SPI 2016 Quick Talk

13 September 2016

Module Quality Assurance: Risk Mitigation and Safeguarding Project Value

countries 27 Client engagements in

125+ solar factories worldwide

Audited

employees

Tech

nica

l Adv

isor

y S

upply Chain

Managem

ent

60

Over

More than Engineering Services

Quality Assurance Certified by Proud member of

Over

8GW experience

A presence in

8 Over

countries

8 years history

35 engineers

2

CEA is a solar PV advisory firm that is able to provide unrivaled insight into the manufacturing process to ensure the success of PV projects worldwide

Design

Phase 1 Pre-production Quality Control

Phase 4 – Downstream Support

! Defining QA standards and requirements

!  Contract/project set-up

!  Examination at job site (thermal imaging tests)

!  Third-party lab tests

!  Factory audits !  Production

monitoring

Construction

Phase 2 Production monitoring

Module Inverters Racking Other BOS

Factory Level System Level

Commissioning

Module Inverters Racking Other BOS

Phase 3 Post-production Quality Control

Module Inverters Racking Other BOS

!  System Design & Equipment Specs

!  Solar Output Analysis

!  Project & Construction Management

!  RFP deconstruct and LCOE analysis

!  System commissioning support

!  Technical Troubleshooting

Independent Engineers (IE)

Technical Advisors in the past have overemphasized downstream quality without looking as closely on the product in more detail

Other third-party test labs

3

•  IEC standards by design offer only a very basic level of assurance

•  Most experts agree that IEC standards are only good for 10 years in the field

•  Even extended testing only proves the manufacturer’s ability to produce a small number (20 pcs) of good modules

•  Bill of Materials compliance during production is not assured

•  Stability of production processes is not assured

•  Long duration (25-30 years) •  Limited terms •  Many exclusions •  Not comprehensive •  Diagnostics costs not covered •  Lost revenue not covered •  Replacement costs not covered •  Lengthy claim process •  Replacement retrofits are

technically challenging

Warranties Certifications

Module Bankability is resting on two problematic pillars: warranties and certifications.

4

CEA’s Quality Assurance Program (CQAP) prevents risk and ensures that downstream project stakeholders maximize the output of their system

Supplier’s certification

Product and factory certification liaising TUV, SGS,

UL, CE, etc. assessment centers

Bill of Materials (BOM) inspection data review

Input material monitoring

Production process control; monitoring of equipment

calibration, stringing, lay-out, lamination, curing and

framing

Production environmental control; monitoring of inventory storage and

manufacturing workshop

ISO 2859 Sample rule

QC standard

AQL rule

Product Inspection

Visual inspection

Functional test

IV test

EL test

Proper Packing

Container type

Container number

Seal number

Shipment plan

Content Listing

Packing list

Inspection certificate

Pre-Production Production Monitoring

Pre-Shipment Container Loading

5

Typical PV Module Production Process Flowchart: multiple steps, leading to multiple potential failure points

1. Material Preparation

4. Backsheet Lay Up

3. String Interconnection

2. Cell Tabbing & Stringing

5. Lamination

6. Trimming 7. EL Testing

8. Framing

9. Junction Box Installment

10. AB Glue Addition

1 2 2 4

5

7 8 9 10

11

6

1 2 11. IV test

12. Packaging

6

CQAP has been performed in various facilities of Tier 1 manufacturers and the distribution of defect rates has showed surprising results.

0.0%

2.0%

4.0%

6.0%

8.0%

10.0%

12.0%

14.0%

16.0%

18.0%

2012 2013 2013 2014 2014 2015 2015 2016 2016 2017 2017

Def

ect

rate

s

2013 – 2016 YTD

Defect data from QA performed on over 100 projects totaling many GWs of modules

15 MW

30 MW

75 MW

100 MW

YTD TTTT

Project size

7

CEA has analyzed 4 case study QA projects and the insights gained can show the value of performing quality work on future projects

8

1. Projects demonstrate highly variable defect rates

2. Defect distribution varies greatly with project, location and in time

3. FMECA risk analysis (IEC 60812) similarly shows great risk variability

4. Even a 1% excess yield loss, can lead to substantial financial loss

4 case studies of typical projects produced at Tier 1 manufacturers shows the variability across manufacturers.

[CELLRANGE] [SERIES NAME]

[VALUE]

[CELLRANGE] [SERIES NAME]

[VALUE]

[CELLRANGE] [SERIES NAME]

[VALUE]

[CELLRANGE] [SERIES NAME]

[VALUE]

5.68%

0.0%

1.0%

2.0%

3.0%

4.0%

5.0%

6.0%

7.0%

Project 1 Project 2 Project 3 Project 4

Average defect rate

Defect rate per project

9

Applying the FMECA methodology, the RPN defects are ranked for severity, occurrence and detectability to produce each RPN.

