

No light-induced degradation

– Avoid immediate 2–3% degradation after first exposure to light (n-type wafer, so

no B-O defect complex)


More kWh/W – Temperature Coefficient

37


High efficiency cells have

lower temperature coefficient

Conventional panels have Tc

of approximately -0.5%/ C

High-efficiency panels can

have Tc as low as -0.35%/ C

Provides up to 4% kWh/W

advantage in real world

conditions

Nuremberg,

Germany

Phoenix,

AZ

SunPower,

-.38%/C dP/dT

0.990 0.923

Conventional silicon,

-.50%/C dP/dT

0.986 0.891

SunPower advantage 0.40% 3.60%

More kWh/W – Spectral response

38

Panels are rated for a defined spectrum, but spectrum of light changes through the day.

High efficiency cells have wider spectral response, hence better energy performance in real

world conditions (which include morning, evening, clouds, and high-noon)

Multi-junction 2-terminal cells are limited by the lowest current in the cell stack, so they can

perform poorly at spectral conditions other than standard


Wavelength, microns

Re

lative

Exte

rna

l Q

uan

tum

Eff

icie

ncy, %

0

20

40

60

80

100

AM 1.5 global spectrum0

1

2

3

4

5

Nu

mb

er

of S

unlig

ht P

hoto

ns (

m-2

s-1

mic

ron

-1)

E+

19

0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2

SunPower cell Conventional cell

More kWh/W – Low light performance

Low light conditions occur

in the morning, afternoon &

cloudy weather conditions

Loss at low light is

dominated by loss

mechanisms at low

injection (e.g. shunt

resistance)

High-efficiency solar cells

have better low light

performance, which

improves energy

production / rated watt

39


SunPower SPR-90

Mono Crystalline Silicon

Poly Crystalline Silicon

CuInSe2

Amorphous Silicon

Sharper knee of curve at low light levels

indicates better low light performance

More kWh/W - No Light Induced Degradation (LID)

40

Light induced degradation is caused by an interaction between boron doped silicon (used in p-

type solar cells) and oxygen. This effect has been documented by many research institutes.

* Excerpt from Photon International article, “A Call for Quality”, March, 2008


Third Party Validation- Best Energy Performance


41

Independent tests show that SunPower® Solar Panels deliver the highest energy performance (kWhs/kWp) at sites throughout the world

Independent Sites

ASU

CREST

University of Cyprus

7% more than CdTe

6% more than px-Si

12% more than a-Si

9% more than px-Si

7% more than px-Si

University of Stuttgart (IPE)

7% more than CdTe

7% more than px-Si

16% more than a-Si

For Stuttgart and Cyprus (3 year studies):

SunPower modules listed as Suntechnics STM200FW

More kWh/W – Field data Data from 3 year field test by the University of Stuttgart and University of Cyprus is

showing superior energy per rated W performance of SunPower‡ modules in Germany

and Greece vs. other 13 technologies tested†

– As with any single test using a relatively small sampling of modules, low performance by a company’s modules

should not be taken proof of typical performance

– Test methodology matters (e.g., an a-Si module would receive an unfairly high kwh/kW result if the rated power

is used as the baseline and the test is less than several years, and would receive an unfairly low kwh/kW result

if the initial power before light soak is used as the initial value.

– The industry would benefit from more independent studies like this (long-term, carefully monitored and analyzed)

42

† Latest data at: http://www.ipe.uni-stuttgart.de/index.php?lang=ger&pulldownID=12&ebene2ID=44

Paper with methodology at: http://www.pvtechnology.ucy.ac.cy/pvtechnology/publications/22EUPVSECucyipe.pdf‡ SunPower modules listed as Suntechics STM200FW.

