Challenges for the PV equipment supply chain in 2011

Part 1: Meyer Burger’s takeover target Roth & Rau confronted by declines in PV revenues & PECVD market-share

Finlay Colville, Senior Analyst, Solarbuzz

In reporting Q1’11 results on May 13, 2011, Roth & Rau highlighted year-on-year revenue growth for group activities (including PV) of 69.1% – from €35.3 million in Q1’10 to €59.7 million in Q1’11. Indeed, analogous to other PV companies recently across the value and supply-chains, emphasis was placed on year-on-year comparisons, not on quarter-on-quarter changes from the preceding reporting period Q4’10.

For PV equipment suppliers, the phasing of equipment spending cycles means that seasonality is not as useful a comparative measure as it is for downstream suppliers to the end-market. Moreover for Roth & Rau, Q1’10 represented the final quarter before the incremental revenues of OTB-Solar appeared on Roth & Rau’s books. These factors combine to make Roth & Rau’s impressive year-on-year growth figure of 69.1% in Q1’11 a somewhat less helpful metric when tracking current equipment trends and changes in supplier market-share.

With Meyer Burger having announced their proposed voluntary public takeover of Roth & Rau back in early April 2011, the key questions being raised within the industry are: What impact would this takeover have on the consolidated group performance at Meyer Burger? How is Roth & Rau performing within the PV industry today? Is the current PV product portfolio matched to equipment demand and PV technology preferences? What might the competitive landscape look like to Meyer Burger in 2012 and 2013?

To understand these issues, it is necessary to extract Roth & Rau’s PV revenues at the process tool level and to estimate market share. Further, by projecting the quarterly revenues on offer within Roth & Rau’s PV SAM over the next four to six quarters, the potential upside to Meyer Burger’s PV revenues can be assessed (if the takeover proceeds as proposed).

Putting aside Roth & Rau’s de-bookings in excess of €200 million over the past few quarters – including the turn-key wafer line previously ordered by Birla Surya – and the change in product mix from discontinuing former turn-key line options, more accurate trend statistics emerge for single-equipment and, in particular, individual process tools.

During Q1’11, Roth and Rau’s PV revenues declined sharply Q/Q by 46% to €40.3 million, at a time when the industry’s c-Si cell TAM grew Q/Q by almost 20%. As a benchmark here, Centrotherm’s c-Si cell/module revenues grew by over 40% during the same period; Amtech’s subsidiary Tempress posted c-Si diffusion furnace growth of 16%; and Applied Materials’ Baccini segment is forecast to return CY Q1’11 sequential growth in excess of 30%.

While Roth and Rau’s single-equipment PV revenues and backlog also declined sequentially during Q1’11, another key takeaway was the order intake for single-equipment which declined by nearly 50% Q/Q. However, to provide better clarity on market-share shift and quarterly revenues on offer, the methodology behind PV equipment spending and supplier market-share requires some explanation now.

PV equipment spending & market-share analysis

Developing equipment suppliers’ market-share estimates is an important part of assessing the competitive landscape. And enhancing PV equipment supplier analysis has become increasingly important with annual equipment revenues now exceeding the US$10 billion level. The need to bring equipment analysis for the PV industry up to the standards of adjacent market sectors such as semiconductor and display has never been more pertinent.

Previously, PV equipment demand was estimated typically on an annual basis, often simply by subtracting year-end (industry-wide) nameplate capacity levels and then performing a quick ‘back-of-the-envelope’ pro-rated arithmetic calculation for a handful of tool types. In practice, this gives accuracy at the ± 50% level from one year to another since this method neglects several factors specific to the supply-chain. In this calculation, equipment spending expansion phases cannot be matched to revenue recognition quarters; no provision is made for the widely varying Capex levels across each production line configuration; and differences in process tool ASPs (depending on tool/line throughput or suppliers selected) are not factored in either.

