Reported by Jeanny H. Lim (swied@infothe.com)

What are the challenges you see ahead for inverter manufacturers in terms of market growth?

The solar PV market, together with the inverter market has a tremendous amount of potential. The solar industry has been keeping up its momentum and is expected to grow at impressive rates in the next years. But the way to growth is paved with quite some unpredictable hurdles. The solar market still shows the behavior of market instability and remains dependent on government subsidies. The same governments which decide on subsidization policies in support of solar also need to justify these policies in relation to their constituencies and sometimes fail to do so. As a result, we keep experiencing abrupt changes in direction and need to be prepared to instantly shift focus from one region to another or from one segment to another. A certain country or region may be a booming market one year and then be off the radar the next, while a different market elsewhere rises. Given such a ‘pop-up’ natured demand, it is a major challenge for inverter manufactures to ensure availability of the demanded inverter capacities and their delivery within a reasonable lead time. In order to do so inverter manufacturers need to maintain a high enough level of flexibility in capacity ramp and distribution chain management. 

Is there anything that makes it particularly difficult for inverter manufacturers to be that flexible?

Inverters include many semiconductor components and are, therefore, more dependent on third party suppliers than for example module manufacturers. In 2010, the industry experienced a shortage in Insulated-Gate Bipolar Transistors [IGBTs] which caused a global inverter shortage just as demand peaked in Germany. Foregoing predictions of an economic crisis had cautioned manufacturers of IGBTs to produce less than what was eventually needed.

With efficiency being one of SolarEdge’s fields of expertise, let me ask you, what can be done to improve inverter efficiency?

This is a really tricky question. It is true that the higher the efficiency of each component in the system, the higher the overall efficiency of the installation. And inverter efficiencies are quite high today ranging between 95%-98%. But even if theoretically both modules and inverters operated at 100% efficiency, the overall system efficiency would still only range between approximately 75% and 85% depending on the circumstance. These losses have nothing to do with the weighted or CEC efficiencies. The energy, while there, is not properly harvested out. So I would prefer to rephrase the question to how inverters can contribute to improving the efficiency of the system as a whole. Traditional inverters face shortcomings in MPP tracking abilities. The central tracking unit can track the ups and downs of the power curve of an array of modules and pull the maximum available current from this array. In a single current, that is, calculated as the average of all modules. In practice, every module provides maximum power at slightly different current and traditional inverters can’t single out the power curves of individual modules. As a result stronger modules are hampered by weaker modules. This is what we refer to as module mismatch loss. Here exactly lies an important topological problem and barrier to improving system efficiency. And really, the only way to overcoming this is by applying MPP trackers for each module individually, which is what we do.

You say that in traditional systems, weaker modules pull down the output of stronger modules because of the mismatch between them. How high is the loss caused by module mismatch?

In PV systems with no shading whatsoever, where modules’ real capacities deviate by a standard ±3% from the nameplate capacity, losses constitute about 2%. This energy retrieved for a 1 MW plant in Italy, for example, where the feed-in tariff is about US$0.4/kWh, would translate into more than US$10,300 in revenue in the first year of operation alone. The mismatch situation is amplified when some of the modules are shaded or soiled as their real capacities now decreased way below the nameplate capacity. Mismatch is also amplified over time when modules age at different rates given different responses to weathering.  

The second problem we experience with central MPPT tracking units in traditional inverters is that they have difficulties to detect the power fluctuations fast enough and can get stuck on so-called local, meaning not the highest, peaks. Usually this happens on these half sunny half cloudy days, on which clouds come and go by quickly. Actually, this is a fascinating thing to watch on the monitoring portal. Our monitoring portal shows the output curve of every module at any given moment and you can see how the shadow of the cloud slides across the array causing module output to rise and drop as if a wave was running through them.

What is your vision of the inverter of the future or should I say system of the future?

I really believe that the way forward is to bring intelligence to all system components in order to make the energy harvesting process better and PV systems more consumer-friendly. This is also why in addition to MPPT per module other features like monitoring of module performance and safety functions gain growing importance. By the way, module companies have recently started to integrate our intelligence into their junction box directly in the manufacturing process. More and more modules will soon have their own individual MPP tracker already included and I would like to spell out what this means. Module manufacturing companies are actually starting to take on some of the responsibilities originally in the hands of inverter companies. The pioneers in bringing such products to the market include Solon SE, Ritek Solar, Helios Energy, and GE Solar. In addition, junction box manufacturers are also heading in this direction, giving module manufacturers the choice to order junction boxes with certain smart features.

And─just something to think about─the inverter working with smart modules doesn’t really need to include an MPP tracker any longer, or at least not a very sophisticated one. It should however, have improved communication features which are able to receive, store and transmit the performance data sent by the modules. As a manufacturer of both power optimizers and inverters we were lucky that we could adjust our inverters accordingly.

So, this kind of restructuring as well as a closer cooperation between the manufactures of the different components is definitely something we should look forward to. At the end of the day, the system owner buys a complete PV installation and should not need to care about the problems of one component handling the other.

The story of SolarEdge begins as a startup company. Within four years you developed a certified and mass produced product. Today, you are closing multimillion-dollar sales deals. In your opinion, which roles do and will startup companies continue to play in the inverter industry?

By definition, startup companies come up with new ways of how to address shortcomings in existing technologies. The solar industry and the inverter industry in particular are known for being rather conservative. I am generalizing here but I am talking compared to, for example, the online or telecommunication industry. Startups, therefore, play an even more crucial role in pushing the industry forward. At the same time it is much more difficult to gain acceptance. At SolarEdge, we realized that the mismatch, shading, soiling and tracking pace problems remained unaddressed by the big inverter companies and we engineered a product which solves all these problems through maximum power point tracking at the module-level instead of the array-level. The idea resonated so strongly it instigated a whole new sector that summed up in market research reports under power optimizers and micro-inverters. Our sector is expected to grow to 1,400 MW in 2012, to constitute around 11% of the total PV market by 2014 and to reach a value of US$1.3 billion by 2015.

Any advice you wish to give to new startups in the field?

Make sure to commercialize your technology fast enough to reach a good head start before competition shows up. If nobody else competes with your idea after a while, drop it.

How was SPI for you?

Thank you for asking. We launched our new US product line. Also, our U.S. office is growing so it was a good opportunity for our new hires to warm up and get familiar with the SolarEdge feel.

And what would that be?

Well, as a SolarEdge employee, you better have a good sense of humor because we have a policy which says that you need to laugh at least once during the day at work. I feel personally responsible for getting this task done. My title may say VP Business Development and Product Strategy, but what I really like to do is go around the office to make the employees laugh. Sometimes they make me cry in return. Now seriously, when you experience tense moments together and sometimes spend a lot of time at work together because you need to finalize an urgent assignment, it is necessary to be able to disperse the stress a little bit in order to get through. The relationship between the executives and the employees is quite informal by nature which makes our work environment cooperative and real. This is characteristic for startup companies and we try to maintain it as we grow. I encourage my employees to disclose what they really think. There is no right or wrong input, only good persuasion. Our employees are filled in on the company’s general targets and are able to work towards them independently and in cooperation with each other. Some of our most successful projects were initiated bottom-up.

Jeanny H. Lim is Editor-in-Chief of InterPV. Send your comments to swied@infothe.com.

ⓒ2011 www.interpv.net All rights reserved.