T e c hno lo g y to Ma rke t I mpa c t: I ntro duc tio n to T - - PowerPoint PPT Presentation

T e c hno lo g y to Ma rke t I mpa c t: I ntro duc tio n to T e c hno e c o no mic Ana lysis (T E A) a nd Ma rke t Ana lyse s

Joe Stekli

F OCUS K ic ko ff Me e ting June 25, 2014

1

With credit Joel Fetter (ARPA-E/Booz Allen)

2

Alte rna tive T itle :

Why are TEA and Market Analyses Important?

Interesting! Hey check out my new FOCUS system that can solve all your problems! It’s exergetic efficiency is 75% and it costs less than $1/W. 3 Your idea is interesting. But I hear about solar companies going

• bankrupt. And what proof do you

have that your numbers are accurate, what is the potential market size for the technology and how much of that market can you capture?. Researcher Potential Investor Hey check out my super- expensive, unreliable gadget that we’re cooking up in lab and are trying to figure out what we can do with it!

Credit: Joe Miler (ARPA-E)

What they really mean: What they hear:

Summary

• Technoeconomic analyses

Why it is important What it is not How they are properly done

• Market analyses

Why it is important An example culled from the CSP industry

4

Why Should I Complete a Technoeconomic Analysis?

“There is too much uncertainty to calculate exactly how much my widget will cost until the design is finalized”

$ Cost WTP

+

TEA Still has Value…

• Most Valuable Technical

Improvements

• Inform Potential Trade-offs,

Targets, and Metrics

• Can Bound Theoretical Limits
• Understand the Minimum

Viable Pricing

Credit: Ned Chiverton (ARPA-E),

Who Will Use This Analysis?

• You

– Building an effective cost model can help direct your R&D efforts, focus networking, and own the follow-on funding discussion

• Your Investors

– Strategic investors will want to understand your thought process on value creation and the cost model is key to this

• Your Partners

– In corporations, managers will use cost models as a tool to help allocate resources

Inform R&D Effort Set Research Targets and Metrics Understand Viability / Pitch investors Inform Analysts, Bank Loans

Credit: Ned Chiverton (ARPA-E),

What a TEA is Not

7

Component Units Value Cost Size of System MW 10 ‐ Cost of Storage Media $/kg 1 $100,000 Cost of Tanks $/tank 250000 $500,000 Cost of BOS ‐ ‐ $1,000,000 Cost of Power Block $/kW 100 $1,000,000 Total Direct Costs ‐ ‐ $2,600,000 Indirect Costs % Direct 20 $520,000 Contingency % Direct 10 $260,000 Total Indirects ‐ ‐ $780,000 Total Capital Cost ‐ ‐ $3,380,000 System Cost $/kWe $338

*These values are fictitious and not representative of any real or proposed system

What is a TEA?

Inputs Costs

MODEL

• Recipes
• Procedures
• Raw Material Prices

(est.)

• Energy Prices (est.)
• Labor Prices (est.)
• Technology

Performance

• + Assumptions!
• Operating Cost
• Raw Materials
• Labor
• Capital Cost
• “Fully-Loaded” Cost

Credit: Ned Chiverton (ARPA-E),

How Does it Work?

Before you Model

• Scope the Problem

– Define your product

• How far downstream is the product you are developing (components vs. system)?

– Define your inputs

• How far upstream are the materials you will purchase (reflectors, cells, receiver tubes)?
• What downstream processes will you need to integrate with (e.g. TES)?
• Considerations

– Cost model should approximate business model – Where your innovation will be disruptive

• Location in the value chain

– Should align both cost and value (requires a product or service)

Credit: Ned Chiverton (ARPA-E),

How Does it Work?

Creating a Process Model – Product

Bill of Materials Process Fabrication Steps Drawn/AutoCAD Factory Design Detailed Construction Drawings

How Does it Work?

Example – The Block Flow Diagram

Stage 1 – Cell fabrication

Raw Materials Encapsul ation

Stage 2 – Receiver fabrication Stage 3 – System Integration

Substrate, dopants, gases

Collector Product

How Does it Work?

