The Future Of Solar Doesn’t Track The Sun – Terraform Industries Blog

PV modules are cheap enough today that the simple fixed East-West arrays are cheaper and faster to install than the industry’s darling, the single-axis tracked array.

Xavier Dedenbach, Terraform Power Electronics Engineer

Originally posted on April 20, 2025.

This article takes you on a journey to explore whether East-West fixed solar arrays are now more economical than the dominant variant, single-axis tracked arrays. If you just want the answer, feel free to jump to the end or read this summary. But if you have 15 minutes to follow along, the juice is worth the squeeze.

We’re hiring at Terraform!

💡Today’s Claim
PV modules are cheap enough today that the simple fixed East-West arrays are cheaper and faster to install than the industry’s darling, the single-axis tracked array.

Arguments Against

• Single-axis tracked arrays still significantly outperform East-West fixed arrays in energy production on a per-panel basis.
• There is one weather pattern (severe hail) so destructive that I would choose the more expensive single-axis tracked array if I experienced it regularly.

Arguments For

• East-West arrays require less material and labor to install.
• East-West arrays use substantially less land for the same power output.
• As PV modules become cheaper, the economic benefits of East-West arrays get even better.

Conclusion

• If price per watt delivered were the only important factor, East-West arrays would be the clear choice everywhere.
• However, nature is metal, and in some regions, we currently lack sufficiently resilient panel technology to support East-West arrays.

Like many technologies, photovoltaic (PV) cells, which convert sunlight into usable energy, started out expensive and uncompetitive. They had to compete in one of the most competitive industrial sectors: the energy sector, which is filled with behemoths like coal, natural gas, and hydropower. On top of that, photovoltaic technology was handicapped by limited daylight hours and an above-average vulnerability to weather conditions.

PV systems did not have it easy, but a collective push from green energy companies, governments, and their citizens willed them into existence. They shared in the renewable energy promise to build an economy powered by sunlight, wind, water, and atoms.

So began the decades-long push to improve module efficiency in every way imaginable: new materials, new PV cell architectures that convert more photons to electrons, and tracking systems that keep modules aligned with the Sun’s position. These trackers gained industry dominance by offsetting their costs through improved efficiency of the PV modules.

Today’s story examines what happens when the very components being optimized, the PV modules themselves, become cheaper than the technologies designed to improve their efficiency.

Tracking technologies are moderately complex, requiring actuators, sensors, stronger structures, and additional energy storage. These technologies fall into two camps: single-axis and dual-axis systems. Single-axis trackers handle intra-day solar movements, following the Sun as it rises in the East and sets in the West. Dual-axis trackers adjust for the array’s latitude, the Sun’s intra-day movement, and inter-day movement, due to Earth’s orbit 1. Unlike single-axis trackers, dual-axis solutions haven’t seen mass adoption due to more complicated mechanisms and increased spacing requirements between rows to prevent shading. For this reason, we’ll focus solely on the costs and benefits of the single-axis tracker, the current king, if you will, and its relation to its much simpler cousin, Earl. Ugh… I mean the East-West array.

The Rapid Decline in PV Module Prices

Why ask this question? When I took over the solar and power systems at Terraform Industries, I was asked to architect a photovoltaic system that capitalized on one of the most remarkable learning curves in modern history: a 44% cost reduction for every doubling of cumulative PV module production. The East-West array was a clear winner, despite not being highly adopted in the United States. I began to investigate why, and this article is a byproduct of this effort.

Starting with PV module costs, PV benefited from the trifecta of expanding economies of scale, increased competition as more suppliers rushed into the market, and increased supply as high-grade silicon became abundant due to the demand for PV modules and integrated circuits, aka chips. That level of cost reduction is hard to compete with. In the 2010s, the cost of PV modules fell by 85%, and so far this decade, they have fallen an additional 58%. All this in the face of the worldwide epidemic of increased inflation in recent years.

Table 1: Module Price Reduction 2010-2025

Over the last decade, trackers rose to dominance, and by 2015, they accounted for more than 50% of the cumulative installed capacity for photovoltaic systems. At that point, they had usurped the previous dominant array type, the south-facing fixed array, to become the new king. Today, they continue to dominate newly installed capacity by a wide margin.

