Toledo-PV: the Oldest Solar >1MW Power Plant In the World

Toledo-PV was the photovoltaic power plant that inaugurated the power scale of the MWs in Europe. It began to operate in 1993, so it has already completed 23 years of operation and, standstill better, good operation, as recent article on the operating experience of previous years shows: «Toledo-PV Plant 1MWp and 20 Years of Operation»


Toledo-PV: The Oldest Solar 1MW In the World has been Operating for 23 years 1993/2016 Abstract/Summary: The paper is a summary of the energy production during the 20 years, 1994-2014, of operation of the photovoltaic plant, Toledo-PV. The photovoltaic power plant, located close to the “La Puebla de Montalbán” village In Toledo (Spain). Toledo-PV plant with a nominal power of 1 MWp was inaugurated in July 1994 and has operated satisfactorily since that time providing electricity to the Spanish grid. The plant is owned equally by three companies: Enel Green Power, RWE and Gas Natural Fenosa.

Author(s): M. Alonso-Abella, F. Chenlo, A. Alonso, D. González

Keywords: Evaluation, Large Grid-Connected PV Systems, Performance


Subtopic: Operation of PV Systems and Plants

Event: 29th European Photovoltaic Solar Energy Conference and Exhibition

Session: 5BV.1.44

Pages: 2728 – 2733

ISBN: 3-936338-34-5

Paper DOI: 10.4229/EUPVSEC20142014-5BV.1.44

Price: 0,00 EUR

Document(s): paper, poster

Some years earlier there had been another power plant with MW power in the USA. But it used conventional photovoltaic modules in combination with V-shaped side mirrors, which practically doubled the incident irradiance. Under these conditions, the modules degraded rapidly, losing 40% of their power in four years of operation, which led to the dismantling of the plant.

Electrical Degradation of the Carrisa Plains Power Plant

Author(s): Schaefer J.EPRI, Palo Alto, CA, USA

Rosenthal A. Southwest Technology Development Inst., Las Cruces, NM, USA

Schlueter L. Siemens Solar Industries, Camarillo, CA, USA

WengerH. Pacific Gas and Electric Co, San Ramon, CA, USA

Book Title: Tenth E.C. Photovoltaic Solar Energy Conference

Book Subtitle: Proceedings of the International Conference, held at Lisbon, Portugal, 8–12 April 1991

ISBN: 978-94-010-5607-6

DOI: 10.1007/978-94-011-3622-8

Print Online ISBN: 978-94-011-3622-8

Publisher Springer Netherlands

Copyright Holder ECSC, EEC, EAEC, Brussels and Luxembourg

Some years earlier there had been another power plant with MW power in the USA. But it used conventional photovoltaic modules in combination with V-shaped side mirrors, which practically doubled the incident irradiance. Under these conditions, the modules degraded rapidly, losing 40% of their power in four years of operation, which led to the dismantling of the plant. Thus, Toledo-PV is the oldest photovoltaic >1MW PV alive utility in the world and the first one that exceeds 20 years of useful life, a ribbon that, although nobody knows very well why, is usually interpreted as conclusive of technological maturity.

Both attributes, antiquity and reliability give Toledo-PV the venerable condition that, coupled with the fact that in its construction concurred some first-fruits of interest, make of Toledo-PV a project deserving of the historical name, whose event deserves more celebration of the Which he has so far. (The first story was published on September 5, 2016).

At this point it is important to insist that this lack of celebration reaches not only the scope of the rumor that could be made with the banner of “… I am the one who has the oldest PhotoVoltaic MW in the world …”. More regrettable is that this lack extends to the field of scientific production. Toledo-PV is a project designed, executed and operated by private companies (Endesa, RWE Energie AC and Unión Fenosa, coordinated by the latter) which, as such, have all the legitimacy not to advertise information that could constitute a competitive advantage for them. But Toledo-PV is also a project that was financed exclusively with Public money and from that, as such, one would expect returns in knowledge of general interest. The aging of the photovoltaic modules is one of the aspects on which Toledo-PV should have generated public and rigorous knowledge. It is not possible at this point to imagine a conflict of interest, since neither of the two types of modules used in its construction – Saturn modules of BP Solar, and modules MIS of Nukem – is realized in the current market. However what has been published about Toledo-PV is minimal and disappointing. After a series of articles published in the years 1994 and 1995:

1st 1MW Solar PhotoVoltaic Power Station In Spain -Plant Description & Gained Experience During Construction- Toledo-PV

Contributors: Hill, R. / Palz, W. / Helm, P.