3000

3100

3200

3300

3400

3500

3600

3700

3800

4000

4200

4400

4600

4800

5000

5200

5400 Cyprus: June 2006 – 7/13/09 Stuttgart, June 2006 – 7/13/09


http://www.ipe.uni-stuttgart.de/index.php?lang=ger&pulldownID=12&ebene2ID=44



http://www.pvtechnology.ucy.ac.cy/pvtechnology/publications/22EUPVSECucyipe.pdf

MODULES

44

20% Module

96 cell, 165 mm, AR

glass, 328 W

Total area efficiency

20.1%*

0

50

100

150

200

250

300

350

0% 5% 10% 15% 20%

Ra

ted

Po

we

r (W

att

s)

Module Efficiency (%)

Photon Buyers Guide, February 2008

* Verified by Sandia National Lab

96 cell, 150 mm, AR glass

96 cell, 165 mm, AR glass

96 cell, 150 mm

72 cell

Module with Gen 2 cells:


45

Module-value enhancements cost less per Watt in high efficiency modules

Anti-reflection (AR) Glass provides:

2.7% relative power gain

4.0% relative energy delivery gain

Field test with AR and non-AR modules

Field site with checkerboard pattern of AR and no-AR coated modules


Some of the other module impacts on LCOE

Modules designed for downstream savings (and thus

lower LCOE)

– Larger and/or easy mount on trackers or other applications

– Integration with end application solution

Module cost reductions

– Design innovations

– Value-chain development

– Includes increased levels of automation

46


SunPower progress on panel ahead of schedule

<$2/W

<$1/W

0

1.5

3

Q4 2009 Q4 2014

Panel Cost: $/WWithout Imputed Efficiency & Energy Delivery Value

47

$3

>20% efficiency module&


RELIABILITY

49

Importance of reliability

Degradation rate, % failure rate, and financier confidence in reliability

are important drivers of the LCOE

Customer confidence in the reliability affects average selling price

Passing the standard tests in necessary, but not sufficient to ensure

world-class reliability. Typical minimum is:

– Retention of electrical, optical, and mechanical properties for 30 yrs

– 25 years with <20% efficiency loss

– Support for claim of reliability


50

Reliability: SunPower approach

Standard tests (e.g., tests in IEC 61215; UL1703; IEC61730)

– HF10, DH1000, TC200, etc.

Test to failure with standard tests then FA

Non-standard tests to try to generate new failure modes

– Combination of stresses essential

FMEA (Failure modes and effects analysis)

– Modes from previous learning, science, and tests

Non-standard tests to develop acceleration factors

– Different stress levels; Measurement of continuous variable preferred

Field tests with acceleration

– Extreme voltage, illumination, air quality, …

Field tests across a range of normal conditions


51

Reliability example: SunPower interconnects

New approach

developed to

eliminate

interconnect

fatigue failure

Extended tests

(past standard of

200 cycles) found

no ribbon failure,

but some solder

creep

Redesigned

interconnect to

reduce force on

joint & changed

to SnAg material

system (less

creep and no Pb)

Validated

expected

improvements

and >30yr life

prediction

Field studies of traditional-cell modules revealed that interconnect failure

(fracture and bond failure) was the #1 failure mode for modules


Validation experiments included:

Acceleration factors calculated by theory and estimated from different

temperature and different dwell time thermal cycling

Coupons with 4 point measurement of all joints with continuous monitoring

during thermal cycling for >2000 cycles (10 times the industry standard) –

results well matched to prediction, with 0% failures with redesigned

interconnect in +90/-40C

52

SunPower interconnects

Module results from both automated and manual manufacture:

-5%-4%-3%-2%

-1%0%1%2%

0 250 500 750 1000 1250

Number of +90/-40C Temperature Cycles

Re

lative

Chan

ge

in E

ffic

ien

cy

IEC 61215 passing threshold

(5% loss in 200 cycles)


SYSTEMS

Residential

54

Plus: Smart-mount system, monitoring, packaged systems …

Building integrated Building applied


55

Germany Italy

United States Australia

Residential


56

Installed Cost Reduction Roadmap: 50% by 2012$/Watt 2006 Downstream

Cell/Panel

Silicon

Efficiency

2012

25%

10%

10%

15%

Target Reduction:

