Field Testing Concept
Understanding the
Geff Calculation
Why the sunlight hitting the front of a bifacial module is only half the story — and how to account for both sides when correcting field measurements to standard conditions.
The Problem with Standard Irradiance Measurement
When you measure a PV string in the field, you almost always record the irradiance at the same time — typically using a reference cell or pyranometer mounted in the plane of the module. That reading tells you how much sunlight is hitting the front face of the module at that moment, and you use it to correct your voltage, current, or power measurement back to standard test conditions (STC) at 1,000 W/m².
For a monofacial module — one that generates power only from its front face — that approach is correct. The front-side irradiance is the only irradiance that matters.
But bifacial modules generate power from both sides. The rear face picks up reflected light from the ground, surrounding structures, and the sky — and that contribution adds meaningfully to what the module is actually producing. If you measure a bifacial string and correct your result using only the front-side irradiance reading, you are dividing by a number that is too small. The result looks like the string is performing better than it should be at STC — which is mathematically correct — but the correction is wrong because you have not accounted for all the light the module was actually working with.
Standard irradiance measurement only sees the front of the module. Bifacial modules also use the light hitting their rear. Geff is the number that combines both into a single effective irradiance value that correctly represents the total light input to the module.
What Geff Actually Is
Geff — effective irradiance — is a single number that represents the total light available to a bifacial module, expressed as if it were all arriving at the front face. It combines front-side and rear-side irradiance into one value that can be used directly in the same STC correction calculations you would use for a monofacial module.
Think of it this way: your correction math expects a single irradiance number. For a monofacial module that number is straightforward — it is just the front-side reading. For a bifacial module you need to roll both the front and rear contributions into one equivalent number before the same correction math can be applied. Geff is that equivalent number.
What the Bifaciality Coefficient Does
The bifaciality coefficient (φ) is the key ingredient that makes this calculation accurate. It answers a simple question: how efficient is the rear face of this module compared to the front face?
A module with a bifaciality coefficient of 0.70 means the rear face is 70% as efficient as the front face at converting light into electricity. So 100 W/m² hitting the rear does not add 100 W/m² to your effective irradiance — it adds 70 W/m², because the rear face does not convert light quite as well as the front.
This coefficient is measured in a controlled lab environment and published on the module datasheet. You do not calculate it yourself in the field — you look it up. Most commercial bifacial modules have bifaciality coefficients in the range of 0.65 to 0.80, meaning their rear face is 65–80% as efficient as their front face.
Imagine you are filling a bucket with two hoses. The front hose delivers water at full flow. The rear hose delivers water, but it is slightly narrower — it only delivers 70% as much per minute. The total fill rate is not just the front hose alone. It is the front hose plus 70% of the rear hose. The bifaciality coefficient is the "narrowness factor" of the rear hose. Geff is the combined fill rate, expressed as if it were all coming through one full-size front hose.
Two Versions of the Formula
There are two versions of Geff depending on what you are measuring. The only difference is which bifaciality coefficient you pull from the datasheet.
| If You Are Measuring | Formula | Coefficient to Use | Why |
|---|---|---|---|
| IV Curve / Pmax | Geff = GF + (GR × φPmax) | φPmax — maximum power bifaciality | You are correcting a power measurement, so you weight rear irradiance by how efficiently the rear face converts it to power |
| Short-Circuit Current (Isc) | Geff = GF + (GR × φIsc) | φIsc — short-circuit current bifaciality | You are correcting a current measurement, so you weight rear irradiance by how efficiently the rear face generates short-circuit current |
For most commercial bifacial modules, φPmax and φIsc are very close in value — typically within one or two percentage points. In practice, using one where the other is called for introduces only a small error. But it is worth using the correct one when the datasheet makes both available.
A Worked Example
Here is what the calculation looks like with real numbers from a typical mid-day commissioning test on a bifacial string.
What Happens if You Skip Geff
Front-side irradiance only
You correct to STC using GF = 850 W/m². The correction divides by a smaller number than the module was actually working with. Result: STC-corrected Pmax appears slightly high relative to nameplate. At commissioning, this creates a false baseline that overstates string performance. Future O&M comparisons against that baseline will show apparent degradation that is partly artificial.
Front and weighted rear
You correct to STC using Geff = 918.4 W/m². The correction accounts for all the light the module was actually working with. Result: STC-corrected Pmax reflects the string's true performance. The commissioning baseline is accurate and defensible. Future O&M measurements corrected the same way will produce comparable results.
The commissioning baseline is the reference point for the entire asset life. An incorrect Geff at commissioning — or the absence of Geff entirely — creates errors that compound every time a future measurement is compared against that baseline. Getting it right once at commissioning protects every subsequent performance comparison for the life of the system.
Frequently Asked Questions
Do I need Geff for every bifacial string test?
Yes, any time you are correcting a bifacial string measurement to STC and comparing it to a nameplate value or a previous measurement. If you are only doing relative cross-string comparisons on the same day under identical conditions — for example, Isc screening to identify outliers — the Geff offset is the same for all strings and does not affect the relative ranking. But as soon as you apply a correction to STC, you need Geff.
What if my IV tracer does not let me enter a custom Geff?
Some older IV tracers accept only a direct reference cell reading and apply a fixed correction formula using that value alone. In that case, you have two options: perform the STC correction manually in post-processing using the correct Geff, or use a tracer that supports a custom irradiance input. For commissioned bifacial sites where the data will be used for performance verification or warranty purposes, manual post-processing is preferable to accepting an incorrect automated result.
What if I do not have a rear-facing sensor?
Without a rear irradiance measurement, you cannot calculate Geff. You can estimate rear irradiance from ground albedo and a view factor model — which IEC 61724-1 Method 1 (albedometer) supports — but this introduces additional uncertainty. For commissioning work with contractual implications, a rear-facing reference cell is not optional. It is the minimum instrumentation needed to produce defensible results on a bifacial array.
Does Geff apply to monofacial modules?
No. For monofacial modules there is no rear contribution, so Geff equals GF. The formula simplifies to the standard correction most technicians already use. Geff only differs from GF when there is a bifacial module producing power from both sides.
Why is the rear irradiance lower than the front?
The rear face sees reflected light — primarily from the ground surface and adjacent structures — rather than direct sunlight. Ground surfaces typically reflect 15–50% of incoming irradiance depending on the material (grass, soil, gravel, concrete). The rear sensor therefore reads significantly lower than the front sensor under the same conditions. That is expected and correct. The bifaciality coefficient then further discounts the rear reading to account for the rear face's slightly lower conversion efficiency.