As trackers productized and their manufacturing scaled, they achieved massive cost reductions of 85% in the 2010s. But trackers have a price floor due to the large amount of steel, actuators, and sensors they require. In the 2020s, they have seen a 17% reduction in cost, compared to a 58% reduction for PV modules, making trackers an increasingly dominant line item in the overall budget of single-axis arrays.

Table 2: Tracker Price Reduction 2010-2025

One must stop and wonder: are trackers still worth keeping around, as the very modules they were designed to improve in efficiency fall to near cost parity with the tracking systems themselves?

A Technology in Transition?

Photovoltaic technologies are no longer a fledgling idea. They have grown into a robust and scalable technology that delivers increasingly significant amounts of energy. We can compare them to the rise of the computer, the Internet, or mobile.

If we take the internet as an example, there were technologies along the way that filled in initial shortcomings, which were no longer necessary once it scaled. Netflix’s mailing of DVDs made much more sense before the proliferation of fiber and high-speed internet that led to the streaming revolution. I am here today to explore the possibility that trackers may one day be thought of this way, as a stop-gap solution that will no longer be necessary.

Single-axis Tracked Arrays 👑

Single-axis tracking systems are essentially long steel torque tubes, motors, sensors, and algorithms that work together to align the PV module with the incoming direction of sunlight. The most popular of these systems rotates from East to West following the Sun’s position. To support these forces of rotation and resist wind loads, the torque tubes are long and heavy. They mount onto deeply driven piles, piles are to the earth what nails are to wood.

Tracking systems function by minimizing the angle of incidence between the Sun’s position and the direction the PV module is facing, later referred to as the PV module unit vector. Ensuring each PV module receives maximum exposure to sunlight, and expanding the power profile by capturing more energy in the morning and afternoons.

The components of the single-axis tracker

• The foundation holds the array to the Earth, typically using piles, which are large steel posts driven into the ground.
• The fixed structure holds the sensors, the fixed side of the actuator and bearing system, and the tracker power system. Modern systems, like those of Array Technologies or Soltec, can be installed over minimally modified terrain, which significantly reduces the cost and time of ground preparation.

• The bearing assembly affixes the rotating structure to the fixed structure.

• The actuator drives the rotation of the rotating structure. It can be a linear actuator and lever arm, or a rack and pinion system. These actuators are sized to withstand lift forces from windy days and have sufficient speed to quickly stow when faced with adverse conditions.
• The rotating structure holds all the PV modules of the row at the same angle, including components such as the torque tubes—the main tube, and the purlins—the arms that hold the solar panels.
• The alignment system processes data such as direct irradiance, time, wind speed, weather conditions, and location to determine the correct angle of the modules. It includes a controller, multiple alignment algorithms, and some electro-optical sensors for detecting sunlight intensity.
  • Solar position tracking: aligning the module face with the incoming sunlight, used during clear days.
  • Backtracking: deviating from the solar position to collect more sunlight, such as lying horizontally on cloudy days to capture ambient light or moving to avoid shading from one row to another during early mornings or late afternoons.
  • Stowing: moving the modules to a safe position during an adverse weather event. This includes a near horizontal position on high wind days or away from the direction of hail during a hailstorm.
• The tracker power system (not pictured) contains the energy to drive the actuators of the single-axis trackers, including independent PV modules, charge controllers, and battery packs.

The tracking system’s competitive advantage is to enhance the effectiveness of each solar panel, requiring fewer of them to achieve a certain target power. However, this is achieved at the expense of higher structural complexity, installation times, and land consumption per panel.

East-West Arrays

Despite being a simpler system, East-West solutions are relatively new on the utility-scale solar scene. Previously, PV modules were so expensive that East-West arrays were not economical due to their lower energy collection per panel. However, they eliminate a ton (literally) of material and assembly time from the system. Solutions, such as the Jurchen PEG system, use the PV module as the primary structural component, which removes 90% of the typical structure of other solar array variants.

