A photovoltaic noise barrier is a noise barrier that also generates electricity, in that order. If the acoustic section is compromised to fit the panels, the project has failed at its first job and no amount of generation recovers it. That is why the cost of a PV barrier is best read as a normal barrier plus a power plant bolted onto it, rather than as a new product with a new price.

The cost stack, in the order it moves the total

1. The barrier itself

Everything that a plain barrier needs: posts, panels, foundations, and the connection to whatever is below. This part is not made cheaper by adding solar, and on a long run it is still the largest single line.

2. The photovoltaic layer

Modules or laminates sized to the panels, the cable routing between them, junction boxes, and the mechanical interface that holds the module in the panel frame without breaching the weather seal. This layer costs more per square metre than plain metal, and it is quoted per square metre of barrier face rather than per panel.

3. The electrical balance of system

Inverters, DC and AC protection, metering, earthing, monitoring, and the connection to the grid or to the site distribution. On a linear barrier the cable runs are long and the inverter has to be placed somewhere accessible, which is a civil cost as much as an electrical one.

4. The structural and access allowance

The most frequently underestimated line. A PV panel is heavier than a plain metal cassette, and the wind load on a barrier is already high because it is a vertical surface. Posts and foundations sized for a metal panel may not carry the module plus its mounting. Access for cleaning and for replacing a failed module also has to be designed in.

Why the module is rarely the deciding number

Buyers usually ask for the module price first, and it is usually the least informative figure in the quote. The module is a commodity bought in a competitive market; the cost that varies between projects is everything around it. A short barrier run with a difficult grid connection can spend more on cable, protection and connection than on the generation surface itself.

Why the yield is not simply area times sunshine

This is where PV barriers differ most from rooftop systems, and where optimistic business cases are usually built.

• Vertical orientation. A barrier face is vertical. It receives much less irradiation than a surface tilted towards the sun at the site latitude, and the loss is largest where the sun is high at midday.

• Fixed azimuth. The barrier faces the road or the track. Whatever direction that is, it is the direction you get.

• Self-shading and structure. Posts, frames, the roof of an enclosure and the barrier on the opposite side of a narrow corridor all cast shade at some hour.

• Soiling. A roadside or trackside vertical surface collects dirt, brake dust and exhaust film, and cleaning access is limited. Soiling loss on a vertical surface in a traffic environment is not negligible.

• Temperature. A dark module behind a sound-absorbing build-up runs hot, and hot modules deliver less.

• Inverter sizing. The array rarely produces its nameplate output, so the inverter is sized below it, and the peak is clipped by design.

Where a PV barrier makes sense, and where it does not

It makes sense where the barrier is long, where the face is reasonably oriented towards the sun, where the electricity is consumed on the same site at the same time, and where land for a separate array does not exist. In those conditions the structure and the land are already paid for by the acoustic requirement, and the PV layer is the only incremental cost.

It does not make sense on a short run, on a face that is shaded or facing away from the sun for most of the day, where the grid connection is the expensive part, or where the acoustic specification is tight enough that the module cannot occupy the panel area the acoustics need. In those cases the honest answer is a conventional barrier, and the money is better spent on the acoustic performance.