By: Luis Reyes
Published: Aug 2, at 6:30am ET
Every solar panel you have ever driven past is tilted to face the sun, and the logic is not complicated. The sun sits highest at midday, midday is when the most energy is falling out of the sky, so you angle the glass at it and collect the peak.
There is a growing amount of solar hardware built to do the opposite. The modules stand dead upright, like a fence, faces pointed east and west instead of up. At solar noon the sun runs almost parallel to both sides of the glass, the angle of incidence goes to garbage, and output drops into a hole.
These arrays skip the noon peak on purpose. In the farm fields of Germany, Austria, Denmark, Sweden and Japan, they have now run long enough to produce operating data instead of brochures.
On July 1, pv magazine reported on a Swedish oat trial where the crop grown between upright panels outproduced the bare field next door. Two weeks earlier, the German company that commercialized the format launched a version aimed at utility-scale power plants. The American branch of this story is going considerably worse, and we will get to it.
A south-facing panel gives its widest opening to the sun when the sun is at its highest. Stand that same panel on edge facing east and west and the geometry inverts.
Morning light hits the east face almost perpendicular. Evening light does the same to the west face. Because the module is bifacial, with glass on both sides, both faces work a shift.
What comes out is not a duck curve. It is a camel.
Researchers at the University of York ran a vertical bifacial rig for a full year against two reference systems and published the results in Scientific Reports. Against a conventional 45-degree tilted monofacial array, the vertical system produced 26.91 percent more in the early morning and 22.88 percent more in the late afternoon.
No spam. Unsubscribe anytime. Privacy policy (opens in new window)
The morning gain landed between 5:30 and 9:00 am, the evening gain between 5:00 and 8:30 pm. Seasonally the advantage peaked in winter, at 24.52 percent.
Read the fine print before you get excited. That was a 1.5 kW bench with rows 50 cm apart on white gravel, a generous reflector, and the paper’s acknowledgements credit backing from Over Easy Solar, a Norwegian vertical-PV company. The shape of the curve is solid. The exact percentages belong to that rig.
Snow behaves strangely here too. It slides off vertical glass instead of burying it, and the white ground turns into a mirror feeding the bottom half of both faces.
A team at Nagaoka University of Technology in Japan spent two winters at a site where snow depth clears a meter. Albedo added 55.3 percent, snow burying the lower cells took back 17.0 percent, and the net effect of having a winter was a gain of 38.3 percent, according to their paper in EPJ Photovoltaics. No cracked glass, no bent frames.
Here is why anyone would build a generator that quits at lunchtime. Solar has gotten good enough at producing in the middle of the day that it is eating its own price.
California’s grid operator curtailed 3.4 million megawatt-hours of utility-scale wind and solar in 2024, a 29 percent jump over the year before, and solar was 93 percent of it, according to the Energy Information Administration. The worst of it lands in spring, when the sun is strong and nobody is running heat or air conditioning.
A generator that delivers at 7 am and 7 pm is selling into a different market than one that delivers at 1 pm. That is the entire commercial argument for standing the glass up.
Germany’s Next2Sun, building these systems since 2016, put numbers on it in June. Its June 16 analysis argues that bolting vertical east-west arrays onto existing south-facing solar parks could raise connected capacity by roughly 60 percent without meaningful curtailment losses. Three days later it launched Fields2Sun Max, a mounting system for utility-scale plants and modules above 700 Wp.
“Not every additional PV system places the same burden on the grid,” Next2Sun chief executive Heiko Hildebrandt said in the release.
The company also claims its vertical systems captured 87 percent higher market values than the overall German solar market between January and May 2026. That figure is Next2Sun’s own, about its own product, presented at its own trade show booth. No grid operator or regulator has confirmed it.
A non-vendor did back the underlying logic. Frankfurt Airport switched on 37,000 vertical modules along 2,800 meters of Runway 18 West last October, rated at 17.4 megawatts, and Fraport’s announcement is blunt about why. Its tilted rooftop panels peak at midday, the runway fence catches morning and afternoon, and running both flattens supply across the day.
That site is managed airport grassland, not cropland, so it says nothing about farming. It says plenty about the curve.
Land is the reason agrivoltaics exists at all, and vertical is the most extreme answer to it. Unlike the raised panels shading Arizona vegetable beds on stilts, nothing grows underneath anything here. The crop grows in the alley.
At Next2Sun’s Löffingen park in Germany, 4.3 megawatts sit across 27 acres with rows 13.5 meters apart and less than one percent of the ground sealed, which leaves the combine somewhere to drive.
Aarhus University put the sharpest number on the land question. Comparing vertical east-west against south-facing tilted at its Foulum site, the team found the panels occupied roughly 10 percent of the field, and calculated that producing the same food and the same electricity on separate parcels would need 18 to 26 percent more land.
The Danish group is honest about the tradeoff, which is refreshing given how this technology usually gets sold. Their vertical panels produced slightly less electricity per year than the tilted reference. The electricity was just worth more per unit.
