Proposed new damp heat test standardsfor thin film PV modules

K. Sakurai, K. Ogawa, H. Shibata, A. MasudaAIST

A. TakanoFWAVE

H.Tomita, D.Schmitz, S.TokudaSolar Frontier K.K.

Presented at SAYURI‐PV, 2016 Oct 4‐5AIST, Tsukuba, Japan

np

←

Current

Backgrounds

• Major international PV standards are now being created/revised under IEC TC82 WG2

• a major “overhaul” of the whole set of test standards‐‐‐ wider, stricter coverage of degradation modes

• includes the IEC61215, Design qualification and type approval of PV modules

MQT 01Visual inspection

MQT 06.1Performance at STC

MQT 03Insulation test

MQT 15Wet leakage current test

MQT 10UV precondition test

15 kWh/m2

MQT 11Thermal cycling test

50 cycles–40 °C to 85 °C

MQT 12Humidity freeze test

10 cycles–40 °C to 85 °C

85 % RH

10 Modules

MQT 15Wet leakage current test

1 Module

MQT 06.2Performance at

NMOT (see Note 1)

MQT 07Performance atlow irradiance (see Note 1)

MQT 04Measurement of

temperature coeff icients(see Note 1)

MQT 19.1Initial Stabil ization

(Note 4)

MQT 06.1Performance at STC

3 ModulesSequence A

1 ModuleSequence B

2 ModulesSequence C

2 ModulesSequence D

2 ModulesSequence E

MQT 19.2Final Stabil ization

MQT 05 & MQT 08Measurement of NMOT

& Outdoor Exposure Test 60 kWh/m2

(see note 2)

MQT 18Bypass diode

thermal test (see Note 2)

MQT 09Hot-spot

endurance test(see Note 3)

MQT 16.1Static mechanical

load test(design load)

MQT 17Hail test

1 Module1 Module

MQT 11Thermal cycling test

200 cycles–40 °C to 85 °C

MQT 13Damp heat test

1 000 h85 °C / 85 % RH

MQT 14.1Retention of

junction box test

1 Module

MQT 19.2Final Stabil ization

MQT 06.1Performance at STC

1 Modulemeasured together with modules from sequences C-E as

control

2 Modules

MQT 14.2Test of cord anchorage

MQT 03Insulation test

MQT 16.2Static mechanical load test

(Test load, γm>1)

old IEC61215’s test procedure revised IEC61215(draft)

wider/stricter coverageof degradation modesfor improved reliability

• Among the test procedures, "heating in dark" is often applied.• Damp heat• Thermal cycling• PID

etc.

However, in REAL conditions, how often does "heat + dark" happen?

Is it always the best test method?

Question...

When hot in field, there's always light(voltage)

➣ In the real field, module temperature seldom exceeds 50 C without > 0.2 Sun irradiation.⇒ "High temperature + dark" condition is rare.

At high temperatures, module is soaked in light and generating bias voltage.

Irradiation (Sun)

Norm

alized Voc (V

oc @1Sun = 100 %

)

Module temperature (Middle east)Module temperature (Japan)

Suns‐Voc resultM

odul

e te

mpe

ratu

re (

C)

Outline of this talk

Two examples of when conventional "damp heat" testsare NOT the optimum test procedure for PV modules.

Example 1: A flexible thin‐film Si module (prototype)

Example 2: Some CIGS products in the market

Example 1

On damp‐heat testing of a prototype flexible thin film Si module

Unique degradation mode observed in a prototypeThin film Silicon flexible module(1)

0123456789

10111213

Year

0.00.10.20.30.40.50.60.70.80.91.01.11.21.3

Effic

ienc

y (a

rb. u

nits

)

Power GenerationRadiationEfficiency

Pow

er G

ener

atio

n (k

Wh/

kWp・

day)

Am

ount

of S

olar

Rad

iatio

n (k

Wh/

day)

Relocation

99 00 01 02 03 04 05 06 07 08 09 10 11 12

(A.Takano et al, EU PVSEC 2013, 3BO.5.4)

Transparent Electrode (+)

Laser-Scribed Linesfor Unit-Cell Separation

Backside Electrode Plastic-Film Subs trate

a-Si Layer

Metal Electrode (-)

Series-Connect ion Holes Current-Collection Holes

Transparent Electrode (+)

Laser-Scribed Linesfor Unit-Cell Separation

Backside Electrode Plastic-Film Subs trate

a-Si Layer

Metal Electrode (-)

Series-Connect ion Holes Current-Collection Holes

Reproducing the degradationby applying forward bias during damp heat test

(A.Takano et al, EU PVSEC 2013, 3BO.5.4)

(Leak points)

EL image

0102030405060708090

100

0 1000 2000 3000Test Time (h)

Out

put P

ower

（ arb

. uni

ts） Damp Heat Test

Damp Heat & Current InjectionTest

Degradation reproduced by forward bias during damp heat testing

→ made it possible to fix the problembefore mass production

Example 2

On damp‐heat testing of the some CIGS products in the market

• The IEC61215‐1‐4 draft, which defines the new damp heat test standard for CIGS modules, requires a light soaking (LS) process before and after the damp heat test. Pass/Fail criteria are also changed from the current standard. 

Then...