The array scales in both directions, like a mesh network of panels, held together by clamps and supported by rebar. This network of PV modules doesn’t have the same risk of shading each other, due to low tilt angles, so they can pack 250% more installed power into the same space when compared to a single-axis array.

East-West arrays are simple. They consist of parallel strings of PV modules that are oriented in opposing directions, one facing East and the other West. The current of the whole array is the summation of these string currents, effectively letting East-West arrays capture sunlight from dawn till dusk, similar to a tracked array.

The energy profile of the East-West array depends on its tilt angle. The more aggressive the tilt, the larger the difference in time between the East and West peaks, but the more that one string shades its adjacent neighbor. The ideal angle extends the time of near peak solar irradiance by increasing the time between the East-facing peak and the West-facing peak.

The components of the East-West structure

• The substructure of the Jurchen PEG array is rebar rods driven into the ground, held in by the friction between the rebar and the earth. There is a bottom plate crimped at the base of the rod to prevent the rod from sinking into the earth, and a “top hat” on the rod to clamp the panel onto. That’s about it. Rooftop East-West array structures can be even simpler, using bent sheet metal pieces or custom molded bricks.
• The modules themselves.

These systems require minimal land prep, reduced land usage, shorter installation time, and lower material consumption. The question is, can these savings make up for using more solar panels to achieve power parity with their tracked cousins?

Number of Panels Required For Equivalent Power

It is important to note that the ratio between the size of the single-axis tracker and the East-West solar array varies from location to location, depending on latitude and weather conditions. Understanding this requires simulation to account for the impact of changing sunlight vectors and the effects of weather on array performance. To learn more, check out this supporting article where I step through the math to show the relationship between the angle of the panel and the angle of the incoming sunlight, as well as the underlying reasons why single-axis trackers are more efficient.

The baseline performance we were looking to achieve from each array is 1 megawatt of DC power. We’ll simulate the performance at four locations in the US that differ.

• Denver, Colorado (Cold and Dry Climate)
• Mojave, California (Hot and Dry Climate)
• Buffalo, New York (Cold and Wet Climate)
• Houston, Texas (Hot and Wet Climate)

The assumptions I made during this are as follows:

• The array voltage is 1500 volts, consisting of 27 panels in series.
• Panel size is 550 watts.
• Panels are 21.5% efficient.
• Changing the array size can only be done by adding or subtracting parallel strings.
• The power target for each array is 1 megawatt of DC power, defined as an array that averages 8 megawatt-hours of energy per day across the year. We assume all power generated over 1 megawatt is unusable.

Important concepts before we get started.

The energy available to photovoltaic systems depends on the Sun-Earth relationship and the specific location of the system on Earth’s surface. Before we start, let’s get a couple of the basics of the Earth-Sun relationship down:

• The Earth’s orbit around the Sun is slightly elliptical, leading to variations in the Sun’s position in our sky by a certain hour from day to day. The equation of time corrects for this variance.
• The Earth rotates on a 24-hour schedule.2
• The solar constant measures the power per area (solar irradiance) of sunlight as it hits Earth’s atmosphere: 1,361 Watts per square meter.

The amount of sunlight available is determined by the relationship between the local position on Earth, the orientation of the PV module, and the solar position vector, based on the relative positions of the Earth and the Sun.

• The solar position vector magnitude is the solar constant, with a direction determined by the time of day and the Sun’s declination angle, determined by Earth’s axial tilt and orbit.
• The local position vector extends from Earth’s center to the PV’s position on Earth’s surface, which is determined by its latitude, longitude, and elevation.
• The PV module vector is based on the azimuth and module tilt.

The solar position vector and local position vector both affect the way sunlight appears at any one spot. The latitude of the position and the declination angle determine the elevation and duration during which the sun appears in the sky. A shallower angle, common in winter in the Northern hemisphere, will lead to a lower elevation and a larger percentage of Earth’s 24hr rotation with the Sun appearing below the horizon. This scenario is the exact inverse in the southern hemisphere, with increasingly lower latitudes leading to the Sun appearing increasingly more northward at a lower elevation.

Impact of latitude and declination angle

I first investigated performance impacts based on the photovoltaic installation’s position and panel orientation. Both arrays were installed as 1,200 kilowatts DC for an apples-to-apples comparison.