To provide accurate SAMs across key tool segments, a more appropriate methodology adapts proven frameworks used within adjacent technology sectors, and applies PV specific detail at the manufacturer and process tool level. Key to this is establishing quarterly resolution for every capacity expansion phase by manufacturers’ individual fab locations – with each of these expansions (or equipment spending phases) broken down into the quarter of tool order (supplier booking), tool delivery (supplier shipment/revenue recognition) and (staged) manufacturer ramp-up to mass production (annualized nameplate).

Process flows are then assigned to every expansion phase, enabling any tool type – such as PECVD tools, boron diffusion furnaces, ion implanters, PSG laser tooling, etc. – to be captured separately by quarter. For example, within the c-Si cell and thin-film segment, this equates to more than 1,000 discrete equipment spending phases within the 2005-2013 period alone, spread across over 100 predefined process flows.

Quarterly tool SAMs are then cross-checked against a separate analysis of equipment suppliers’ quarterly PV revenues and ASPs at the key process tool level, and by triangulating this data directly with leading tool suppliers for accuracy. Typically, this methodology for PV equipment spending gets to within 5-10% of forecasted tool SAMs each quarter when compared to quarterly revenues recognized by dominant suppliers of each tool type.

This approach populates the PV book-to-bill metrics from both an equipment spending (manufacturer) and revenue recognition (supplier) standpoint. In addition, discrete bookings, revenues and backlogs can be filtered across any geographical region, PV technology or process tool segment – eliminating the requirement on tool suppliers to populate monthly or quarterly spreadsheet returns.

Finally, to ensure that market-share trends are captured accurately – and to remove any lumpy one-off revenue recognition spikes in quarterly process tool revenues – it is necessary to use three or five quarter rolling averages for both equipment spending (SAM) and supplier tool revenues to determine quarterly market-share data. This also allows staged revenue recognitions associated with tool shipment and final payment/factory-acceptance to be factored into the analysis.

PECVD anti-reflective coating & surface passivation deposition tooling

Let’s return now to the tool segment within which Roth & Rau is most active today – PECVD tools for c-Si ARC/passivation-layer deposition. It is worth emphasizing again that turn-key variants are no longer options from Roth & Rau. In fact, the PV backlog at March 31, 2011 is dominated by orders on hand for ‘SiNA-based’ tool types. (For simplicity here, revenues from former OTB-Solar and ‘MAiA’ deposition tooling are grouped as ‘SiNA-based’ tooling).

During 2010, Roth & Rau was actively promoting several different turnkey options, through which Roth & Rau effectively acted as a contractor, supplying one or two tool types in-house and sourcing all other tooling from third-parties. Options here included c-Si wafer lines, standard process flow c-Si cell lines, two types of high-efficiency c-Si cell lines (laser-doped selective emitter and heterojunction ‘Sanyo-type’ cell), and even an option for turn-key CdTe production lines. Each of these options (as a turn-key line) is no longer offered by Roth and Rau.

Adapted from analysis provided within the PV Equipment Quarterly, the Solarbuzz estimate of Roth & Rau’s market share is shown in the accompanying figure here, starting from Q2’10 as the first quarter when former OTB-Solar deposition tooling was consolidated in Roth & Rau’s reporting. The analysis (based purely on revenue recognition) illustrates market-share decline (mainly at the hands of Centrotherm) during the past few quarters. In fact, as shown in the pie chart, Centrotherm’s growing dominance (within the highest-revenue process tool SAM across c-Si cell production) moved to a new level during Q1’11 with estimated PECVD revenues approaching 50% market-share.

Challenges Facing Meyer Burger

The proposed takeover of Roth & Rau by Meyer Burger captured headline media coverage during April 2011. Both companies featured in the Solarbuzz top-10 ranking of equipment supplier’s PV revenues for 2010. And their product portfolios – at first glance – suggest the scope for synergistic offerings from a sales and marketing perspective across most of the c-Si value-chain from wafering to module manufacturing.

However, during 2010, approximately two-thirds of Meyer Burger’s PV revenues were derived from organic product types, as opposed to new products incorporated as part of previous acquisitions. From a field sales perspective there is limited synergy in offering volume wafer cutting tools to a repeat (upstream) customer in Asia compared to selling an automated turn-key c-Si module line to a start-up in Europe for example.