3,290

2,758

1,944

1,463

0

500

1,000

1,500

2,000

2,500

3,000

3,500

Project 1 Project 2 Project 3 Project 4

RPN (Risk Priority Numbers): high values show relative high risk

10

Projects with seemingly low defect rates, may have a high RPN, and vice versa, when defect risk analysis is performed

Scattered defect distribution show that even on the same production lines as module assembly processes are not stable

CELL 17%

BACKSHEET 13%

OTHER 1%

EL 69%

PRO JECT 1

CELL 36%

CELL STRING 6% GLASS

10%

BACKSHEET 7%

FRAME 29%

JUNCTION BOX 5%

OTHER 3%

EL 4%

PRO JECT 2

CELL 16%

CELL STRING 51%

GLASS 6%

FRAME 8%

OTHER 11%

EL 8%

PRO JECT 3

CELL 28%

CELL STRING 37%

GLASS 6%

BACKSHEET 6%

FRAME 13%

EL 10%

PRO JECT 4

11

CEA model for assumed module degradation per defect category*

12

8%

7% 40%

30% 8%

79%

73%

53%

29%

CELL

CELL STRING

GLASS

BACKSHEET

FRAME JUNCTION BOX

FINAL ASSEMBLY

OTHER

EL

0% 10% 20% 30% 40% 50% 60% 70% 80% 90%

100% Assumed losses

* in excess of warranted degradation, at project midlife, based on reasonable assumptions about degradation caused by defects

Correlation of defect rates to average power loss derived from 4 project case studies. This loss will be mitigated if Quality Assurance is performed. Average defect rate

5.68%

Average avoided power loss due to defects

(calculated according to degradation model)*

1.10%

0.0%

1.0%

2.0%

3.0%

4.0%

5.0%

6.0%

4 projects average values

13

* in excess of warranted degradation, at project midlife, based on reasonable assumptions about degradation caused by defects

QA Virtuous Effect: The presence of a 3rd party performing Quality Assurance leads to a “virtuous effect”. Overall quality improves. The absence of a 3rd party oversight, inevitably leads to higher defect rates ,and subsequently to higher calculated power losses by a factor.

Example of calculated NPVs for 2 projects, produced with and without Quality Assurance program, using three scenarios

14

$14.6 M

$12.5 M $11.5 M

$10.4 M

$25.4 M

$23.0 M $21.8 M

$20.6 M

$ M

$5 M

$10 M

$15 M

$20 M

$25 M

$30 M

Projects with Quality Assurance (no excess loss)

Good scenario Medium scenario Bad scenario

Project A Project B

Sizes: 100 MW DC Project A, fixed tilt PPA $80 MWH EPC cost: $95m Project B, tracker PPA $60 MWH EPC cost: $107m

Projects without Quality Assurance

A B Lending rate 6.00% 6.00% Equity 50.00% 50.00% Loan tenor 10 10 O&M $1,000,000 $1,000,000 Land $500,000 $500,000 Inflation 2.50% 2.50% Tax rate 30.00% 30.00% PPA rate escalation 2.00% 2.00% Degradation of modules 0.65% 0.65% Energy yield per year MWh/MW 2,200 1,400 PPA in $/MWh 60 80 Pdc in MW 100 100 Discount rate 5.00% 5.00% EPC cost $/Wdc 1.05 0.95 Interconnection cost $1,805,000 $1,805,000

Bad scenario: manufacturer shows a big drop in quality w/o QA program (4 x 1% defect loss, or 4% in excess of warranty in project midlife) Medium scenario: manufacturer shows a medium drop in quality w/o QA program (3 x 1% defect loss, or 3% in excess of warranty in project midlife) Good scenario: manufacturer shows a small drop in quality w/o QA program (2 x 1% defect loss, or 2% in excess of warranty in project midlife)

NPV loss in $/W for the 2 projects, without applying Module Quality Assurance Programs, using three scenarios

15

$0.021

$0.031

$0.041

$0.024

$0.037

$0.049

$0.000

$0.010

$0.020

$0.030

$0.040

$0.050

$0.060

Good scenario Medium scenario Bad scenario

Project A Project B

Bad scenario: manufacturer shows a big drop in quality w/o QA program (4 x 1% defect loss, or 4% in excess of warranty in project midlife) Medium scenario: manufacturer shows a medium drop in quality w/o QA program (3 x 1% defect loss, or 3% in excess of warranty in project midlife) Good scenario: manufacturer shows a small drop in quality w/o QA program (2 x 1% defect loss, or 2% in excess of warranty in project midlife)

The information herein has been prepared by CEA solely for the exclusive use of recipient. No representation, warranty or undertaking, express or implied, is made as to, and no reliance should be placed on, the fairness, accuracy, completeness or correctness of the information or the opinions contained herein. Neither CEA or any of its affiliates, advisors or representatives will be liable (in negligence or otherwise) for any loss howsoever arising from any use of this presentation.

16

Thank you for your attention