60%

Residential Example


Commercial

57

No roof penetrations;

easy to install

New integrated

module/frame/mount,

developed under DOE

SAI program


Power plants

58

Nellis AFB, 14.5MW FP&L, 25MW


59

PORTUGAL11 MW Serpa10 MW Ferreira

GERMANY10 MW Bavaria I3.0 MW Bavaria II1.6 MW Pfenninghof

KOREA6.0 MW Samsung2.2 MW Mungyeong2.0 MW JeonJu1.4 MW Hampyeong1.0 MW Gwangju

HAWAII1.5 MW Lanai

CALIFORNIA1.7 MW Lake County1.2 MW Peninsula Packaging1.2 MW Napa Valley College1.1 MW Rancho Water1.1 MW Grundfos Pump1.1 MW North Bay Reg. Water1.1 MW Inland Empire Utility1.1 MW Chico Water Recycling1.1 MW Agilent Technologies1.1 MW Skinner Water Facility1.1 MW Gap Pacific Distr. Ctr.1.1 MW Marine Corp AGCC1.0 MW Sonoma County Water1.0 MW Applied Materials

NEVADA15 MW Nellis AFB3.1 MW Las Vegas WD

SunPower Global Power Plant Presence 325+ MW by Q3’09

SPAIN29 MW Naturener23 MW Trujillo23 MW Jumilla18 MW Olivenza14 MW Lorca12 MW Almodovar11 MW Magasquilla11 MW Ciudad Real9.9 MW Zaragoza8.4 MW Isla Mayor8.3 MW Guadarranque6.9 MW Caceres6.0 MW Atersa4.8 MW Llerena3.8 MW Lebrija

PG&E Contracted250 MW CA Valley Solar Ranch

EAST COAST – U.S.26 MW FPL–Desoto10 MW FPL–SpaceCoast1.6 MW Merck1.0 MW FPL–NASA1.0 MW J & J1.0 MW QVC Network1.0 MW SAS Institute

ITALY2.3 MW Toletino1.0 MW Ferentino1.0 MW Siron/Soleto

Australia0.3 MW Marble Bar0.3MW Nagline

© 2009 SunPower Corp. 59

Track record of energy production: Bavaria Solarpark

60

Expected Energy Production Actual Energy Production

Actual vs Expected Production: 106%

2005 2006 2007 2008

12,000

10,000

8,000

6,000

4,000

2,000

Pro

du

cti

on

in

MW

h

60


In summary, high efficiency silicon modules reduce LCOE by helping with:

Capital cost

Module $/W

Area-related BOS

Electrical BOS

Project-related $


61

Capacity Factor

Environmental conditions

– Total solar radiation

– Diffuse/direct & spectrum

– Ambient temperature,

wind, soiling conditions

Mounting

– Tracking vs. fixed

– Tilt angle

– GCR and shading

Performance

– Response to solar

radiation conditions

– Operating temp & temp

coef; Soiling

– Degradation rate

– System availability

Cost of capital

Perception of risk

Average financing

conditions

O & M

Reliability x

number of

components

Operation (cleaning, …)

Better response to low

light, varying spectrum,

and off-axis illumination

Lower operating temp.

and temp. coef;

Reduced soiling with AR

glass

Lower degradation rate

High reliability modules and fewer connections

Fewer components.

Grouped trackers.

Experience.

Solid financials backing

good warranty

30% increase in CF with

low cost tracking

Can decrease GCR while

still meeting energy/area

requirement

Reduced silicon, wafering,

and module conversion in

some cases can result in

lower $/W module

Less power loss

with partial

shading;

Ability to design

around

obstructions

Proportionate reduction of

material, installation, and

shipping – can have very

large impact.

Smaller footprint and fewer

connections

More Watts/project in

area-constrained jobs

means less $/kW fixed

costs;

Higher efficiency gives

lower ground prep and

other land costs

Experience; proven

performance

Thank you

Technology and Economics of High-Efficiency c-Si PV · 2016-04-06 · c-Si PV industry back on cost...

Documents

Transcript of Technology and Economics of High-Efficiency c-Si PV · 2016-04-06 · c-Si PV industry back on cost...