Collecting/Estimating Prices

• Process modeling has yielded metrics that can be priced

(warehouse square footage, raw material usage per day, etc.)

• Depending on the sophistication of your cost model, you will

need varying accuracy in your price inputs.

• Try to understand the variability inherent in estimating prices

Free Internet Sources Market Reports and Vendor Discussions Vendor Quotes Negotiated Prices Focus

How Does it Work?

Distribution of Capital Cost

• Money now is worth more than money later
• A discount rate to relate future money to current money
• Current $ = Future $ / (1 + discount rate)^(time)
• For most energy projects at this early

stage, using a 10% discount rate is a reasonable estimate.

• As your project advances, an accurate

financial model will be required with an updated discount rate.

Capital Costs: Money spent on things (land, buildings, machinery, etc.) in order to bring a project to operability

How Does it Work?

Distribution of Capital Cost (cont.)

• If we assume a 10% discount rate, we can approximate the

“doubling time” as 7.2 years

• This will allow us to allocate capital costs to our product.

Capital Costs

+

Labor and

• ther OpEx

Materials Allocated CapEx “Fully Loaded” Cost

How Does it Work?

Using your Model

• Sensitivity Testing for Key Variables

– Try to evaluate the range of scenarios that could potentially occur. – Some places call this a P10/P50/P90 (in reference to probability notation) or a tornado diagram (in reference to the natural disaster it most resembles) Example:

• Currently Targeting Gold usage of 0.1g/unit

– Reasonable Target, need some modest productivity improvements (P50) – Currently using 0.3g/unit, but have put much effort toward reduction (P90) – If you can make a great, but unlikely, breakthrough, you will be able to use only 0.01g/unit. (P10) $43/unit $39/unit Gold Use per Unit Gold Price per troz. $xx/unit $54/unit $xx/unit Input 3 $xxx/unit $xx/unit Input 4 $xxx/unit $xx/unit

Cost Model E volution

16

Increasing Technology Readiness Level

Data Inputs

Small‐scale experimental data / conceptual prototype Small‐scale process unit data / "Minimum Viable Product" Pilot plant data / Scale Product Demonstration scale data / For Sale Production

Process Model

Block Flow Diagram / Bill of Materials Process Flow Diagram / BOM + rough scematic Process Simulation / Computer Aided Design Simulation verified with

• perating data / Detailed

CAD

Price Inputs

Published prices, estimates based on similar products / processes Vendor discussions to inform major costs, estimates on others Vendor quotes for most equipment Negotiated contract data

Level of Detail

Major cost drivers only Estimates of majority of

• perating costs and

capital equipment 90% of equipment and

• perating costs included

as a line item 98% included and verified by an independent 3rd party

Capital Costing

Recognizing that it will have a required return "Rule of 72" ‐ 10% discount rate Discount rate based on variability of free cash flow with market Full Weighted Average Cost of Capital with all Tax Shields included

Approximate effort

40 man‐hours 200 man‐hours 2,000 man‐hours 5,000+ man‐hours

Review/Input

Co‐Worker Review Several co‐workers from varying disciplines Input from a potential investor under a NDA Fully shared with EPC and bank, open to modification/scrutiny

Cost Model Use

Focusing Research and Development Effort Developing research targets/goals Understanding long‐term viability of the technology, pitching VC's Securing Bank Financing, projecting earnings, activity‐based costing

Summary

• Technoeconomic analyses

Why it is important What it is not How they are properly done

• Market analyses

Why it is important An example culled from the CSP industry

17

Why Should I Complete a Market Analysis?

“My design will produce electricity that is too cheap to

• meter. I don’t need to worry about what is currently there.”

$ Our cost Competition

Market Analysis is still key…

• Avoid Pitfalls That Others

Made

• Inform Potential Supply Chain

Partners

• Inform Geographical Location

and Therefore Design

• Follow-on Funders Want to

Know Size of the Market

19

How Does it Wor k?