Conference: European photovoltaic solar energy conference; 12th; 1994; Amsterdam

Size: 4 pages

ISBN: 0952145243, 0952145235

Type of media: Conference paper

Type of material:Print

Language: English

Keywords:  Photovoltaic solar energy, solar energy

100 kWp tracking PV plant at the Toledo PV Project,

Jiménez C, Lorenzo E, Yordi B. Proc.12th European PVSEC, Amsterdam, 1994; 822-824.

The Data Acquisition System for the 1MW PV Plant in Toledo, Spain. 

Mukadam K, Chenlo F. Proc. 12th European PVSEC. Amsterdam 1994; 818-821.

PV modules and Array Testat Toledo PV plant.

Lorenzo E., Zilles R., Proc. 12th EU PVSEC, Amsterdam 1994, 807-809.

Operational Results of the 100 kWp Tracking PV plant at Toledo-PV Project.

Lorenzo E., Maquedano C., Valera P. Proc. 13th EU PVSEC, Nice (1995), 755-758.

The 1 MW PV Photovoltaic Plant in Toledo, Spain. First Operational Results and Operating experiences, Mukadam K., Chenlo F., Proc. 13th EU PVSEC, Nice (1995).

All these old publications, created in the heat of the constructive impulse and basically describe aspects related to the construction of some of its elements, such as the trackers or the data acquisition system, there have been only two communications to the European photovoltaic congress, one in 2005 and another that opens here the list of references.

There is here a double pity. On the one hand, for the loss of what could have been learned in terms of resistance to weathering materials that have a presence in the current technological landscape, such as encapsulation polymers, oxides of the AR layers or metals of the electrical contacts of the cells. On the other hand, because the absence of rigorous information pays for the flourishing of another, more rumorous than rigorous, which, based on only half-truths, conveys frankly erroneous messages.

A paradigmatic example of the latter is the recent publication in journalistic media [1], sometimes more prone to rumor than to rigor, of the news that says that the original modules of the Toledo-PV plant have been replaced by new ones “Retrofit“. The news, true in itself, is published together with comments suggesting, first, that the original modules of Toledo-PV had been much degraded – “… in the last study, carried out in August 2015, the effective power had been reduced by 37%” – second, that the current modules are technologically much superior to the original ones – “… these are modules with a new technology that produces 20% more in unfavorable conditions Of ‘mismatch’ caused by shadows, dirt, aging, temperature gradients, etc. – and, thirdly, that the fact of changing the modules has made it possible to learn a lot – “… this ‘revamping’ has allowed us to know, with real and quantifiable magnitudes, that the useful life of a photovoltaic plant is 22 years …”.

Well, it is easy to understand that none of these three things honors the truth. What appears to have actually occurred with the original modules of Toledo-PV was not natural degradation but a consequence of a manufacturing failure in an internal welding of the modules. The difference is a lot, because the natural degradation is general and irreversible, whereas a welding failure is particular and reversible. Thus, the news attributes a quality to the general without more support than a particular circumstance. Something like 1) Pepe has mumps, 2) Pepe is man, ergo 3) All men have mumps. If Aristotle raised his head! That a 37% power reduction is not a general health condition but a particular disease is well done without comparing the figure with the many others available in the literature and even with what was published on the same Toledo-PV shortly before , In the aforementioned reference for the year 2014, where it can be read that in July 2012 (ie 20 years after being put into operation), the modules had not degraded more than the equivalent of 0.2% per year, In the case of modules supplied by BP and 1% per year, in the case of those supplied by Nukem.

What is more, the news says that change the modules (operation that the author of the news calls ‘revamping’, which is still indicative of scorn, because ‘vamp’ in English means rather better than substitute) has made it possible to know the useful life of a photovoltaic plant “with real and quantifiable magnitudes”. In view of this, at least until other more complete information on such magnitudes has been published, that the logical consistency of the link between what is done (the change of modules) and what is learned (the useful life of the plants) is not greater than Of stating that by changing the spare tire of the car, after suffering a puncture, much is learned about the wear resistance of the rubber with which the tires are made. Finally, the technological improvement that the news attributes to the new modules can be put in solfa its wording, when it says that the new technology produces “20% more” but does not say what is the reference for this more. The question is, More than what? Of course it can not be the technology of the solar cells in itself, since the efficiency of the new ones is substantially similar to the previous ones. In this same order of things, a query to the official website of the project ( does nothing but increase disappointment. There is not even a mention of the recent change of modules! In short, nothing further from our intention than criticizing news of the press, let alone to those who do honestly fulfilling their work. But this does not prevent the conviction that Toledo-PV deserves much more and much more rigor than has been published until now.

