Fixed Tilt vs Single-Axis Tracker: Why Yield Math Is Not Enough

Most fixed tilt vs single-axis tracker comparisons stop at one question: does the extra energy pay back the extra steel, motors and maintenance? That calculation is necessary but incomplete. What a tracker’s additional megawatt-hours are worth depends on when they are produced, how the support scheme settles them, and which constraint actually limits the project: land, grid capacity, an awarded capacity figure or terrain. My position is that the right mounting choice comes out of the revenue model and the project’s binding limit, and the CAPEX-to-yield ratio only confirms it.

Why fixed tilt vs single-axis tracker is not only a CAPEX and yield question

The physics is well documented. A horizontal single-axis tracker rotates its rows from east to west, keeping the modules closer to normal incidence through the day. IEA PVPS Task 13 puts the resulting gain at 15 to 20% more annual energy than an equator-facing fixed-tilt array of the same capacity, depending on latitude and layout, with bifacial modules adding a further 2 to 10% absolute in its US simulations ^[1]. The same report is clear that trackers cost more to build and maintain, and that the extra energy has to exceed that extra expense ^[1].

Two details matter at Central European latitudes. First, the gain depends on the share of direct irradiance. Measurements at a 43.5°N site with an oceanic climate in northern Spain found an annual gain in incident irradiation of only about 8.5% for a backtracking tracker, against about 13% in theory ^[2]. Second, the gain is seasonal. At that site the tracker collected more from March to September and the fixed array collected more in winter, with measured monthly differences ranging from about +30% in July to about -24% in December ^[2].

So a tracker adds energy mostly in the high-sun months and away from solar noon. What those megawatt-hours sell for is a market question, not a structural one. Having evaluated more than 5 GWp of projects in Hungary and Turkey, I treat the mounting structure as a revenue decision first and a steel decision second.

Price shape: what a tracker’s extra megawatt-hours are worth

Across Europe, solar is reshaping the daily price curve. ACER’s 2026 monitoring report describes a deepening duck curve, with low prices at midday, higher prices in the morning and evening, and daily price swings around five times larger than in 2020 ^[8].

Hungary sits at the sharp end of this trend. Solar supplied about 18% of the country’s electricity in 2024, and the solar capture price fell from about 90% of the average market price in 2020/21 to about 55% in the first half of 2025 ^[3]. Hungarian day-ahead prices were negative in 261 hours in 2024 and 277 hours in 2025, and the 2025 average daily price profile ran from about 58 €/MWh in its cheapest hour to about 186 €/MWh in its most expensive ^[9].

This is where the tracker’s profile pays. A fixed south-facing array peaks at solar noon, when prices are weakest. A tracker produces a broader plateau, with relatively more energy in the morning and late afternoon. An analysis by EWI at the University of Cologne, using German 2024 prices, found that a tracked plant generated 29% more full-load hours than a fixed one but earned 43% more market value, with the difference concentrated in the shoulder hours, where the tracker’s market value was about 68% higher ^[10]. Those two figures together imply roughly 11% more value per MWh on top of the extra volume.

Treat that study with care. It was commissioned by a tracker supplier, the generation profiles and cost inputs came from that supplier, and the fixed-tilt reference used a shallow 20° tilt, which likely flatters the tracker’s energy gain ^[10]. The direction of the result is robust; the magnitude does not transfer to a Hungarian site without your own simulation.

There is also a seasonal offset. Because the tracker’s gain is concentrated in spring and summer ^[2], part of its extra energy arrives when the system already has the most solar, while fixed tilt at its optimal angle holds up better in winter. A merchant model needs hourly prices for the whole year, not a single capture-rate assumption. As IEA PVPS notes, only a flat tariff or flat-rate PPA lets revenue be calculated as annual energy times one price; under merchant sales, contracts for difference and even fixed-price PPAs, the hourly profile is priced in, directly or by the offtaker ^[1].