Global installed c apac ity almost sur passed 3 GW in 2012

20

551 475 1151 1094 733 3982 1839 2955 950 250 475 623.5 1288.5 6 55.5 117.7 2032.3 541.7 1000 2000 3000 4000 5000 6000 7000 Israel South Africa Brazil Australia India China Spain US Announced / Planning Begun Permitted Finance Secured / Under Construction Commissioned

How Does it Wor k?

Spain leads c umulative c apac ity, while the United States has the lar gest pipeline*

*As of August 2013 Source: BNEF, SEIA

160 55.5 325 596 300 2,064 1,190 775

500 1000 1500 2000 2500 MENA India South Africa Capacity (MW) Current and Planned Capacity in Emerging Markets Current Capacity Financed Announced

Source: BNEF, CSP Today, SBC Energy Institute

Regional highlights Morocco: ACWA secured financing for 160 MW Ourzazate Project Abu Dhabi: Shams 1 connected to the grid, becoming the world’s largest operational CSP plant Saudi Arabia: Announced a 25 GW target for CSP by 2032 (part of 41 GW goal for solar)

1,000 2,000 3,000 4,000 5,000 6,000 7,000 8,000 9,000 Capacity (MW) Projected Growth in Global CSP Capacity Spain US China MENA India ROW

In Spain, economic uncertainty and cuts to renewable energy subsidies are capping CSP growth; it is unlikely that new plants will be built in the coming years

21

How Does it Wor k?

Signific ant gr

• wth is expec ted in the ME

NA r egion, India, and South Afr ic a

• Local manufacturing – especially of mirrors/receivers –

has been critical to growth in CSP to date

• The MENA region’s ability to expand manufacturing

capacity will likely impact its viability as a major market

Adapted from World Bank (2011) 22

Source: BNEF, World Bank

100 200 300 400 500 600 Capacity (MW)

Current and Expected CSP Capacity in MENA Countries Project abandoned Announced / planning begun Permitted Financing secured / under construction Commissioned

Strengths Low labor costs High solar potential Expected growth in electricity demand Proximity to Europe Existing glass- manufacturing capabilities Weaknesses Small existing market Limited financial markets for new financing’ Higher capital costs Highly-subsidized energy Limited local experience with renewable energy Opportunities Cost reduction (components) Investors attracted to growing demand Interest in building energy technologies industry Export potential Threats Insufficient training of local workforce Limited local awareness/understanding of CSP Access to financing Higher manufacturing costs

MENA SWOT Analysis

Although MENA countries currently have only ~160 MW of installed and operating capacity, there is another 1.9 GW in the pipeline

How Does it Wor k?

Over all, gr

• wth in the ME

NA r egion will depend on its ability to expand loc al manufac tur ing

0.04% 3% 64% 33% Dish Linear Fresnel Parabolic Trough Power Tower

0.1%

2% 95% 3%

Operational

0.1%

6% 70% 24%

Under Construction

21% 79%

Under Development*

2000 4000 6000 8000 10000 12000 14000 Parabolic Trough Power Tower Linear Fresnel Dish

Capacity (MW) Global Capacity by Technology Project abandoned Announced / planning begun Permitted Financing secured / under construction Commissioned Global Capacity by Technology, Current and Pipeline (MW) 23

Under Development: Projects having a signed agreement, but actual construction is still pending Source: NREL SolarPACES, BNEF, BAH Analysis

2,676 MW 1,762 MW 1,895 MW 510 MW 602 MW 99 MW 144 MW 46 MW 4,948 MW 2,561 MW

How Does it Wor k?