The IES-UPM was lucky (and never better, because the opportunity was given and without doing anything for it) to participate in the construction of Toledo-PV. To put here in writing the memories of that participation is his way of contributing to extend that celebration. It is also their way of honoring the duty to record the firsts in which they had something to do, for information of those interested in the history of solar energy. And it is finally his way of thanking to the life the exquisite fortune of having participated in Toledo-PV.

They have not had any official relationship with Toledo-PV since 1995, so nothing can report on the lessons learned there over their more than 20 years of operation. The cause of the problem that led to the change of modules came to them via radio and non-existent delator. From what they heard, they made the diagnosis, aided by the fact that they had already encountered a similar problem in other plants with similar modules [2] [3]. They may even have mistaken a plan. Then, they would certainly be very grateful if someone who has more and more compliant information takes the trouble to report their error and provide other more truthful information here.


Toledo-PV project was a project with an investment of approximately 10 M€ (then they were called Ecus, and still were not emitted in paper) that was covered with public subsidies contributed by two Programs (Joule and Thermie) of the European Union, by the German Ministry of Science and Technology (BMFT) and by the Spanish Electrotechnical Research Program. The proponents were Unión Fenosa, Endesa, RWE, BP Solar, Nukem and WIP.

The project was structured around two main lines: the use of photovoltaic modules of last generation and the participation of electric companies. The declared objectives were, on the one hand, to demonstrate the technical feasibility of the technology and, on the other hand, to involve the major players in the conventional electric sector, to pave the way for future photovoltaic penetration in the network. The price of photovoltaic systems was of the order of 10€ /W, which resulted in electricity generation costs of more than €1/ kWh [4]. Because this cost is at least 15 times higher than generating electricity using conventional technologies (water, coal and nuclear), very few believed in the possibility of photovoltaic technology getting to where it is today. Here comes as a finger to quote Machiavelli’s wise assertion about “the unbelief of men, who-in fact-never believe in the new until they acquire a firm experience of it.” [5]

THE SITE: 39° 49´51.65″N 4° 17´52,2″ W; hasl: 430m

Toledo-PV is located in lands of the reservoir of Castrejón, in the Tajo, near La Puebla de Montalbán. It is a medium-sized reservoir (S=750 Ha & 45 Hm³ of Volume) that feeds the hydroelectric power station of “Carpio del Tajo”, of 8 MW, and on whose banks you can enjoy the extraordinary landscape of the “Barrancas de Burujón”. Demonstration projects are always a fertile ground for finding reasons, including pilgrimages, to justify their spending. In this case, and although we do not know if came to appear in the papers, there were repeated mentions to the possibility of taking advantage of the coincidence of the reservoir and the photovoltaic power station to study the complementarity of the sun and water as energy resources. Neither such a study came to fruition nor does it seem to make sense. Happens that the flow of water through this reservoir depends, of course, on the rains in the river basin, but also and especially on the irrigation needs in the area, which is the priority application of the reservoir water.

PV GENERATORS: Saturn and MIS-IL Technologies

The 1-MW of Toledo-PV is divided into three photovoltaic generators: two static 450 kW each (nominal value) and one with horizontal axis tracking and power of 100 kW (figure 1).

The cells of the modules that make up one of the large generators and the small generator are of Saturn technology, manufactured by BP; And those of the modules that constitute the other generator are of MIS-IL technology (metal-insulator-silicon inversion layer), manufactured by Nukem. Both technologies were then presented as “state of the art”. The encapsulation of the BP modules is totally conventional (glass-EVA-Tedlar), whereas the one of the modules Nukem presents some variation (glass-resin-glass).

Saturn technology introduced the novelty of “buried contacts” in laser-cut grooves on the surface of cells. The technology had been developed in Australia by the team of the University of New South Wales, directed by Martin Green. BP had bought the patent and opened a factory in Spain, in Tres Cantos, for its industrialization. In addition to reducing the surface of the metallized plug-in cell, which is inherent to the burial of the contacts (Figure 2-b), this technology significantly reduced the emitter’s doping, compared to conventional technology. The air gap resulted in an increase in the short-circuit current and the reduction of doping in increasing the open-circuit voltage, thus achieving the highest efficiency of silicon cells at that time: 18% compared to 14% offered by the conventional.

