Bifacial modules

A bifacial module generates from light reaching its rear face as well as its front. The rear face sees mostly light that has bounced off the ground between and under the rows, so the gain it produces is a property of the site as much as of the module: the same module over dry sand and over dark wet soil does not give the same answer.

Two inputs control it, and they sit in two different places. The module's own bifaciality lives in Module Specifications; the ground reflectance lives with the energy settings. Both are needed before any rear-side energy is counted.

Turning it on

Tick Bifacial module by hand, or let a module file do it for you. Load .PAN reads the manufacturer's .PAN file, and when that file declares the module bifacial it ticks the checkbox and fills Bifaciality factor (φ) with the manufacturer's value. You therefore do not have to know whether a given part number is bifacial to get the rear side modelled — see Module specifications.

Neither field moves a single table. Turning bifaciality on changes the energy calculation only; the placed layout, the module count and the DC capacity are identical either way. Run the energy calculation again after changing either one, but there is no need to generate the layout again.

The bifaciality factor

Bifaciality factor (φ) is the module's rear-side efficiency expressed as a fraction of its front-side efficiency. At φ = 0.70, the rear face converts light to power at 70 % of the efficiency of the front face, so one unit of irradiance landing on the back is worth 0.70 of the same unit landing on the front.

• Typical published values are 0.65 to 0.80. The field accepts 0.50 to 0.95 and ships at 0.70.
• On a datasheet it appears as a bifaciality, bifaciality factor or rear-side power ratio, usually as a percentage with a tolerance — 70 % ± 5 % is a common form. Divide by 100 before entering it.
• It is a property of the module and nothing else. Take it from the datasheet or from the module file. Do not tune it to reach a target yield; the site inputs below are where a design legitimately differs from another.

Ground albedo

Ground albedo (ρ) is the fraction of the light striking the ground that the ground reflects back. Fresh concrete reflects a large fraction; dark tilled soil reflects little. It is the one input in the rear-side model that describes the site rather than the hardware, and on a real project it is the number a lender's reviewer will ask about first.

With Bifacial module off there is no rear-side term at all, so raising the albedo changes no rear-side figure. The field is live in the panel either way, which is the first thing that catches people out.

The second is where it lives. Albedo is set with the energy settings, several groups below the module fields it works with, so a reader who ticks Bifacial module and stops there has accepted 0.25 without deciding on it. The split is deliberate — bifaciality is a module property and albedo is a ground property — but it does mean the two halves of the same calculation are never on screen together. Set both in one pass.

The field affects the rear side only

Ground albedo (ρ) changes the bifacial gain and nothing else. In particular it does not change the front-side in-plane irradiance. The transposition that turns horizontal irradiance into in-plane irradiance has a ground-reflected term of its own, and that term uses a fixed internal reflectance of 0.20 — the field you set here is never passed to it. See Weather data for the transposition itself.

The practical reading: on a monofacial plant the albedo field is inert, exactly as its own tooltip says. On a bifacial plant it is the site input with the most authority over the rear-side gain — but if you raise it expecting the front-side GTI to move as well, it will not, and neither will the yield of a monofacial design.

How the gain is estimated

Three relations, applied in order. The first estimates the irradiance reaching the rear face, the second is the geometric factor inside it, and the third turns irradiance into a gain.

GTI_rear ≈ GHI × ρ × F_ground_rear × (1 − GCR)

F_ground_rear = (1 + cos(tilt)) / 2

bifacial_gain = φ × GTI_rear / GTI_front

Every term:

The view factor falls as tilt rises: at zero tilt the rear face looks straight down and the term is 1, and at 90° it sees half ground and half sky and the term is 0.5. A flatter array therefore has a rear face pointed more squarely at the reflecting ground.

Two consequences follow from these relations directly, and both are design levers rather than curiosities.

A denser array gains less from bifaciality

Rear-side irradiance carries the factor 1 − GCR. Ground coverage ratio is the collector width as a fraction of the row pitch, so packing rows closer raises GCR and shrinks 1 − GCR — there is less exposed ground left between the rows to catch light and bounce it under the modules.

The trade is the familiar one, now with a second term in it. Reducing the row pitch fits more capacity onto the same site, but it costs row-to-row shading and rear-side gain at the same time. On a bifacial plant the optimum pitch sits wider than the same exercise on monofacial modules would suggest. Pitch is set on Spacing and tilt.

A more reflective ground gains more

Rear-side irradiance is directly proportional to ρ. Doubling the albedo doubles the estimated rear irradiance and doubles the gain, everything else held still. That is why ground treatment under and between the rows is a genuine design decision on a bifacial plant, and why the albedo you enter has to match the ground you are actually going to build on — including how it will look after a few years of vegetation management, not on the day of handover.

How much gain to expect

Typical bifacial energy gain over the same plant built with monofacial modules is 5 to 15 %. Where a specific design lands inside that band is decided far more by the ground under the array and by how densely the rows are packed than by the bifaciality factor, which barely varies between current products.

A result outside that band is a signal to check the inputs rather than a result to report. The usual causes are an albedo entered for the wrong ground cover and a row pitch that does not match the design being priced.

Where the gain appears in the results

The gain is an energy effect, so it appears inside the energy figures and nowhere else:

• The first-year and lifetime energy figures, and the three exceedance probability columns at the right-hand end of the summary table, all carry it.
• The capacity factor carries it, because the capacity factor is computed from first-year energy over the same installed DC capacity.
• The layout columns — table counts, modules, DC capacity — do not change at all. There is no bifacial column in the summary table.
• The gain itself is computed as a percentage, and it is printed on the energy pages of the exported report, beside the inputs and the performance ratio breakdown that produced the figures. See The PDF report.

There is no separate on-screen readout of the gain. Nothing in the panel or the summary table shows it as its own percentage, so if you need the figure itself, take it from the report's energy pages.

Failing that, the cleanest way to see the effect on a design you have already run is to calculate energy with Bifacial module off, note the first-year figure, then tick it and calculate again. The layout is untouched between the two runs, so the difference is the rear-side contribution and nothing else.

Where to go next