Par abolic tr

• ugh has been the tec hnology of c hoic e, but

power tower s ar e gaining tr ac tion

Average Capacity (MW) Operating

• Temp. (˚C)

Peak Plant Efficiency Annual Solar- to-Electricity Net Efficiency Annual Capacity Factor Concentration Factor

• Max. Solar

Field Slope Water Required (m3/MWh) Parabolic Trough 10-300 350-550 14-20% 11-16% 25-28% (w/o TES) 29-43% (7

• hrs. TES)

70-80x <1-2% 3 (wet cooling) 0.3 (dry cooling) Power Tower 10-200 250-565 (1,000 possible*) 25-35% 7-20% 55% (10

• hrs. TES)

>1,000x <2-4% 2-3 (wet cooling) 0.25 (dry cooling) Linear Fresnel 10-200 390 18% 13% 22-24% >60 <4% 3 (wet cooling) 0.2 (dry cooling) Dish Stirling 0.01- 0.025 550-750 30% 12-25% 25-28% >1,300 >10% 0.05-0.1 (washing)

Despite its many benefits, dish technology has a variety of technical and engineering drawbacks; as a result is not expected to experience major growth in the near term

• Namely, Stirling engines produce electricity directly and do not require a “heat sink” (and is thus poorly suited for energy storage);

this makes it comparable technology to PV, which is much cheaper

• Technology is still in its demonstration phase, so scalable systems costs are unclear and have deterred investment

24

*with advanced working fluids (not yet developed) Source: CSP Today, CSP Today, IRENA (2012), Fitchner Best Worst

How Does it Wor k?

While eac h tec hnology has pr

• s and c ons, tr
• ugh and tower

ar e c ur r ently the most matur e & ec onomic ally viable

Primarily Power Tower

*approximate breakdown (costs vary based on plant size, TES, etc.) Source: Company websites, BAH Analysis, IRENA (2012), CSP Today

Although parabolic trough technology is proven and relatively established, power towers are expected to expand their market share

• Increased competition with PV and cutbacks in renewable energy subsidies are making it more difficult for CSP to

compete

• Power towers, however, have certain inherent technological benefits (especially when molten salt is used as the HTF

and storage medium) that make them more economically viable

• Higher temperatures attained by power towers improve the technical and financial viability by:
• Lowering energy storage costs (or increasing how much energy can be stored for the same cost) due to

greater temperature differentials

• Yielding higher achievable capacity factor (up to 55%) and peak plant efficiency (23-35%)
• Other benefits include reduced water consumption for cooling and the ability to provide firm output
• Additional experience is likely to reduce the risks associated with power towers and reduce costs

25 Primarily Parabolic Trough Technology Agnostic Key Technology Players

How Does it Wor k?

Most c ompanies pr

• duc e both tr
• ugh and tower

tec hnologies, with few pur e- play tower manufac tur er s; this is expec ted to c hange as the mar ket matur es

Turbine Receivers Reflector / Mirror Design & Integration EPC 26

Source: World Bank, NREL SolarPACES, BNEF, BAH Analysis, CSP World Map, CSP World

HTF System Storage OVERVIEW Large, often vertically integrated glass companies; some from the auto sector Highly concentrated market Companies with experience in thermal power plants, grid connection, & power distribution Large, well- established suppliers Current state-

• f-the-art is

two-tank molten salt storage system Large int’l companies; experience w/ project dev., financing, & engineering Experience with large- scale energy, construction, transport, & infrastructure projects Experience with large- scale energy, construction, transport, & infrastructure projects INPUTS Glass Polymer films Metal tubing Anti-reflective glass coating HTF (synthetic

• il, water,

molten salt) Piping, pumps Insulation Molten salt Storage tank Heat exchange Pumps Foundation Financing Engineering

• KEY PLAYERS

TR Primarily Parabolic Trough TO Primarily Power Tower AG Technology Agnostic

CURRENT FOCUS

Project Dev. Flabeg Rioglass Alanod Skyfuel / ReflecTech Saint-Gobain Guardian Glaston eSolar Rioglass* Schott** Archimede Sener Siemens Kraftanlagen Siemens ABB MAN-Turbo Alstom GE Ormat Areva Dow Solutia Sener Flagsol Abengoa BrightSource Cobra Ibereolica Acciona SolarReserve eSolar Solargenix Torresol ACWA Areva Solel Schott Archimede Flagsol Siemens Cobra Abener Sener Bechtel Lauren

TR TO AG TO TO TO TO TR TR TR TR TR TR TR TR TR TR AG AG AG AG AG AG AG AG AG AG AG AG AG AG AG AG AG AG AG AG AG AG AG AG AG AG AG AG

Integrated Specialized

*Rioglass acquired Siemens’ receiver technology (previously Solel) in Sept. 2013 **As of Sept. 2013, Schott is reportedly seeking a buyer

How Does it Wor k?