Nukem’s MIS-IL technology had been proposed in Germany by Professor Hezel from the University of Erlangen [6] and characterized because of the PN junction, which generates the internal electric field separating voids and electrons, instead of being created by the  conventional process of diffusing phosphorus impurities (emitter “N”) onto a silicon wafer with boron impurities (base “P”), was induced by the deposition on the front surface of the “P” wafer of a very thin layer ( Tunnel) of silicon oxide incorporating positive electric charges (Figure 2-c). The result was a cell with a very low-conducting emitter and which had to be accompanied by a particularly delicate metallization grid (many fingers at the same time very thin) to compensate for the negative effect of the low conductivity of the emitter on the cell’s series resistance . This “Induced Junction” technology allowed the cell manufacturing process to be carried out entirely at temperatures below 500°C (in contrast to diffusion, which requires approaching 1000°C), which supposedly offered some advantages for the manufacture of solar cells on thin wafers. The foregoing paragraphs are written in the past, because these innovative cell technologies used in Toledo-PV have in common that none of them are currently in production, so that “last generation” is almost an invitation to jest. However, this does not tarnish the heavy weight of this project in demonstrating the durability of photovoltaic technology, since the non-innovative components of the technology, namely the use of crystalline silicon as a base material and the techniques of Encapsulated modules, also present in Toledo-PV, are the main responsible for this durability and maintain their validity in the current industrial landscape. The technology of the cells was the most visible but not the only novelty relative to the photovoltaic modules that Toledo-PV attended. So was the quality control of its power. As IES-UPM states in EraSolar, Toledo-PV was the first project to establish a contractual relationship between the amount paid to suppliers of photovoltaic modules and the results of the characteristic power measurements carried out by an independent body.This, which is fortunately today a widespread custom, had to face, then, not a few resistors, rooted in the practice of supplying modules whose actual power was significantly lower than the nominal power. Surely it is a crowd who can rightly feel that they has done much to eradicate this harmful practice. But with the calendar in hand, it’s easy to see that the scoop took place in Toledo-PV. The IES-UPM was the one who proposed and carried out the quality of modules. The power measurements were made against real sun and by comparison with previously calibrated by CIEMAT references. The following is a brief history of what happened.

STATIC GENERATORS & 1-Horizontal Axis Sun-Trackers

The original Toledo-PV design was only generators mounted over 30° static structures. Simple and robust, the solar radiation captured by these structures is limited by their own statism. Toledo-PV’s are made of galvanized steel and employ approximately 40 kg of steel and 0.1 m³ of cement (in the shoe) per square meter of generator [7]. In today’s plants, piles driven by hammer blow are more frequent than the shoes, but with respect to the support structures themselves, the figure of 40 kg of steel per m² still indicates very optimized designs; Although with a certain downward tendency, as a result of which the inclination angles are also somewhat lower, rather in the environment of 20 than of 30 degrees. Later on and at the proposal of the IES-UPM, the project included a follower of a horizontal axis. The followers, of course, increase the uptake of solar radiation, but at the price of incorporating moving elements that usually arouse suspicions about their reliability.

Monitoring on a horizontal axis represents a good compromise between reliability and radiation uptake. Because it is horizontal, the axis can be very long (it does not rise from the ground as it grows) and can be strongly anchored to the ground at regular intervals, as with static structures. In fact, the eyes of the uninitiated often have difficulty distinguishing prima facie between static structures and horizontal followers. The axis is oriented in the direction NS, so the tracking is such that ideally the normal to the surface of the generator describes daily a semicircle that starts pointing east (in the morning) and ends pointing west (in the afternoon) , And the angle of rotation is adjusted at each moment so that the plane containing the axis is normal also contains the Sun. Strictly speaking, in a field of several followers and if the movement were just so, there would be important shadows both by the Morning and afternoon, when the angle of elevation of the Sun is relatively low. To avoid this, the tracking strategy incorporates a “back-tracking” algorithm, which diverts the angle of rotation from the ideal position to avoid shadows. By exerting a little visual imagination, the reader should understand that, with this algorithm, the 1Haxis-tracker dawns and lies horizontally. Solar trackers with horizontal axis are an invention prior to photovoltaic cells.