What jumps out is the difference in performance based on the latitude of the array’s location. Buffalo, having a higher latitude, has a more southerly Sun and fewer sunlight hours. Meaning, less time in which to capture sunlight, leading to an increased importance in the ability to capture sunlight in the mornings and afternoons, the strong suit of single-axis arrays. Houston has the lowest latitude of this group and hence sees the smallest variance in performance. For all cities, we see the impact of the declination angle, as performance varies from season to season, since this also affects the duration of the sun in the sky.

Performance including weather

To capture real-world solar panel performance, we must account for clouds and adverse weather conditions. I updated the tool to add weather factors and re-ran the simulation at 1,200 kW for each array type.3

The most immediate impact is in the cities with the most cloud cover. Clouds don’t delete all sunlight, but they do they scatter it, turning direct sunlight into indirect or diffuse light. The best way to catch scattered sunlight is to face horizontally, which gives the highest chance of diffuse sunlight to bounce into each respective PV module. East-West arrays are always ready for high diffuse light periods, allowing them to perform relatively well when clouds are overhead. Single-axis trackers will backtrack to the horizontal position to capture diffuse light once they determine that the sky is cloudy.

The high performance of East-West arrays in inclement weather makes performance in places like Mojave look comparatively worse, where the largest differences are still based on the advantages the single-axis tracker has during early morning and late afternoon periods.

Meanwhile, Buffalo goes from being the largest deviation in performance, when we ignore weather, to the smallest when we account for cloudy days.

Sizing the array

A note on installed capacity—arrays that deliver 1,000 kW of power are often not installed at 1,000 kW. The actual size is referred to as installed capacity, which is defined as the total of all PV modules installed in an array. They are used by solar designers and installers to determine the number of materials and labor required for the array installation. The critical dimension, measured in dollars per watt, by which all arrays are evaluated, uses the installed capacity to determine the system’s wattage.

After accounting for weather effects, the goal was to find the PV array configurations that deliver 1000 kilowatt of DC power for both East-West and single-axis tracked systems. This simulation assumes that a power conditioner or inverter is being used, limiting the power to a maximum of 1000 kW of DC power. 4 It is worth noting that typically larger sized East-West arrays will perform better when used in a direct DC connection where more of the middle of the day peak power can be utilized effectively [5]. The sizing calculations ensure that all proposed configurations deliver annual energy yields within 2% of each other across all test locations, allowing for fair comparison of performance between the East-West and single-axis tracked systems.

Table 3: Installed capacities per city.

If we roll up these array sizes into costs, we see that single-axis trackers are quite effective at saving PV module costs. Which looks pretty good for the single-axis array. Using fewer modules to get the same amount of power is a big benefit of these structures. However, these structures also have increased material, labor, and land costs per installed watt when compared to the simple East-West array.

Table 4: Costs of PV modules per array type based on city.

We will explore these costs in the next sections and ultimately answer the question—are PV modules cheap enough today that array types that use more of them, like the East-West array, are cheaper than those that save on module costs, like the single-axis tracked array?

Balance of system material and construction costs

The balance of systems includes all components beyond the PV modules and inverters. The most relevant to today’s article is the structure itself—whether it’s a single-axis tracker or the simpler East-West array structure that we’ve been comparing throughout this article. Other components include fuses, wires, and all other systems that enable an array to function. There are two types of costs in this bucket: material costs and labor costs associated with assembling and installing those systems. Like all other solar costs, they are typically listed in USD per watt terms. The single-axis array is more complicated, resulting in more parts, which in turn leads to a higher cost of both materials and labor for assembly in the field.

Table 5: Costs of the balance of system per watt.

If we plug these numbers into our array sizes per state, we see that East-West arrays are, in fact, cheaper; they consume less material and much less labor in the field. Skilled installer labor is especially in short supply, even with the boosts in wages to be eligible for IRA tax credits. So the benefits extend beyond just the cost savings, by improving the speed of installation, the team can be smaller, leading to less management overhead and increased delivery speed.

