Dirt Accumulation on Module Frames or Laminate Edges

In either framed modules or pitched-roof-integrated laminates with sealing elements that are sized for high annual output and have relatively small tilt angles (0° < b < 35°), after rainfall a small amount of water containing dirt is always trapped behind the glass and frame of each module. A few centimetres above this area, a layer of dirt that is not completely washed away by rain tends to accumulate.

This phenomenon mainly poses a problem in cases where, to avoid shading from solar generators mounted further in front (e. g. on shed roofs; see Figure 1.11), modules are mounted horizontally, and if there is virtually no clearance between the module frames and the solar cells. This same phenomenon can


Figure 4.90 Moss growing on the underside of the fouled lower edge of a framed Siemens M55 module

also occur with solar roof tiles that exhibit high frames. Although striving to achieve optimal module efficiency by minimizing the spacing between solar cells and module frames may seem like a good idea on paper, in actual installations this practice promotes dirt soiling that reduces output. Even a few millimetres of clearance with tilt angles around 30° substantially improve the situation and also ameliorate module surge and lightning current resistance.

After manual cleaning, the output of such modules improves by several percentage points. In the case of the 60kWp Bern University of Applied Sciences test installation, which is around 50 m from a heavily used railway line near a railway station and therefore subject to soiling in the form of a rust film, the potential output increase from cleaning was just under 10% after more than four years of opera­tion [4.6], [4.7]. Figure 4.92 displays a cleaned and an uncleaned array at the test installation, along with the lower module edge soiling that accumulated over a five-year period. Figure 4.93 displays a close-up of such soiling in a solar generator with a 30° tilt angle.

After such modules are manually cleaned, dirt gradually re-accumulates. When evaluating the long­term performance of solar generators, the solar generator correction factor kG = YA/YT (see Chapter 7) as a function of time should be determined.

For PV installations with relatively shallow tilt angles, it is best to determine the kG waveform for April to September, when module output is not impaired by either snow or shading – providing no unusual events such as an inverter failure, cleaning or alterations occurred during the period. Inasmuch as temperature is already factored into YT, ideally kG should be as close to 1 as possible and not time dependent. Figure 4.94 displays the solar generator correction factor waveforms for the summers of 1994


Figure 4.91 Large expanse of lichen growth that has developed over the years on the edge of a framed M55 Siemens module. Slight front-contact delamination is also visible here


Figure 4.92 Solar generator for the 60kWp PV test facility at Bern University of Applied Sciences in Burgdorf (b = 30°). The module on the left has just been cleaned. Dirt that accumulated between June 1993 and May 1998 is visible at the lower edge of the module on the right. The module surface exhibits relatively little dirt

to 2009 for the Bern University of Applied Sciences PV Lab test installation array that was monitored for the longest period.

Figure 4.94 shows that: (a) the solar generator kG and thus energy yield decreased slowly at first, but at an increasingly rapid rate after a number of years; and (b) lengthy periods of winter snow cover slowed the decrease in kG. The installation cleanings realized in 1998, 2002 and 2006 using a strong cleaning agent (Transsolv) successfully reversed most of the measured power loss. Prior to these cleanings, the power loss relative to the installation’s baseline state was around 10%, 12% and 8% respectively. When measurements were performed in the summer of 2002, a faulty solar generator


Figure 4.93 Close-up of pollen – and insect – dropping-induced soiling of the solar generator in Figure 4.92 (framed Siemens M55 modules) during a lengthy springtime dry spell. The clearance between the cells and frames is extremely small (around 0 to 3 mm). Most of the soiling occurs just above the lower edge of the module

PV Sub-Plant West of BFH-TI in Burgdorf:
Generator correction factor in sunny season (April – October)

Figure 4.94 Solar generator correction factor kG for an array at the Bern University of Applied Sciences PV Lab installation, with the installation cleaning dates. The installation has horizontal framed M55 modules with a tilt angle of 30°, is located near a railway line and tends to become heavily fouled. The low baseline kG is partly attributable to the elevated DC power loss in this test installation relative to that of a normal PV system in the west array that was responsible for around 1% of the observed kG reduction was replaced. Cleaning in 1998 and 2006 eliminated 8% of the power loss in each year, and an additional 10% was eliminated via the 2002 cleaning. Part of the approximately 3% power loss incurred over a 12-year period appears to be irreversible, however, probably on account of (a) slight soiling that occurs soon after cleaning and (b) permanent changes in the glass surface properties at the soiled lower edges of the modules. Changes in the module cells and foil may also be a contributing factor in this regard.

Soiling is less of a problem with: (a) vertically mounted modules as in Figure 2.44; (b) relatively large modules; and (c) 40°-50° tilt angles insofar as rainwater runoff is relatively robust (and thus has a stronger cleaning effect) and there is sufficient rainfall. At another installation located near a railway station and a heavily used railway line, after around 20 months of operation the installation’s vertically mounted modules exhibited a soiling-induced power loss of up to 7% [4.9]. Comparable power losses have been observed at other installations (for which no operating periods are indicated), in most cases ranging from 2 to 6% and in rare cases as high as 18% [4.8].

However, at steeper tilt angles, the cleaning effect of rainwater may be somewhat negated. Figure 4.95 displays the lower edge of a b = 65° and g = —61° module whose bottom solar cells exhibit appreciable expanses of deleterious moss and lichen.

Updated: August 8, 2015 — 7:03 am