What Is the Best Tilt Angle and Direction for Solar Panels?

It all reduces to one cosine

A flat surface intercepts a beam in proportion to its projected area. Turn it away from the beam and the projection shrinks as the cosine of the angle between them — the angle of incidence.

Two consequences follow immediately, and they are the whole article. First, a panel aimed anywhere near the sun collects nearly everything, because cosine is flat near its maximum. Second, since the sun's position changes through the day and the year and a fixed panel's does not, no single angle is right — the choice is which times of year and which hours to favour.

There is also a secondary effect worth knowing: at very oblique angles a rising share of the light is reflected off the glass before it enters at all, so the real penalty at extreme angles is slightly worse than the cosine alone suggests.

Why latitude is the starting point

The sun's noon elevation depends on latitude and season. At the equinoxes it stands at ninety degrees minus your latitude above the horizon; it climbs higher in summer and sinks lower in winter.

Setting the tilt equal to the latitude points the panel's normal at the equinox sun, which splits the seasonal difference evenly and maximises the annual total for most sites. Facing the equator — south in the northern hemisphere, north in the southern — follows from the same geometry, because that is the direction the sun occupies at noon.

That is a starting point rather than an answer. A site with heavy winter demand wants a steeper angle than the annual optimum, which is why an off-grid array sized by worst-month load calculation is often pitched more steeply than a grid-connected one on the same roof.

The optimum is remarkably flat

This is the part that saves a great deal of worry. Because the cosine is flat near its peak, and because the sun sweeps through a large range of positions anyway, the annual energy curve against tilt has a very broad top.

In practice the first ten degrees of error in tilt cost a small single-digit percentage of annual output, and the azimuth curve is broader still. A roof that faces roughly the right way at roughly the right pitch is not a compromise worth engineering around. Twenty or thirty degrees off, or facing sideways, starts to be a real number — but at that point the roof has made the decision, not the optimisation.

The useful consequence is that effort is better spent elsewhere. A fortnight of accumulated dust, or a shadow crossing the array at nine every morning, will move more energy than the last five degrees of tilt ever will.

Direction changes when, more than how much

Azimuth has a second effect that tilt does not, and it is often the more useful one.

An east-facing array front-loads the day; a west-facing one back-loads it; splitting an array east and west produces a broad plateau instead of a peak. Annual totals fall somewhat in each case, but generation arrives when it may be more useful — and a flatter curve is less likely to run into the limits that cause surplus to be curtailed.

This is also why mixed orientations are worth taking seriously rather than avoiding, provided the electrical design can handle modules that no longer match each other.

Tilt does more than catch light

Three practical effects routinely override the optimal angle, and all three are about what happens on the glass rather than in it.

Shedding is the one people underestimate. A shallow array holds water, and water that evaporates in place leaves its minerals behind — the cementation problem that makes a shallow array need cleaning far sooner than a steep one. In snowy climates the difference between shedding and not shedding is measured in weeks of lost production.

Row spacing is the one that costs money in area. Steeper rows cast longer shadows, so they must be set further apart, and on a fixed site that means fewer rows. At some point the extra spacing costs more output than the extra tilt gains — and the shadow of one row on the next is exactly the partial shading problem, arriving at the same time every morning.

When to ignore the rule of thumb

Steeper than latitude if winter matters more than the annual total: off-grid systems, heating loads, high-latitude sites where the winter sun barely clears the horizon. Shallower if summer dominates, or if wind loading and mounting simplicity are the constraint.

Vertical, occasionally: east-west facing vertical arrays give up the midday peak entirely in exchange for morning and evening generation and ground that stays fully usable between rows — the reasoning behind vertical mounting in agrivoltaic systems.

And whatever angle the roof already has, if it is within fifteen degrees or so of the ideal and faces roughly the right way, take it. The energy available from arguing further is smaller than the energy available from a clean panel and an unshaded morning — smaller, too, than the gap between cell technologies, and unlike tilt, those are things you can still change after the array is bolted down.