Table 6: Costs of the balance of system for the East-West and single-axis tracked arrays

Land Consumption and Prep

East-West arrays may require more PV modules, but they are far more land-efficient. Single-axis tracked arrays must be spaced out row to row to prevent panels from shading one another. Trackers, depending on the manufacturer, can have maximum tilts of 45-60 degrees towards the rising or setting sun, extending the shadow each row can cast based on its height. The superpower of East-West arrays is their moderate tilt angle, ranging from 8 to 15 degrees, which limits their shading to this tilt angle or less.

At dawn or dusk, when the Sun is just above the horizon, the light cast is considered low productivity. Sunlight at angles below 15 degrees is considered nearly uncapturable due to the effects of the ozone layer scattering sunlight. Therefore, the East-West arrays lose very little opportunity to capture light.

The difference in land efficiency is significant, with East-West arrays able to pack in 250% more installed power (at peak) into the same land area compared to their single-axis trackers or southward fixed facing counterparts. In an area where we install 1 MW of DC power using a single-axis tracker, we can install 2.5 MW of DC power using an East-West array. East-West arrays have a lower cost of land and, more importantly, land preparation.

Table 7: Land consumption per array type

The cost of land varies per state. Solar installations can occur on Bureau of Land Management (BLM) land, converted agricultural land, or M3 heavy industrial land. The most common type is converted farm or pasture land. I’ve listed the costs below.

Table 8: Ag land costs per state.

In general, the cost of land prep is much larger than that of the land itself. These are the common types of land improvements required for a PV array construction project:

• Land clearing – the demolition and removal of vegetation and structures to clear the land of obstacles.
• Site stripping – the removal of loose topsoil with root structures.
• Rough cutting and grading – the removal of dirt from large hills and the addition of dirt from large divots that exceed the array’s maximum allowable slope angle.
• Road construction – excavation, layering in 3-4 different gravel sizes, grading to have a crown to manage water runoff, and making a ditch to drain water into. Roads are required in most counties to give access to emergency vehicles when fighting fires on the site. Strangely, this is not required for row crop agriculture, despite the fact that corn burns far better than glass and aluminum.

The most competitive and modern solar structures require as little land preparation as possible, typically stripping no more than 5% of the land, and only when absolutely necessary. In general, I provide the costs of land clearing and stripping on a per-acre basis.

The costs of cutting and filling are typically per cubic yard or cubic meter. The low preparation solutions minimize the amount, so I will assume another 5% of the land will need 1 yard of depth cut or filled.

Finally, roads are cost-based on a per-mile or per-km basis. Most counties in the United States require a 20-foot fire access road that supports a fire truck along the entire perimeter of the solar array.

Table 9: Land prep costs

The East-West array looks quite attractive when considering land costs, both in total project cost and on a dollar-per-watt-installed basis.

Table 10: Total land costs per state and array type

Environments with challenging weather

Arrays have a design life of 20-30 years, over which they will experience a large variety of weather events. The most significant are lightning strikes, wildfires, high winds, and hail.

Heavy rains don’t harm panels and, in some ways, make them perform better since it’s ultimately just a cloudy day with a free panel cleaning service. There is a risk of scouring, which is the slow erosion of dirt around the piles due to wind and water. Still, this is typically accommodated during the design phase by determining the depth to which the piles or rods will be driven, so that they can survive the maximum scouring. Therefore, survivability in heavy rains is no differentiator between these array structures.

Lightning strikes pose similar challenges to both structures. Both have methods of conducting high current surges to ground. Both rely on surge protection systems in the electrical balance of system to reduce the risk of surges frying individual panels. This, too, does not give us any means of differentiating the system performance.

Designers and system operators mitigate wildfires by clearing the setback distance around the solar array. However, if a wildfire near the array is hot enough on a windy day, those ashes can fly right over the setback distance into the precious PV modules.

Surviving wind, caused by tropical storms, near hurricanes, or tornadoes, is a moderate differentiator of the systems. The East-West arrays perform best due to low angles of attack, resulting in less lift. The small gaps between panels prevent the formation of large vortices on the leading edge of the panel. So while the East-West array has less material affixing it to the ground, it experiences far less lift in high wind scenarios, leading to very high survivability against high winds, whether from Santa Ana winds or a hurricane. One company, Jurchen, claims that its PEG system can survive 180 mph winds, which exceeds the speeds expected in a category 5 hurricane. However, to my knowledge, no real-world testing has verified this claim.

