Introduction to the new MCS Solar PV Guide
The UK's Microgeneration Certification Scheme (MCS) has published its Solar PV Guide called 'Guide to the Installation of Photovoltaic (PV) Systems' detailing in 124 pages the updated requirements for new solar PV installations under rated outputs of 50 kW.
As a result of this, MCS installers working under many of the subsidy schemes will soon be required to record a shade or horizon line on a sunpath diagram to present to clients. Other changes include a revised performance calculation and amended electrical safety advice. These were recently announced by the Energy and Climate Change Minister, Greg Barker, with the intention of raising the level of MCS certification.
In this respect, the PV Guide has provided much-needed clarity on the inspection processes.
However, its scope indicates that the guide is also a statement of best practice. This statement deserves further scrutiny, as there have been recent technological advances which have often come from the well established global markets of Germany and USA. One ‘litmus test’ of best practice is whether the industry has been given sufficient opportunity for consultation. This is normally assumed for government policy as evidenced by the existence of their Code of Practice for Consultation. Given the minister’s involvement in the guide’s launch, it might be expected that the consultation need has been met. However, as a recent poll in the Solar Power Portal shows, the majority of respondents do not agree with all of the PV Guide and it seems that a large part of the industry would like to have further say in the guide contents.
Shading implications for design
Identifying shading is one of the cornerstones of successful performance prediction for solar arrays. It is not always appreciated that at every point in the UK mainland, at least 2% of the sky is already obscured by the landmass that sits above sea-level. This reduction can be far greater than this minimum percentage due to additional landscape features such as hills and buildings on the horizon. However, it is shading objects near to an array that have the greatest effect as these are most likely to cause hard shadows by blocking out the direct or beam solar radiation. Therefore, it is no longer considered sufficient to record a simple silhouette outline of the horizon, as it is also important to check a nearby object’s distance from the array.
For this reason the MCS method calls for any object less than 10 metres from any part of the array to be treated differently in the calculation. Installers are required to record a horizon line identifying all shading objects, both near and far, on a sunpath diagram (see Image 4).
Shading objects that are within 10 metres then require a circle to be drawn onto the sunpath diagram. The apex of this circle should sit at the top of the shading object and the radius of the circle should be the height of the object (see Image 5). In this example, the green part of the line represents shading objects that are further than 10m away while the red part of the line represents a large building which is within 10m of the proposed array.
In accordance with the MCS method a circle has been added (dotted), with the top of the object at the apex of the circle.
Here we see divergence from German best practice which identifies the risk of nearby shading objects according to the width of the object as well as its height. In this case, the MCS method does not correctly identify objects (such as buildings) that will cause a hard shadow even from over 100 metres away. The MCS method also makes the calculation of shade effects by nearby overhead power or telephone lines quite onerous, when in reality these objects are often less than 5cm in diameter and so do not cause hard shadows over 5 metres away. More details of the MCS method can be found here.
Furthermore, the MCS method does not make a distinction between types of shading objects: in particular their colour and seasonal variance. Trees, for example, can either be deciduous or coniferous, meaning they have substantially different shading potentials even though they measure with equal silhouettes in the summer.
However, it is the reflectance of nearby brightly coloured surfaces which is the most seriously underestimated factor. Reflectance can increase the solar radiation reaching a vertical array by as much as 20%, such as in the case of a large brightly-coloured commercial building facade. It is imperative that this is considered before an inverter is selected as otherwise this could be hopelessly undersized.
In terms of achieving a successful design, perhaps the biggest flaw in the MCS shading method is in the way it averages shading across a large array. Best practice here would suggest that shade measurements should be taken from all the extreme array edges as well as the array centre or base, allowing the results to be amalgamated. Instead, the guide suggests that the surveyor will need to ‘stand’ at a single position and gather all shade measurements from there. Of course, on a pitched roof that would be quite difficult and risky. If we are seeking best practice then we can compare this to the established techniques from the USA which quickly use electronic devices to collect multiple readings from around the array location, all from the safety of ground level using an extendable pole.
In effect, the USA method digitises the surrounding landscape, allowing instant results to be shown to the customer as well as further detailed analysis to be undertaken on a computer.
Given that there is already strong encouragement from the government to streamline the design process, it is surprising that we do not see this represented as best practice rather than requiring installers to become skilled at drawing landscapes onto graph paper. Instead of automating the process electronically, the MCS method requires extra steps of manually counting over 80 segments in a sunpath diagram, requiring careful levelling and drawing skills to achieve any accuracy. This is not a robust method that would enable different surveyors to acquire the same measurement and it is therefore likely to lead to disputes.
Further errors occur with the MCS sunpath diagram method as this is given as the only one for all of the UK. This does not fairly account for the almost ten-degree variance in latitude between SW England and the highlands of Scotland. The industry has already provided fast, accurate digital solutions to many of these tasks and there seems little benefit in trying to reinvent the wheel.
Other aspects of the design process
In reality, the effect of shading and reflectance varies not only according to the number of module cells that are affected, but also the detail of the cell and bypass diode connections. The overall effect then relies on how the modules are connected in either series or parallel mode as well as the presence of maximum power point trackers (MPPT) in the inverters.
Electronic equipment such as inverters have certain ‘sweet spots’ which give optimum efficiency and can also offer extra contingency between voltage limits. The best installers will select array voltages that give plenty of leeway in the event of module degradation, which over time can vary. If this is not adequately considered, customers will be faced increasingly with the dreaded red flashing inverter warning lamp indicating their FiT payments have effectively stopped until the array voltage is back in tolerance.
The difference between string inverters and those that are module-mounted also dramatically changes the response of an array to a shadow as it passes over. Modules that are strung together with one MPPT are effectively linked together, which means that one shaded cell starts to affect the whole string voltage. This is why professional software simulation can now work in 3D to accurately track potential shadows over the module cells. MCS installers will be required to explain the difference in performance prediction between any other methods used, and there is bound to be a significant difference between manual and digital methods that input many more of the variables. My advice would be to hand the confused customer a copy of this article!
The best locations for PV arrays will always be those in which near-object shading can be avoided. The best way to avoid the onerous MCS shading calculation method is to pick sites with no shading above the ‘raw’ horizon. However, as pressure grows to use the south-facing aspects of buildings and land plots, there will inevitably have to be compromises. Local roof features, nearby construction developments, trees and inter-module shading of rows of modules will always present a challenge. To provide customers and loan providers with a reliable lifetime performance prediction requires the best in site surveying and shading simulation techniques.
Irrespective of shading, the MCS guide performance prediction calculation is insensitive to the different equipment efficiencies. It uses a ‘one size fits-all factor’ in its annual energy calculation set at 80%, which means that there is no difference as to whether the worst or best choice of equipment is made. This inflexible principle does not provide the motivation for installers to optimise designs or for equipment manufacturers to improve their materials. Indeed, it is the part-load performance which is most revealing about product choices for the UK. These margins for error are added to those from voltage mismatching that frequently cause inverters to fail to deliver their peak array power. Additionally to the above, any variance in D.C. electrical cables losses will not be considered despite the fact that this varies according to the cable current, diameter and length.
In many respects, the guide has missed many of the industry’s latest techniques for achieving a successful PV design. In this author’s eyes at least, the UK’s best practice for the design and performance prediction of modern PV systems of all sizes has yet to be published.