Printing:This  poster  is   wide  by  36”  high.  It’s  designed  to   printed  on  a  large-format ter.

Cust tent:s in this poster are

formatted for y Type in the placeholders t ext, or click an icon t able, chart, SmartArt gr e or multimedia file.

To add or remo e bullet points from text, jus Bullets button on the ab.

If you need mor s for titles, content text, just make a cop t you need and drag  it  into   PowerPoint’s  Smart Guides w ou align it with everything else.

Want to use y wn pictures instead of our No problem! Just right-click a e and choose Change Pictur tain the proportion es as you resize by dragging a c .

s?• MFG1 and MFG2 show very little degradation after 1000hrs of light soak • MFG1 has begun to recover to match MF2

-1.0%

0.0% Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

48”Laboratory Testing at STC: Necessary but Not Sufficient(Real World System Testing Picks Up Where Lab Testing Falls Short be

prin

MPropst, NAOlsson, CRichardson; pearllaboratories, 2649 Mulberry Unit 15, Fort Collins CO 80524

omizing the ConAbstract Outdoor Energy Yield Detailed Failure Analysis • Historically performance evaluations have been performed at standard test • MFG1 and MFG2 were installed outdoors in identical 5kW systems DLIT imaging

1A forward bias on full module conditions (STC). A number of pitfalls may skew accelerated lab test results The placeholder• Additionally, 4 modules of each were installed on individual channel MPP when evaluating performance at STC. Skewed results may over-estimate or trackers and IV curves were swept every 5 minutes.

Emission from center of cell

o add t(under-estimate real world performance often leading manufacturers astray. ou.• Energy yield appeared very similar, noticeable differences occurring on A comprehensive look at module performance in real world conditions is

lowlight days required to compliment the lab test results. For example, metastabilities in

o add a t aphic, pictur

v

thin film technologies are inherent in the measurements and can result in incorrect conclusions. Parametric values like temperature coefficients and low light performance may not be stable and can degrade more rapidly than measurements at STC will show. These parameters play a big part in the

In this study we present a detailed analysis comparing and contrasting results between accelerated lab testing and outdoor performance testing. This study highlights the shortfalls of STC only performance assessments.

0

1

2

4

5

6

7

70.0%

75.0%

80.0%

90.0%

100.0%

105.0%

110.0%

SunH

rs (k

Wh)

%P

erfo

rman

ce (S

TC)

MFG1 vs MFG2 Outdoor %Performance (Normalized to STC)

EL imaging2.3A forward bias on full module

economics of solar installations due to loss in overall energy yield.

95.0%

Accelerated lab testing is necessary in providing some assurances in stability and durability but field performance is the critical and complimentary

3

piece of testing required to accurately predict performance of installations. 85.0%

SEM micrographs found anomalies on film surface

Film

t click the

our o

ill help y(

Day1

MFG1 MFG2

MFG1 MFG2

Day45

y of wha(or body

e placeholder(

Home tand normalized to sticker giving a %Performance (STC) value. Protective • MFG1 and MFG2 were subjected to 1000hrs of indoor 1 sun continuous light Coat soak. Modules were pulled from soak and tested at STC every 100hrs. • The days total sun hours in kWh are plotted on the secondary axis to CdTe highlight lowlight vs full sun days. CdS • MFG1 shows significantly poorer performance when days are cloudy or TCO higher percentage of lowlight hours

• Post 45 days performance vs irradiance has degraded at lowlight in MFG1

Indoor Light Soak • Using module temperature and Irradiance, each value was corrected to STC 10/12/2011 10/19/2011 10/26/2011 11/2/2011 11/9/2011 11/16/2011 11/23/2011 11/30/2011

MFG2 MFG1 Sun Hrs

Performance vs Irradiance • This defect was traced back to the CdTe source form factor which resulted in “spitting”  during  sublimation. • Initial performance vs irradiance was very good for both MFG1 and MFG2

place.Conclusions • Very different conclusions can be drawn about the equivalence of these two

thin film manufacturers when looking at laboratory STC testing and actual outdoor performance.

• This particular type of Rsh defect degrades over time and manifests itself in lowlight performance first.

Observations %Energy Yield Loss Due to Degrading Rsh, Poor Lowlight Performance

pictur• IV curves appear nearly identical under STC IV Analysis -4.0% -3.0%

%E

nerg

y Yi

eld

Lost

-2.0%

e. Main of pictur

orner

• A more in depth analysis of the IV curves after 1000hrs of indoor light soak shows a significant difference when tested off of STC

• Using neutral density filters to assess the curves at varying IRR levels reveals MFG1 has degraded performance under lowlight conditions

-0.5

0

0.5

1

1.5

2

2.5

-10 0 10 20 30 40 50 60 70

Cur

rent

(A)

Voltage(V)

MFG1 vs MFG2 Post 1000hr Lightsoak

Time Zero 1000hrs

Cur

rent

• The PVSYST PAN files for these 2 manufacturers show identical dark Rsh and exponential relationships because at time zero they do match.

-7.0%

-6.0%

-5.0% $$$ LOST! %Loss MFG2 %Loss MFG3

• Adjusting the PAN files to account for the degraded Rsh values reveals a

significant loss in energy yield each year.

• Above shows the %energy lost when MFG2 is modeled using a 50% Rsh relationship and MFG3 is modeled using a 25% Rsh relationship.

`

Voltage