How Hungary’s METÁR premium changes the answer

How the premium is settled

METÁR’s green premium is awarded by tender for plants of 1 MW and above, and bidders compete on their offered supported price. The premium equals the supported price minus a reference market price; if the result is negative, the producer pays the difference back ^[4]. In practice it behaves like a two-way contract for difference.

The detail that matters for the mounting decision is the reference price. For solar plants, it is the monthly average of day-ahead prices weighted by the combined output of all supported solar plants in each settlement period ^[4]. The reference is therefore the same for every supported solar plant in a given month, whatever its own production profile, while each producer sells its own energy in the market and bears the risk of that sale ^[11].

Three further rules shape the economics:

• The regulator’s decision fixes the supported price, the maximum quantity of electricity that receives support, the support period and the deadline for commercial operation ^[4]. The maximum quantity is based on technology-specific benchmark utilization factors published in the tender documentation ^[12].
• In the 2020 round, the support period was 15 years, and the benchmark was 1,440 hours per year for fixed-tilt solar and 1,680 hours for trackers, with no larger supported quantity allowed above those values ^[5]. The requested quantities in a published analysis of MEKH bid data match those ratios ^[11].
• No premium is paid for production during runs of at least six consecutive negative-price settlement periods ^[4]. Since a 2025 amendment moved settlement to 15-minute units, that window is 1.5 hours rather than six hours ^[13].

Why this favors trackers

This is the first half of my position. Under a fleet-weighted reference, a plant whose profile is better than the fleet average earns more than its supported price. A tracker sells relatively more energy into the higher-priced morning and late-afternoon hours, so its own capture price tends to sit above the fleet reference, and the producer keeps the difference.

A hypothetical illustration in round numbers: take a supported price of €60/MWh and a monthly solar reference of €45/MWh, so the premium is €15/MWh. A plant that sold its output at an average of €50/MWh that month ends at €65/MWh; a plant that sold at €42/MWh ends at €57/MWh. The scheme removes price-level risk but leaves the profile difference with the producer.

On top of that, the tracker’s higher benchmark gives it about 17% more supported energy per MW. Tracker projects were competitive: in the 2020 round, 86 of the 255 solar bids used trackers, and the three lowest bids all came from tracker projects ^[11].

Where fixed tilt wins: the size-limited project

The second half of my position is that for a size-limited METÁR project, fixed tilt can be the better investment because it maximizes production across the year. Whether that holds depends on what limits the size.

Where the plot is the limit, fixed racking fits more capacity on the same land. In US plants built in 2019, the median fixed-tilt site carried about 0.86 MWdc per hectare against about 0.59 for trackers, and still produced more energy per hectare, roughly 1,100 against 970 MWh per year ^[6]. Latitudes and layouts differ from Hungary’s, but that gap in capacity density is far larger than the 17% benchmark advantage the tracker receives per MW. IEA PVPS makes a related point about financial models: for land-constrained projects, value is often higher with tighter row spacing and a larger plant, even at the expense of specific yield ^[1]. And because fixed tilt at its optimal angle performs better in winter ^[2], its output is spread more evenly through the year.

The honest limit is that this logic reverses when the binding constraint is grid capacity or an awarded MW figure and land is plentiful. Then the tracker produces more MWh per MW, and the METÁR benchmark lets it claim more supported energy per MW.

Context matters too. MEKH’s METÁR page lists six tender calls between 2019 and March 2022 and no newer round, alongside a notice extending implementation deadlines for awarded rights ^[14]. Of Hungary’s roughly 4.6 GW of solar plants above 50 kW, only about 1.4 GW is exposed to market prices ^[3]. The METÁR reasoning therefore matters for awarded projects still in delivery and as a template for any future CfD-style scheme: always ask what reference price the payment is settled against.

Grid-limited projects: DC/AC ratio, clipping and export caps

Where a project is limited by its connection rather than its land, the comparison is easy to get wrong.

