Getting solar: from research to install

East- and west-facing roofs are still fine for solar installations, if shading is at a minimum
How our “no good for solar” roof was actually just fine. By Stephen Zuluaga.
This article was first published in issue 142 (Jan-Mar 2018) of Renew magazine

Until recently, I’d thought solar wouldn’t work well on our house. With little north-facing roof to speak of, I just assumed that solar wouldn’t be worth it. But then I began to read about some of the good outcomes possible with an east/west array—our roof has lots of east/west space and shading issues only at the extreme ends of the day.

Although an east/west array will produce less overall than a north-facing one, it can extend generation hours, both earlier in the morning with an east-facing array and later in the day with a west-facing system. Long generation hours are important if you don’t have battery storage and the gravy train of premium feed-in tariffs has left the station. It means you can match more of your generation to usage, particularly before and after work usage, and hence increase your ‘self-consumption’ of solar—this will mean lower grid imports and a shorter payback period.

Modelling the economics

Before committing to a solar purchase, I was interested to more fully understand the financials. I found ATA’s free Sunulator tool which helped me model a scenario based on my actual electricity consumption and the combined north/east/west PV configuration I was contemplating. Sunulator is a great tool—if you’re planning solar you should use it [Ed note: ATA also has a simpler tool available to give you an indication of the financials without the full modelling of Sunulator.] The energy analysts at the ATA helped with understanding the Sunulator results as one of the ATA member benefits.

Using a few PV system sizes, our location and orientation, Sunulator provided an estimate of generation over a year for these different sizing scenarios. We ultimately went with 9.72 kW but also modelled 10.2 kW and 10.8 kW. It then matched generation with our household’s actual electricity consumption over the past year and our electricity and feed-in tariff rates (allowing for forecast increases), to provide an expected payback period of seven years. The expected payback is based on the real quote prices you put into it. In my case, this represents a 16% ROI. I wonder how many people think about that when they consider solar—they should! I’ll be interested to track the estimated generation and payback against the actual results over the next few years.

Choosing the technology

As well as looking at the financials, I also investigated different panel options. There are a wide range of panels available, with varying price tags and panel wattages, and options such as DC optimisers and microinverters to consider. I looked at both DC optimisers and microinverters as I was concerned to maximise generation (these maximise generation for shaded arrays and also if there’s dirt on a panel), but in the end I decided the additional cost per watt was too high; Sunulator helped a lot with that conclusion.

One thing with solar is that different installers will work with particular panels and inverters, so you can’t just choose any panel that you want. I got around this by getting a wide variety of quotes from different installers.

My quoting experience was mixed. I used a quote aggregator service to find companies to quote. I also had an existing relationship with a supplier who helped me convert my house to all-electric last year, and they’re the ones I ultimately bought solar from. Some of the companies I spoke to about quoting were super proactive and others didn’t even bother to call me back. There were a few suppliers who would argue about how their one brand or type of product was amazing and everything else was rubbish. I walked away from those operators, particularly as every question I asked was met with an answer explaining why my question wasn’t relevant.

The quote I ultimately bought from was done completely via the internet and Nearmap (a paid high-resolution mapping service similar to google maps, but with higher resolution and more up-to-date images). My chosen installer came out to confirm the final details of the install before they started.

In the end, I chose an installer/system based on price and online reviews—the latter pointed to no known quality issues with the hardware. I have 36 Jinko 270 W panels (14 north, 11 east, 11 west) with a Fronius Primo 8.2-1 inverter. With 9.72 kW of panels on an 8.2 kW inverter I could lose some generation (it gets clipped by the inverter if it goes above 8.2 kW), but my north/east/west system will rarely get to this maximum output anyway, and the extra panels mean I get the benefit of more generation earlier and later in the day, or in cloudy conditions.

Don’t overlook the detail

My switchboard needed to be upgraded as all the circuit-breaker slots were full. This wasn’t identified until after I purchased, but I had seen it coming; and I now have space for a circuit for an electric vehicle charger one day. I’m not an electrician, but my understanding of electricals was helpful in communicating with the electricians, particularly about how I’d like to maintain a tidy switchboard and not have modules and conduits scattered all over the garage wall.