The first practical design [8] was proposed in 1884 by John Ericsson, a Swedish engineer who a few years before had achieved world renown for inventing the propeller for naval propulsion. And the solar pump that Frank Shuman designed and installed in Maadi (Egypt) and which operated regularly for almost 5 years (providing a mechanical power of more than 50 HP) consisted of 5 solar collectors each 60 meters long by 4 meters Wide and 8 meters apart. These collectors, which were thermal and concentrated (such as the parabolic trough collectors of the current solar thermal power plants), were a very important milestone in the history of solar plants. A few years after the Maadí solar bomb, the World attended the deployment of an impressive infrastructure for the extraction and distribution of coal and oil, which managed to make these fuels very cheap even in places far from mines and wells, opening a parenthesis in the history of solar energy, which did not resume until, in the 1970s, the first sounding of the oil crisis sounded, whose resonances spread throughout the world in the form of concern for oil-dependence Industrial society. A very well-known manifestation of this concern was the PSA (Solar Platform of Almería), in which thermal collectors were also equipped with monitoring in a horizontal axis. And this is related to Toledo-PV because they were, in fact, the same engineers who had designed the trackers structures that makes the motor torque in the very large tracker axis, allowing to move loosely the 200 m² of the collector, ready for extreme wind situations. In addition, the high multiplication factor also allows to easily control the tracker angle [9], without counting the laps of the motor (1 lap of the motor corresponds to half a minute of degree, which is a huge precision for the requirements of this application). The Toledo-PV tracker was manufactured by Jupasa, a Spanish company specialized in the manufacture and assembly of metallic transformations of great dimensions and high precision. Nowaday the oldest living PV-Tracker in the world and still enjoys excellent health, which saves more comments on the goodness of his design. The control system was developed by the IES-UPM. The following is a brief history of this development.

PV Inverters: Tyristhors and IGBTs

DC/AC converters are widely used power electronics equipment: AC motors, power supplies and, most recently, photovoltaic inverters.

The heart of these equipments is constituted by a block of switches, or switches, whose ordered action allows to obtain pulse trains from dc. How these switches are to a large extent defines technology, and Toledo-PV coincided with the dawn of a transition in the field of these devices: the thyristors, which at that time were the mambo dance kings, were replaced by the faster IGBTs, notably easier to handle. Thus, in 1994, Toledo-PV included two inverters with thyristors of 450 kW each, associated with static generators, and an inverter of 100 kW IBGTs, associated to the generator with horizontal tracking. All of them were manufactured by Enertron, and there was the remarkable circumstance that the 100 kW inverter was for some time the European size record in power electronics equipment with IGBTs. All three had a lot of prototypes and their operation and maintenance associated the difficulties of this condition, so that some 7 years later they were replaced by inverters of IGBTs that were already standard equipment in the market. This change is not discussed here in greater detail, to understand that the traffic from the thyristors to the IGBTs was a process that was developed in the general scope of power electronics and in which photovoltaic did not have a particularly relevant role.


[1] E. Lorenzo, R. Zilles. Era Solar Nov-Dic 2016

[2] Progress in Photovoltaics: An investigation into hot-spots in two large grid-connected PV plants.

Muñoz J., Lorenzo E., Martinez-Moreno F., Marroyo L., Garcia M.,

Research and Applications 16 (2008), 693-701. DOI:10.1002/pip.844

[3] Progress in Photovoltaics: Observed degradation in photovoltaic plants affected by hot-spots.

Garcia M., Marroyo L., Lorenzo E., Marcos J., Pérez M., Research and Applications 22 (2013), 1292-1301.


[4] Toledo-PV official web: does mention the figure of 1,1 €/kWh in this project; don’t worth noting as this website has not been updated since 1998, which seems to be a symptom of the scarce celebration for more than twenty years ago.

[5] N. Machiavelli, The Prince, chapter VI.

[6] R. Hezel, R. Schörner, J. Appl. Phys., vol 52(4), pp 3076-3079, 1981.

[7] Estimated from data contained in: M. Alonso et al. 1 MW PhotoVoltaic Power Station Toledo/Spain – Plant Description and Gained Experience during Construction – Toledo-PV. 12th European Photovoltaic Solar Energy Conference, 1163-1166, Amsterdam (1994).

[8] K. Butti y J. Perlin, A Golden Thread (1980). (There is a castillian translation: “Un hilo dorado: 2.500 años de arquitectura y tecnología solar”, Ed Blume, 1985).

[9] The design of this tracker included two pulleys of almost one meter in diameter, separated almost 100 meters, strongly anchored to the ground and through which a thick steel cord passed, which transmitted the rotation of the pulleys to the axes of the follower’s collectors . This design had a defect that proved fatal: this steel cable, which by its very function had to be well-tensioned, easily oscillated (like a guitar string), causing considerable mechanical damage.


Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s