Solar Farms: String or Central Inverters?

Solar Farms: String or Central Inverters?

solar_farmWith the current boom in the installation of new large scale (>MWp) solar farms in the UK, we have ended up being instrumental in the development of several of these projects so far. Often the developer involved has an inverter solution and manufacturer in mind already, but occasionally we have been asked to provide guidance on the most suitable solution for this kind of large project.

There are essentially two ways to go. You can either use very large central inverters (100s of KVA usually), or use smaller string inverters (10s of KVA usually) although these are not to be confused with the single phase string inverters used in domestic installations. We’re still talking some pretty big bits of kit.

Most of the big inverter manufacturers offer both options. SMA, Samil Power, Power One, Solar Max etc all offer both in various configurations and of various design philosophies. It would be perfectly possible to get bogged down forever in the details of inverter design, but I’d just like to consider inverter choice for now.

There doesn’t seem to be much existing guidance out there, which I suppose is understandable for big capital projects, as they are always bespoke to one degree or another or at least not widespread enough for forums to spread outside of the organisations planning and designing them.

An inverter that is undersized in this way will spend more of its life operating at a more optimal load. The advantage of this approach is that during the large part of the year, and the course of a single day, when the PV array is producing less than about 20% of its peak output, the inverter will be working at a higher efficiently and more energy will be captured than with a larger inverter. This does mean that a small amount of energy will be lost on the rare occasions that the PV array is operating at peak power (Usually only around mid-day in mid-Summer) but this small penalty is worth the gains at every other time.

I don’t find it helps a great deal approaching the inverter manufacturers for guidance for various reasons. I’ve found that they, as you would expect, will often promote different products at different times depending on what they are pushing right then. Their sales people are often reasonably knowledgeable (Their technical people are almost always excellent, but you probably wouldn’t be dealing with them up front) but will not usually provide actual guidance to you, rather taking their lead from what you first enquire about. They are also often unwilling or unable to seriously consider the other design aspects for each option such as cable runs and ohmic losses and BOS costs, which can have a huge impact on overall project cost and long term yields.

If you’re planning the installation of one of these projects it’s worth considering your inverter choice as early as possible. With roof based solar systems it’s usually the roof area available that dictates most of the further project details, but with solar farms the whole approach of the project, from G59 applications, system design all the way through to installation, project cash flow, commissioning and O&M contracts can all be dictated by the choice of inverter which may have been made early on in the project without careful thought.

First let’s look at the really big Central Inverters.

central_inverter

Power One Aurora Station- shows the inverter housing on the left and the MV transformer housing on the right – approx 2.5m high weighing several tons

Central Inverters are usually somewhere between 35kVA and 2MVA depending on context, but are usually free standing units either sitting in a plant room or an enclosure outside. More and more they are being designed with Solar Farms in mind and so come in an IP54 or IP65 enclosure and often associate with a MV transformer for connection to the 11KV or 33KV local distribution network. As I’m sure that conveys, these are pretty serious pieces of kit. They are almost always made to order and costs tens or hundreds of thousands of pounds each. They are actually usually a collection of smaller inverters packaged together into one unit, although not always.

Advantages of Central Inverters are (To my mind anyway)
  • Engineered with large projects in mind
  • Comprehensive support from manufactures (design, delivery, commissioning, O&M
  • Well defined O&M practices
  • usually can be repaired in-situ (Modular construction)
  • Simplified LV design – connect straight to MV, by the DNO or other contested works provider
  • Lower ohmic losses (given correct design of DC system, as high voltages and low currents are maintained for longer
  • Better monitoring options, with a string combiners and comms packages usually offer on the fly Power Factor control, which isn’t much of an issue in this country yet as renewable penetration is still low, but will become more of an issue as we install more, and may be required retroactively by the DNO’s. Lower /Wp cost

 

Disadvantages of Central Inverters
  • Very Expensive individual units – Cash flow issues
  • Long Lead times from some manufacturers as made to order
  • Usually very heavy and could require the construction of new roads for delivery
  • Relatively low MPPT resolution, often with only 1 or 2 covering thousands of modules – Mismatch losses
  • Long and elaborate DC system required – outside a lot of designers and installers comfort zones
  • Usually require 1 MV transformer for every 1 or 2 inverters which can be expensive and complicated depending on system size

sunny_tripowerAlthough large Central Inverters have been used extensively on the continent and the rest of the world we seem reluctant to use them in the UK. I think that one of the main reasons is probably because a fair number of the people involved are more comfortable with string inverters. This may change as more solar farms experts from the continent are starting to get involved in the UK market.

String Inverters (often also called Mini-Central Inverters, as they are actually still pretty big) are usually somewhere between 10KVA and 35KVA and are usually IP65 sealed units that are wall, or frame mounted. You would usually have 10 or 12 of these combined at a LV distribution boards, and then as many of these boards as required connected at a main panel board close to an MV transformer. They have their own characteristics that either lend them to a project or make them less desirable.

 

SMA Sunny Tri Power- one 20KVA unit, 665mm/680mm on a side and weighing 45Kg

String Inverters Advantages
  • Much cheaper unit costs, easier cash flow
  • much lighter units so no special access ways required
  • Can usually be mounted on module mounting frames
  • Not that different (sometimes the same) to inverters used in normal commercial PV installations, so more familiar
  • Shorter and less complicated DC systems
  • Higher MPPT resolution, so in principal, less mismatch loss
  • MV transformers can be sized to existing ranges more easily
Disadvantages of String Inverters
  • Usually Higher /Wp cost
  • Less likely to be well integrated for system monitoring
  • Usually need to be replaced rather than repaired in-situ
  • Require much more elaborate LV system, so long high current runs at LV – high ohmic losses
  • Require more elaborate LV protection systems with large numbers of IP65 LV distribution boards

It seems that the current advice from those involved in solar farm projects is that string inverters work out as the more cost effective solution below about 10MVA.

During the course of my work I’ve carried out reasonably detailed turnkey costing of both 5MVA and 15MVA solar farms, and analysed both options, string or central. As far as I could tell the costs worked out broadly the same whichever way you go, maybe with a great deal of detailed work you could find a distinct cost advantage one way or the other, but in a fast moving market it’s rare to get the chance for such an in depth study before a project is up and moving. So I suppose it comes down to a judgement call. If you assess the limitations of your project and decide appropriately then either can be made to work.

If you’re going for lowest possible cost, fit and forget, I suspect that most people would chose string inverters. There are more people available who are familiar with them and LV electrical systems so you have more labour options and cash flow will be easier to manage.

For those looking to own and operate a solar farm for its whole lifetime I’d probably go for the central inverters myself as they are much more carefully engineered for this kind of project and offer much better monitoring, control and operation and maintenance possibilities. They might be more difficult to install in the first place, but should reward the effort in the long run.

If you have any questions I’d love to hear from you, please leave a comment or send an email.

Want to know how we can help you?

Contact Form Callback Request