Working with distributor export limits was pretty simple in my case as my system was smaller than the distributor limit: I have an 8.2 kW inverter and my distributor, United Energy, limits a single phase to 10 kW. However, it’s important to check this with your installer as many distributors limit systems to less than this. It will affect the size of system (or inverter) you can install—or the inverter will need to be export limited via a setting on the inverter, something the distributor needs to approve. The installer should manage this approval process, but it’s good to be in the know, as this may also affect your payback period.


Once I approved the quote, installation happened about two months later. This was partly due to having my roof restored and painted and it not being walkable for a few weeks after, and also delays coordinating a time that suited both me and my installer.

On the day of the installation, there were a couple of things that I wanted input on. One was the location of the inverter—at my house we have a garage which worked as a good location. The installer will let you know the constraints, but it’s good to avoid north- or west-facing walls. A shaded area is ideal as the inverter runs less efficiently when it’s hot and may even overheat and shut down. Heat can also accelerate the deterioration of the electronics over time. On a warm day, my non-contact thermometer shows the inverter’s front panel to be at 55 °C with its fan running at full speed.

The installation involved several steps:

  • installing the inverter
  • installing the circuit breakers and meter  in the switchboard
  • cabling from the inverter to the switchboard
  • installing the tilt frames on the flat roof section (each frame cost about $50)
  • installing the panel rails
  • installing the panels
  • installing the DC isolation switches
  • cabling between the panels and the inverter via the isolation switches.

I’ve heard of longer installation times, but my installation took two half days for three people. Watching them work on the sloping roof was impressive. They used safety harnesses and it’s physical work, hefting the panels up from the ground and then into position. These guys obviously opt-in to the functional fitness classes at their local gym.

Signing over STCs was easy—that was done during the sign-off and invoicing process after installation.

Being in Victoria, once the system was installed, the installer did a basic commissioning of the system to check its operation, then explained that the system needed to remain off until the distributor had inspected the system. [Ed note: this connection process can differ by state; see for details.] About two weeks after the installation, an inspector phoned to arrange a time, as they needed physical access to the inverter and switchboard. The system got the tick of approval and the inspector turned it on.

The next step was meter reconfiguration (which happens remotely) to enable the meter to measure the solar feed-in, required before the retailer starts to pay a feed-in tariff. I’ve heard that some distributors have taken two months to do this (in Victoria, the requirement is for this to be done within 20 business days of the request from the retailer). I was fortunate—mine was done the next business day after the inspection report was forwarded to the retailer. I had a $67 fee passed on to me in my next bill for the meter reconfiguration. This fee varies across distributors.

You can monitor everything

The inverter came with a built-in communications card and I also had a meter installed in my switchboard which communicates with the inverter to give me up-to-the-second generation, consumption and grid usage/feed-in data. Both the inverter and smart meter are Fronius and I’ve found this a highly impressive system, for both the networking and analytics. All the solar electricians I’ve spoken to say the electrical side is equally good.

Actual generation so far has exceeded expectations (slightly higher than the Sunulator estimates).

There were no issues along the way during the project and I’m very happy with the company I bought the system from, Pure Electric. I have no regrets post installation, perhaps because I was so proactive before everything went in. All the parties involved in the solar install were easy to deal with, very professional and patient with my many questions. I know this isn’t everyone’s experience with solar. I think my experience in the purchase and configuration of complex IT equipment transferred over to solar purchasing quite nicely.

About the author
Stephen is a co-founder of training business Peer Camp and a teacher at RMIT, with a strong interest in technology and energy efficiency. You can see his article on his home's conversion to all-electric in ReNew 140.
This article was first published in issue 142 (Jan-Mar 2018) of Renew magazine
Product profile: Portable solar recharge hubs

Product profile: Portable solar recharge hubs

We’ve covered a few electric scooters over the last few years, and a recent entrant, Benzine Zero (Italian for zero petrol) has added some nice looking offerings to the market.

Read more
Product profile: Solar in a box

Product profile: Solar in a box

We’ve covered a few electric scooters over the last few years, and a recent entrant, Benzine Zero (Italian for zero petrol) has added some nice looking offerings to the market.

Read more
Big solar at Melbourne Water

Big solar at Melbourne Water

Rechargeable lithium batteries are critical for our modern world, but they do have a somewhat variable safety history. Lance Turner looks at the issues and what to do about them.

Read more