WHITE PAPERS

Solar Panel Performance Survey

by James Hoffman

Motivation

Solar panels have two dominant performance metrics: economy and efficiency. These metrics are commonly measured as ratios of a panel's peak power output to its purchase cost (economy) and to its aperture (efficiency). For PV panels, peak power output is the electrical specification peak watts (Wp). Given that, economy is computed as Wp per unit of equipment cost, and efficiency is computed as Wp per unit of maximum aperture (corresponding to the panel's frontal area).

Whereas comparisons of solar panels on the basis of economy are straightforward, comparisons on the basis of efficiency are often muddled by the mixing of different measures such as module efficiency and cell efficiency. A comprehensive comparison of existing PV panels that uses a consistent method to measure efficiency and examines the relationship between economy and efficiency could provide important data on the economic value of efficiency. The stronger the inverse correlation between these two variables, if any, the stronger the economic value indicated for efficiency. The present study plots panel models and types of panels in the two-dimensional space of value and efficiency to examine the relationship of these two variables.

Data Capture

A database was created to record, for each of a broad cross-section of available PV panels, data including:

  • Peak power output
  • Physical dimensions
  • PV cell type
  • Weight
  • Manufacturer
  • Zero or more price quotes

PV panels are typically manufactured and sold as product series, each series having three to seven models sharing the same cell type and physical characteristics such as dimensions and weight, and distinguished by power output. Thus it was natural to organize the database as three types of objects:

  • Panel Family: A PV panel product line sharing physical characteristics, including PV cell type, physical dimensions, and weight
  • Panel: A PV panel model belonging to a Panel Family and having specified electrical characteristics including peak power output
  • Source: A quote for a PV panel model including the unit price, minimum quantity, date of the quote, and the publication source
PanelFamily
name:SW155-175mono
manufacturer:SolarWorld
type:MCSi
dimensionsMm:1610*810*34
weightKg:15
url:http://lib.store.yahoo.net/lib/wind-sun/SW-175-Mono.pdf

Panel
model:SW-175
familyName:SW155-175mono
power:175

Source
panelModel:SW-175
priceD:555
date:2009.10.11
url:http://store.solar-electric.com/so175wamosop.html

The database is stored as a set of flat-files containing entries describing objects of these types. The excerpt to the right contains three entries: the first specifies the panel family designated 'SW155-175mono'; the second specifies the panel model 'SW-175' belonging to that family; and the third gives a price quote for that model.

The database is instrumented with different views of its contents. One view is the following list of panel models with links to the specifications and quote sources.

Each row in the table corresponds to a model of solar panel. The Specs column provides links to the specification sheets describing the physical and electrical characteristics of each model of panel, usually as part of a family of models sharing the same physical characteristics. The Economy column provides links to sources of price quotes, if any, for each model of panel.

Some panel specifications provide module and/or cell efficiencies, and the database records these when present. However, in order to provide consistent performance metrics, efficiencies reported by views of the database are computed by dividing its peak power output by frontal area, where a panel's frontal area is found by multiplying its height and width dimensions.

Plotting Performance

The main objective of this survey is to examine the relationship between the two most important performance characteristics of PV panels, which can be described by the dependent variables 'efficiency' and 'economy'. Alternatively, they can be described by the two related dependent variables, 'space' and 'capital', which are the inverses of efficiency and economy, respectively. Whereas efficiency and economy measure panel performance in the domain of value to the customer, space and capital measure their performance in the domain of resources expended or footprint. The two domains are simply inverses of each-other.

Domain
Resource or Footprint
(smaller is better)
Value
(larger is better)
variable:
description:
typical units:
Space
Occupied space per unit of output
m2 / kWp
Efficiency
Output per unit of space occupied
kWp / m2 * 100 (percent)
variable:
description:
typical units:
Capital
Equipment cost per unit of output
$ / Wp
Economy
Output per unit of equipment cost
Wp / $

The relationship of the pair of dependent variables -- whether space and capital or efficiency and economy -- is visualized by plotting the locations of solar panels in a two-dimensional space, where the variable relating to occupied space maps to the horizontal axis, and the variable relating to equipment cost maps to the vertical axis. The horizontal coordinate for a given panel model is well-defined by its physical and electrical specifications. The vertical coordinate is a function of the price, which typically spans a range when there are multiple quotes. In the plots, a panel model appears as a disk when it has one quote, a series of disks connected by a vertical bar when it has multiple quotes, and a faint vertical bar when it has no quotes and it is assigned an estimated price range.

Results

Graphical views of the database based on the two-dimensional domains of footprint and value enable visualizing the relationship of the dependent variables.

One set of views plots positions of panel models in the two domains and another set of views plots the extent of groups of panel models sharing the same PV cell types.

Resource domainValue domain
Panel models
Panel types
Click on graph to view full-sized SVG version. NOTE: Use the Opera Browser for best results.

The green trend line -- which is the same contour in both domains -- is a hyperbola connecting points of equal resource cost defined my multiplying capital and space footprints.

The distribution of the data indicates a strong inverse correlation between the pairs of variables, such as economy and efficiency. Several features of the plots are noteworthy:

  • Some of the amorphous silicon based panels have about double the economy of most crystalline silicon panels, but have less than half of their efficiency.
  • Hybrid and enhanced mono-crystalline silicon panels with percentage efficiencies in the upper teens command double the price per peak watt of mono- and poly-crystalline silicon panels in the large cluster whose efficiencies are in the lower teens.
  • Panels of types that score lower in the value domain and depart from the trend indicated by the green line, such as CIGS and laminar silicon, tend to be used in speciality applications that do not compete directly with the more mainstream panels.

No panels for which we could obtain specifications surpassed 20 percent in efficiency. However, the panels at the upper end of the efficiency range tend to cluster closely around the green trend line. A doubling of efficiency from 10 to 20 percent roughly corresponds to a doubling of the value of the panel, given that a 20-percent-efficient panel commands about twice the price of a 10-percent-efficient panel of comparable output. Continuity of the underlying economic equations suggests that increases of efficiency well above the 20 percent limit would result in further substantial increases in value.

(C) Sun Synchrony and James Hoffman 2010