A tracker’s flatter profile spends more hours near peak output, so at the same DC/AC ratio it clips more. In a minute-resolution US study, moving from fixed tilt to single-axis tracking roughly doubled annual clipping to about 1% at a ratio of 1.25, and raised it from 16% to 22% at a ratio of 2.0 ^[15]. Fixed-tilt designs can therefore carry a higher DC/AC ratio before clipping erodes value, and a high ratio flattens a fixed plant’s output into a plateau of its own ^[15], cutting the cheapest noon peak while adding output in lower-irradiance hours.

The EWI analysis found that, for the same connection size, the tracked plant always earned more market value than the fixed one in 2024, and argued that trackers use scarce connection capacity more efficiently ^[10]. But it held DC capacity at 150 MWp for both designs. The fair question for a grid-limited site is a tracker at its optimal DC/AC ratio against fixed tilt at its own, higher optimal ratio, both filling the same export limit. With low module prices, extra DC on fixed tables can be the cheaper route to more MWh per MW of connection. Sometimes it is not; the answer is project-specific.

IEA PVPS recommends treating grid sizing as a separate optimization track that includes curtailment from limited or non-firm connection agreements, and checking whether the curtailed hours are high-value or low-value hours ^[1]. If a battery is planned at the same connection point, rerun the mounting comparison with it: a battery that moves midday energy into the evening captures part of the same price spread the tracker is chasing.

Terrain, weather and O&M: risks the yield model does not show

Terrain. In the IEA PVPS survey of tracker suppliers, most offered products rated for a maximum overall north-south grade of 15% (8.5°), with more tolerance east-west; the one supplier claiming 30% also had the shortest rows, at 25 m ^[1]. Beyond those limits a tracker needs grading, varying pile heights or a terrain-following design, and each option adds cost or mismatch losses. Much of Hungary’s Great Plain is flat enough; hillier sites need a closer look before the tracker is assumed.

Wind and hail. Trackers rely on stow strategies, and survey responses put wind-stow thresholds at 15 to 22 m/s ^[1]. Long, slender torque tubes are exposed to torsional galloping, an aeroelastic instability that can lead to failure of the row ^[1]. Hail is rare but expensive: an insurer cited by IEA PVPS reported hail at 1.4% of its PV claims but 54% of incurred costs, with damage made worse when trackers did not stow or stowed incorrectly ^[1]. A working hail stow is an advantage over fixed tilt; a failed one is a liability. Only about 30% of surveyed owners and operators reported that their trackers had a hail response ^[1].

Availability. Tracker failures rarely disconnect the array, so they usually cost less than an inverter outage ^[1]. They are still common enough to model. A 2026 assessment of more than 2 GWp of plants, mostly in temperate climates, found median tracker availability between 66% and 88% depending on data treatment, against industry claims as high as 99%. It flagged missing tracker data, and on-site teams without access to or expertise in tracker control systems, as systemic problems ^[7]. If the O&M contractor cannot diagnose a tracker controller, part of the yield gain in the model is theoretical.

Decision framework: when fixed tilt wins and when the tracker wins

A practical sequence for the design stage:

1. Name the binding constraint first: land, connection capacity, awarded MW, or budget.
2. Simulate both options at their own optimum in PVsyst or RatedPower, including ground coverage ratio, DC/AC ratio, backtracking, terrain and bifacial gain.
3. Value the hourly output against an hourly price forecast, or against the exact support formula, including the negative-price rule.
4. Run a tracker availability sensitivity instead of a flat 99%, and price O&M, spares and stow-related losses.
5. Compare the options on NPV or IRR under the same constraint, not on LCOE alone.

Conclusion

The tracker is the default in much of the world for good reasons, and with Hungary’s midday price dip, its morning and evening output is worth more than its kWh count suggests. METÁR reinforces that. But a project rarely competes on yield alone. When land is the limit, when connection capacity can be filled more cheaply with extra DC on fixed tables, or when terrain and O&M capability raise the tracker’s real cost, fixed tilt can still be the better investment. Settle the binding constraint and the revenue formula first, then let the CAPEX and yield numbers confirm the choice.