Light soaking effect of CIGS

Pmax@STC

Exposure time in field

Initial(Nameplate)

Change of Pmax by light soaking(sometimes takes years until maximum)

long‐term degradation

Damp heat test defined by the “old” IEC61215

Pmax@STC

Time

Initial(NamePlate)

decreased output(loss of light soaking effect+ real degradation) increased Pmax by light soaking in field

Damp heat testpass/fail decision

pass criterion

Original draft for IEC61215‐1‐4Pm

ax@STC

Time

(NamePlate)

Damp heat test

pass/fail decision

pass criterion

LS LS

Use Pmax after light soaking for pass/fail decision(not the nameplate value)

decreased output(loss of light soaking effect+ real degradation)

What actually happensPm

ax@STC

Time

(NamePlate)

Loss of LS effect+ “test‐specific degradation”

(not observed in field)

Damp heat test

pass/fail

pass criterion

LS LS

does not fully recover by LS(though no problems areobserved in real field)

Exploring the "right" test options for CIGS

5 types of tests below were performed with 170W-class CIGS modules.

* Performed at AIST. Others were performed at SF.**NMOT : Nominal Module Operating Temperature.

Test sequence2nd Pre-LS 1st Post-LS1st Pre-LS Test 1000hrs 2nd Post-LS

・Open-circuit state・Each LS process:1kW LS chamber for 21 hrs (SF) orOutdoor for one week (AIST)

・Open-circuit state・Each LS process:1kW LS chamber for 21 hrs (SF) orOutdoor for one week (AIST)

Normal DH

DH with forward bias

Test type

DH with irradiation

Normal dry heat test

Dry heat test with forward bias

ー

ー

0.1~0.65 Sun*

ー

ー

ー

ー

ー

Vpm@NMOT**Vpm@STC*

Vpm@NMOT**Vpm@STC*

Irradiation Bias

Wet

Dry

Experimental setup

Module temperature: 85℃Chamber humidity: 85 %r.h.

Module current : ~ 0.2 A(with NMOT‐Vpm, Tmod 85 ℃)

Test with forward biasTest with light irradiation

Module temperature: 85℃Module humidity: 85 %r.h. (raised the chamber humiditydepending on chamber temperature)

np

←

Current

np

(light) Current

Effect of light illumination

Test‐specificdegradation

illuminated

not illuminated

Compared to illuminated DH,DH in dark invokes a degradation not observed in field

Normalize

d Pm

ax

Initial(Light soaked)

Effect of light illumination + short‐circuiting (preliminary)

Short circuiting of modules → test‐specific degradation→ not the light, but forward bias voltage?

illuminated+short

illuminated

dark

Test‐specificdegradation

Normalize

d Pm

ax

Time

Initial(Light soaked)

Effect of forward bias + DampHeat(DH)/DryHeat(HT) tests

Bias voltage gives same results to light illuminationTest‐specific degradation does not depend on humidity

with bias voltage

without biasvoltage

Normalize

d Pm

ax

Damp heat test85℃ (module)

Damp: 95% r.h., Dry: <10% r.h.(chamber)

Initial(Light soaked)

Does the test‐specific degradation recover by loooong light soaking?Part 3. Experimental result

Normalized

 Pmax (%

)

Test‐specificdegradation

~ 3+ %DH with NMOT‐Vpm (N =4)

DH without bias (N = 4)

Pre‐LS DH 1000 h Post‐LSOC‐LS OC‐LS Pmpp‐LS

Total Post‐LS :OC‐LS 42hrs

& Pmpp‐LS 126hrs

No, it does not

Summary

➣ Understanding of mechanism of test‐specific degradation.• Wavelength dependency of light• Work on other samples• Destructive analysis• etc.

➣ In the real field, high temperature is always accompanied with illumination&voltage.("Hot+dark" does not happen ‐‐ it's the MOST unnatural test condition!)

➣ The examined CIGS modules after DH with light irradiation or forward bias showed similarbehavior to real field. Meanwhile, conventional DH invoked “Test‐specific degradation”,which is not observed in field.

➣ Test‐specific degradation is irreversible by light soaking.

"DH+forward bias" option added to IEC 61215‐1‐4 (draft).

Homeworks

Mysteries➣ The cause of test‐specific degradation➣ Some other products show same behaviors, while some others do not

(private communications)

Important message

Over 50 IEC standards are being created or revised right now;Though people are working hard, mistakes can happen.Keep watch on the progress of IEC TC82 WG2!Next face‐to‐face meeting: October 2016, Colorado

Thank you for your attentionContact : k‐sakurai@aist.go.jp

Backup: How to determine the chamber RH value

Chamber temperature (C) 

Cham

ber R

H (%

)

➣ Chamber RH should be set to realize the module surface RH 85%.In case of our study, chamber temperature was set 84 C to realize the module temperature 85 C.Then, according to this graph, we set chamber RH 88 %.

The curve of chamber relative humidity as a function of chamber temperature can be plotted using Tetens’ formula:

85

6.11 10.

.

1100

RHMOD: Relative humidity of the module surface (= 85 %)e(85): Saturated vapor pressure at 85 °C (= 581 hPa)RHCHM: Relative humidity inside the chamberTCHM: Temperature of the chamber