Wheat and grass-clover between the rows showed no yield decline at all. “The crops don’t seem to mind the presence of solar panels,” said Uffe Jørgensen, a professor in the university’s agroecology department, who credits the wind shelter as much as anything.
That wind effect is the underrated part. Next2Sun’s research office reports wind speeds between rows running about 50 percent slower than open farmland, and estimates the drop cuts evapotranspiration two to three times more than the lost sunlight costs. In a dry August, less wind across a wheat field beats the sunlight the steel took away.
The Swedish oat data is the newest piece. Researchers at Mälardalen University grew Avena sativa under a 60-module vertical array at Kärrbo Prästgård and compared it against a ground-mounted array and open field.
Even after subtracting the 10 percent of land the structures occupy, total oat biomass beat the open field in one season and came within 106 kilograms per hectare in the other, with no significant difference in protein or fat.
Watch that land-loss footnote across all of these trials. Per plant the crop does fine. Per hectare of farm you are still down the strip the steel is standing on.
This is the part the trade press tends to walk past.
pv magazine visited a 1.9 MW vertical plant at Neudorf an der Mur in Austria, built in 2022 across 12.6 acres and now growing pumpkins and soy between the rows. The owners put the structure cost at roughly €200,000 per megawatt. Conventional ground-mount in Austria runs about €110,000.
Close to double, on a layout that fits fewer megawatts per acre to begin with.
The maintenance column looks unusually good against that. Rows sit 9.4 meters apart on poles driven up to 2.5 meters into the dirt with no concrete footings, and of 4,500 panels, seven have taken mechanical damage from farm equipment. Nobody has cleaned the array since 2022, because rain handles it and vertical glass does not hold dust the way tilted glass does.
So the honest version of the economics is that vertical has to earn its entire premium back on the price it captures. If midday power keeps collapsing in value, it does. If storage keeps filling the midday trough, and battery buildout in California is already pushing those prices up, the margin narrows.
In December 2023, Next2Sun and the Vermont solar contractor iSun announced the first vertical agrivoltaics system in the United States. A 50 kW array on 3.7 acres at the University of Vermont’s Horticulture Research and Education Center, three rows running north to south, 30 feet apart, with carrots, beets and saffron between them. Construction was set for early 2024.
It was never built.
The final report on the USDA-funded research grant attached to that array, a $199,998 award to UVM, lays out the sequence. Vermont approved the plans in June 2022. Legal agreements took until January 2023. The contractor requested state extensions in May 2023 and again in February 2024. Then iSun filed for Chapter 11 in Delaware in June 2024, and its assets were sold that August.
“Hopes were crushed by the dissolution of iSun,” the report says, in what may be the most unguarded sentence in the federal agricultural research literature. The university terminated the contract, is now in final negotiations with a different installer, and hopes to see the array standing by late 2026.
Meanwhile the researchers had a grant to spend and no solar panels. So they built fences out of black knitted shade cloth, sized with a shading tool from the Massachusetts Department of Energy Resources to reproduce the light a vertical array would have cast, and grew Boro beets and Negovia carrots around them across three trials.
They found no clear yield differences between shade levels, which is encouraging and also, as the report concedes, needs redoing inside an actual array.
Real American data exists, just not much of it. Colorado State University ran silage corn between north-south rows of vertical bifacial modules through the 2024 season and found no significant difference in silage or grain yield against unshaded controls, with the plots that caught morning light showing the highest photosynthetic rates. Corn is shade-sensitive, so that result matters. It is also a research plot, not a farm.
Standing solar panels on edge in a field is not a better way to make electricity. Measured over a year it makes slightly less, and the Danish researchers say so plainly. It is a better way to make electricity people actually want, on land that stays in production, in weather that flattens conventional arrays.
Whether that is worth roughly double the steel depends entirely on how badly your local grid has stopped paying for noon.
American farmers get to find out whenever somebody finishes building one. Germany has been standing panels on edge since 2016, from wheat fields to 2,600 of them floating upright on a Bavarian gravel lake. Four years into the Vermont project, the crop research is still happening next to a fence made of shade cloth.
Don’t bite your tongue. Speak up.
Luis Reyes · Jul 31, 2026
Luis Reyes · Jul 10, 2026
Luis Reyes · Jul 16, 2026
Dave McQuilling · Jul 8, 2026
Luis Reyes · Jul 19, 2026
Luis Reyes · Jul 31, 2026
Luis Reyes · Aug 1, 2026
Luis Reyes · Aug 1, 2026
Luis Reyes · Aug 1, 2026
Luis Reyes · Aug 1, 2026
Luis Reyes · Aug 1, 2026
Autonotion is the English-language automotive editorial by Autonocion.com — car news, reviews, and industry analysis for American readers.
Other links
Company
Subscribe
Get the latest car news in your inbox:
By submitting your email you allow autonocion.com to send you news or promotions. More info

source