With the addition of PECVD tooling for c-Si cell lines, the gap in customer profiles is certainly reduced. But there is no lack of obstacles at play for Meyer Burger if the takeover proceeds as outlined. If this happens during 2H’11, c-Si cell spending will be near the bottom of the current equipment spending cycle, and Roth & Rau’s market-share could be eroded by then over several quarters. And if tool upgrades for c-Si cell lines (automation, high-efficiency, etc.) then become the focus for the equipment supply-chain, other tool options here may simply outrank deposition tool upgrade spending.

Leading wafer manufacturers such as GCL and LDK are also busy positioning themselves in the long term as suppliers to cell manufacturers, not as direct competitors. And the vertical integration trend of recent has been mainly upstream by cell/module manufacturers to wafering: not downstream by leading poly/wafer manufacturers into an over-crowded midstream cell segment.

During the Q1’11 analyst call, Roth & Rau conceded that new orders for April and May 2011 did reflect a softening in demand from their customer base, and correspondingly indicated that new order intake for PV would be down sequentially in Q2’11. For observers tracking the Solarbuzz PV book-to-bill metrics in the past couple of quarters, there is no great surprise here. However, with Roth & Rau banking on revenue upside for SiNA tools in 2H’11, this signals a greater threat than exists for competitors. For example, Centrotherm announced record revenues for Q1’11, aligned with industry spending trends this year, and the PECVD market-leader is likely to accumulate the bulk of single-equipment revenues in 2011 during the first two quarters – not back-end loaded.

PECVD competition struggles to offer credible challenge

Looking forward, competition to Centrotherm (and Roth & Rau) within the c-Si PECVD segment will continue to target market-share gains. However, when c-Si cell equipment spending does finally return to positive growth territory (possibly as late as 2H’12), the competitive landscape may be fundamentally different than during 2010.

Local Chinese tool makers of low-cost PECVD alternatives are at risk of being frozen out the market at this time. These suppliers benefited mostly from the aggressive Tier 2 c-Si cell expansion in China during 2009 and 2010. However, it is debatable how many of these cell manufacturers will survive through the demand constrained midstream environment of the next 12 months, or will be competitive enough to merit any immediate capacity expansions – short of a large upswing to 2012/2013 demand. Few, if any, Chinese PECVD tool makers are therefore projected to have meaningful revenue opportunity within a post midstream shake-out, where leading Tier 1 vertically-integrated c-Si cell players dominate the top spots in PECVD sales prospects lists.

Turning to some of the more credible PECVD tool suppliers who have sought to gain market-share during 2010, fortunes have been mixed to date. Jusung Engineering’s selection as deposition tool supplier to one of the top-10 cell producers during 2010 came as a surprise to many last year, with Jusung offering limited track record in this space. Having launched the Singular PECVD product line to the market over two years ago, Singulus has struggled to obtain significant customer uptake so far.

Manz Automation and Orbotech’s subsidiary OLT-Solar have voiced their intentions to enter the PECVD market as competition to Centrotherm. Having previously been somewhat bullish during 2010 on a tool launch at the PVSEC exhibition in Hamburg later this year, recent rhetoric from Manz suggests a softening in attitude towards committing to any release date. Orbotech too has been warming the analyst community for some time regarding a significant adjacent market opportunity for OLT-Solar, and the first PECVD tool is slated to ship this month with initial revenues recognized in 2012.

Any appreciable upside to Meyer Burger’s PV revenues (if the takeover goes ahead) may be pushed out as far as 2013, at which point the challenge for the collective field sales organization could be more heavily weighted to tool adoption and customer ‘wins’, than to repeat order consolidation at Roth & Rau’s legacy customer base.

In Part 2, contributing author Finlay Colville will provide a comprehensive overview of an emerging tool type considered a key component within many next generation cell types within the industry – laser tools for PV. Questions addressed will include: What is the market size today for laser tooling used within c-Si and thin-film fabs? Who are the market leaders for tool types? Which laser-based processes are gaining traction and which appear to be resigned to perennial R&D activities?