Due to the industr y’s immatur ity, ther e is a fair amount of c onsolidation, entr ies, and exits; this lac k of supply c hain stability has pr evented c osts fr

• m

falling signific antly

EuroTrough UltimateTrough Aperture width Aperture length

100% increase 30% increase

27

Source: Company websites, sbp, 50% increase 40% increase

SkyTrough SkyTrough DSP Aperture width Aperture length 2008 1990s 2011 Present

• Doubling the aperture per collector

expected to reduce costs 20% compared to the EuroTrough (due to efficiency gains)

• Also cuts the number of drive units,

sensors, control elements, pylon foundations, loop specific piping, and construction procedures in half

• 25% increase in concentration ratio
• 40% increase in aperture width
• 50% increase in length

How Does it Wor k?

L eading tr

• ugh manufac tur

er s ar e inc r easing mir r

• r

size to r educ e c osts & impr

• ve effic ienc y

Turbine Receivers Reflector / Mirror Design & Integration EPC HTF System Storage Project Dev. Utility $130M Early Investment $40M Investment and licensing agreement $10M investment for 3 projects totaling 500 MW; rights to PPAs to operate 11 eSolar units International industrial services provider; supply agreement Energy and telecom company; $30M Investment; licensing agreement to help meet 1 GW goal for India by 2019 Licensing agreement to build up to 2 GW in China over 10 years

Source: Company websites

28 Investors/Partners Formal partnership Supply agreements Manufacturing partnership PPAs / generation offtakers

How Does it Wor k?

Companies like eSolar foc us on signing for mal ar r angements with established developer s to deploy their tec hnology

Feature Cost Reduction Opportunity Heliostat size

• Smaller heliostats sit lower to the ground and weigh less, reducing cost of steel support structures
• Heliostats are controlled individually to improve accuracy
• Locally manufactured

Heliostat height off ground

• Reduces costs of ground penetration
• Accommodates shorter towers, thus reducing equipment costs and expediting permitting process

Scalability

• Smallest modular installation only requires ~200 acres of land
• Smaller plant size means they can located closer to the grid

Fabrication and construction

• Pre-fabricated heliostats/frames can be shipped in compact standard shipping containers,

minimizing onsite construction costs

• Simple installation process reduces labor costs
• Deployed in less than 18 months of construction

Heliostat cleaning

• Robotic cleaning technology significantly reduces water consumption

29

Source: eSolar, Greentech Media

“Less is more” approach

• Smaller heliostats can be shipped and assembled more quickly

and at a lower cost

• Lower-profile heliostats require less steel, reducing support system

costs Business model transition: from plant developer to technology provider

• Initially focused on plans to build power plants, but difficult

financing environment convinced eSolar to shift focus to providing technology licenses to larger developers

How Does it Wor k?

eSolar ’s manufac tur ing and tec hnology appr

• ac h is foc used on r

educ ing c osts

Conclusions

• TEA is valuable to your R&D effort

– Helps show the most valuable technical improvements – Informs potential trade-offs, targets, and metrics – Can bound theoretical limits – Understand the minimum viable pricing

• Market analysis provides key insights to

the development of your technology

– Avoid pitfalls that others have made – Inform Potential Supply Chain partners – Inform Geographical Location and Therefore Design – Follow-on Funders Want to Know Size of the Market

30

Thanks! Joe Stekli

joseph.stekli@hq.doe.gov

31