Single-axis tracked arrays are less reliable in extreme wind scenarios, but have added stow features to reduce their angle of attack during high wind days. The individual rows of the single-axis arrays are prone to more aerodynamic loads, leading to fluttering, which can change the panel’s tilt angle, create ground vortices, twist the torque tube, and, in the most severe conditions, eventually cause array collapse. Check this out for a more in-depth understanding.

Figure 24: Simulated wind flow analysis on the single-axis tracker

Modern solar tracker designers understand these risks and have incorporated aeroelastic loads on the tracker, leading to stiffer torque tubes or the addition of dampers. These upgrades add to the system’s costs, but allow single-axis tracked arrays to withstand up to 120 mph winds, such as those seen in a category 2 hurricane, and all winds expected under normal conditions.

Single-axis tracked arrays excel in one type of weather pattern, however: large hailstorms. Hail has been a problem in the solar industry since its inception, leading to huge insurance payouts, especially in Texas, such as $70 million for the Midway Solar Project in 2019 and $50 million for the Fighting Jays solar project in 2024.

Texas may be especially at risk today, as it is located in Tornado Alley and has some of the worst hailstorms in the country, and due to its high solar density. A storm with hail larger than 1.8 inches (45mm) used to be considered a once-in-five-hundred-years storm, but such storms are becoming even more common due to the impacts of climate change, with at least two occurring in the last five years.

To combat this, single-axis tracker systems have implemented hailstorm stow protocols that reduce impact energy. During a hailstorm, the PV modules are oriented facing away from the hail’s angle of attack based on the wind speed and direction, turning a direct impact into a glancing blow. Three solar plants located near the Fighting Jay project, which experienced the once-in-500-years storm, survived the same event with minimal damage. The limited losses they did incur were confined to tracker systems that were either not equipped with hail stow features or were disabled at the time.

Trackers are not the only way to address this issue; the industry has demanded more from its manufacturers. For the past couple of years, PV modules have required testing to demonstrate their resilience to hail up to 1 inch in diameter under UL-61730, with additional tests defined (but not mandated) for hail up to 3 inches in diameter.

The dominant PV module in the US is a bifacial panel with 2 mm of tempered glass on both sides, supported by an aluminum frame. The modules are stiff and do relatively poorly when pelted repeatedly with balls of ice. Lab testing has shown that these panels typically break when impacted by ice balls of 1.8 to 2.4 inches in diameter, hitting them dead on, which represents a 12 to 30 Joule impact, respectively. The ability to withstand hail up to that size ensures these panels are safe for all array configurations in areas such as the southwest and northwest.

New PV modules are being marketed with higher impact ratings due to an increased thickness of the glass sheet, from 2.0 mm to 3.2 mm or even 4.0 mm, which should enhance survivability enough for the northeast and southeast in any array configuration. Future improvements hope to reach the survivability levels needed for Texas and the rest of the Midwest.

The hail stow feature is a relatively new addition to single-axis tracked arrays, but it constitutes one of the largest moats this configuration has against the faster, simpler, and cheaper East-West arrays in areas where hail is likely. There are instances of PV array operators reducing their insurance premiums by up to 72% through the construction of hail-resistant systems, single-axis tracked solutions with a validated hail-stow function, and higher impact-rated panels. This can save a lot of money when insurance costs average 1.5 cents per watt installed per year. Since the example listed went to significant effort to negotiate these insurance savings, it would be incorrect to assume all single-axis arrays automatically qualify for reduced premiums with hail-resistant systems.

What happens as modules get cheaper?

As PV modules continue to fall, the situation for single-axis tracked arrays only gets worse. And I’m betting on this with Terraform Industries, where we aim to leverage cheap solar to manufacture cheap natural gas and other products. So let’s find the crossover point, the cost of the PV module, where the East-West array reaches cost parity with the single-axis tracked array.