Solar panel orientation and tilt: what you lose vs. ideal south
By Fotovol·Updated 10 July 2026
1. The short answer, in numbers
A perfectly oriented system — south-facing, tilted at 30-35° — produces between 1,150 and 1,400 kWh per kWp per year in Romania, depending on the region (the monthly table by region is in how many kWh 1 kWp produces in Romania). Any other orientation produces less, but "less" is not dramatic:
- south-east or south-west: around 3-5% lost per year;
- east-west: typically 10-15% lost;
- any tilt between 20° and 45°: only a few percent of difference;
- north: 30-40% lost — the only genuinely problematic orientation.
The takeaway before any table: a non-south roof is not a problem, it is a 10-15% correction. Below are the full numbers, how to check them yourself in PVGIS (#6), and when a "badly oriented" roof still pays back (#7).
2. Losses by orientation (azimuth): the full table
Azimuth is the direction the panels face. For a typical 25-40° roof pitch, losses versus south look like this:
| Orientation | Output vs. south | Annual loss |
|---|---|---|
| South | 100% | — |
| South-east / South-west | 95-97% | 3-5% |
| East / West | 85-90% | 10-15% |
| North-east / North-west | ~70% | ~30% |
| North | 60-70% | 30-40% |
Two details that often get lost:
- South-east and south-west are practically south. A house rotated 45° off axis loses about one cloudy week per year.
- The flatter the pitch, the less orientation matters. At 10° tilt, an east-facing panel "sees" almost the same sky as a south-facing one. The big losses come from steep pitch combined with wrong orientation.
3. Losses by tilt: 10°-60°
For south orientation, output by pitch:
| Tilt | Vs. the 30-35° optimum |
|---|---|
| 10° | −4-6% |
| 20° | −1-2% |
| 30-35° | optimum (100%) |
| 45° | −1-3% |
| 60° | −6-8% |
The correct reading: between 20° and 45° — which covers virtually every pitched roof in Romania — the difference is a few percent. There is no scenario where rebuilding the roof structure or paying for tilt-correction frames earns those percent back.
A steep pitch (45-60°) even has hidden upsides: it produces better in winter, when the sun sits low, and snow slides off by itself. It loses in summer, when you have surplus anyway. We covered angle selection in detail in the optimal tilt article — the short answer stands: mount flush with the roof and move on.
4. Why east-west is underrated
East-west loses 10-15% of annual production — but that number hides an advantage no table shows: the production profile matches household consumption better.
A south system peaks aggressively at noon, exactly when many people are not at home; the surplus goes to the grid at the netting value, not at the value of a kWh consumed on the spot. An east-west system has a flattened curve: more in the morning and evening — precisely when a household actually consumes.
The practical result: higher self-consumption, so every kWh produced is worth more. For many prosumers, an east-west system at −12% annual production delivers monthly savings very close to a "perfect" south one.
One non-negotiable technical condition: east and west panels never share the same string — each orientation goes to its own MPPT input on the inverter. Why, and what happens when an installer mixes them, is explained in string sizing.
5. Flat roofs: ballast, rows and inter-row shading
A flat roof (terrace, industrial hall) gives you full freedom of orientation — paid for in structure. Panels go on ballasted frames (concrete blocks), usually at 10-15°, without drilling the waterproofing.
Why 10-15° and not the optimal 30-35°? Two reasons:
- Wind: the force on the panel grows quickly with the angle, and a steeper frame needs more ballast — weight the slab has to carry.
- Inter-row shading: the higher the panels sit, the further apart the rows must be to avoid shading each other in winter — so fewer panels fit.
Hence the standard configuration on industrial halls: east-west "tent" layouts, panels back to back at 10-15°. Each panel loses a few percent, but the layout packs 30-40% more panels onto the same surface than spaced south-facing rows. Per roof, it produces more.
How many panels fit on your roof, and in which layout — try it visually in the panel positioning tool.
6. Check it yourself in PVGIS, in five minutes
You don't have to take the tables above — or your installer — on faith. PVGIS is the European Commission's free tool, the same one system designers use, built on satellite irradiation data for any point in Europe:
- Open
re.jrc.ec.europa.eu/pvg_toolsand drop the pin on your roof. - Choose "Grid connected", set the system size (say 5 kWp) and leave system losses at the default (14%).
- Fill in "Slope" (roof pitch) and "Azimuth" (orientation). Mind the PVGIS convention: 0° = south, −90° = east, 90° = west.
- Hit "Visualize results" — you get annual and monthly production in kWh.
The useful trick: run the calculation twice — once with south/35°, once with your roof's real pitch and orientation. The difference between the two figures is your exact loss, computed for your location rather than estimated from a generic table. Any installer claim about orientation can be verified this way in five minutes.
7. When a "badly oriented" roof still pays back
Orientation loss translates directly into payback years, and the math is gentler than it sounds. A well-sized residential system facing south typically pays back in 5-8 years without subsidy — faster with the Casa Verde programme. The full calculation, with scenarios, is in solar system payback, and starting prices in how much a solar system costs.
An east-west system at −12% production pushes payback out by roughly one year. With panels warrantied for 25 years, one extra payback year doesn't change the decision — it changes one line in a spreadsheet.
North is a different story. At −30-40%, payback stretches toward 10-12 years and usually isn't worth it. Exceptions exist — very low pitch, high consumption with a large cheap-to-cover roof, or a south face that is already full — but exceptions get calculated in PVGIS, not assumed.
8. Common orientation and tilt mistakes
- Mixed orientations on one string. The most expensive silent mistake: east panels cap the west ones and you lose 10-20% with no error shown anywhere. Check the string layout before installation (#4).
- Tilt-correction frames on an already pitched roof. Paying extra for structures that win back 2-3% almost never pays off — and it ruins the roofline.
- Optimizing the angle while ignoring shade. A chimney shading two panels for a few hours a day costs more production than the entire east-west "loss". The correct priority order: shade, then orientation, then tilt.
- Oversizing to compensate for orientation. More panels do not mean more value if the surplus leaves cheaply into the grid. Size for consumption with the calculator, then correct for orientation. In the ranking of prosumer mistakes, oversizing is always near the top — orientation doesn't even make the podium.
9. What this means for your roof
The numbers worth remembering:
- South at 30-35° is the reference: 1,150-1,400 kWh/kWp/year in Romania.
- South-east/south-west lose 3-5%, east-west 10-15% — corrections, not problems.
- Tilt between 20° and 45° changes output by a few percent: mount flush with the roof.
- Only north requires a serious calculation before deciding.
- East-west claws back through self-consumption part of what it loses in total production.
Practical steps: size the system for your consumption with the calculator, see how many panels fit with the panel positioning tool, verify any offer's figures in PVGIS, and get quotes from several verified companies so you can compare the proposed orientation layouts too. And if you want to see what real production looks like month by month, we published the numbers of a 5 kWp system in Brașov.