Solar Test Array: 16 Panels Compared Side by Side

By Brett Fisher, Connect Energy Solutions · Connect Energy solar test facility, Darra head office · Results recorded January 2021 · Page last updated September 2026

We stopped taking the brochures’ word for it. We have installed a solar test facility at our Darra head office, where we track the performance of sixteen leading solar panel brands side by side. The results of that testing determine which products we are comfortable installing for our customers — so your decision rests on measured performance rather than a datasheet claim.

This page publishes a full set of those results: sixteen panels from twelve manufacturers — budget names, mid-tier names, and the premium brands commanding a 60% price premium at the time — mounted together, wired individually, and monitored over the same period. The figures below were recorded in January 2021. We are publishing them because the finding still holds and the method is still the right one, not because every model is current. Several of the panels have since been discontinued and panel wattages have moved on; what has changed and what has not is covered further down.

How the test array was set up

The principle was simple: remove every variable except the panel itself.

All sixteen panels are installed on one array at our Darra head office, at the same orientation and pitch, and monitored across the same period. Same weather, same temperature, same soiling, same everything. The only difference between them is the panel.

The measurement comes from the hardware. Every panel type on the array runs through its own panel optimiser, which reports that panel’s production individually instead of folding it into a string total. That is what makes the comparison real: we are not inferring performance from a whole-of-system figure, we are reading each panel’s own output, side by side, in the same conditions, year after year.

That leaves one problem. The panels were not the same size. The array included a 280W Talesun and a 400W Canadian Solar, and a bigger panel producing more electricity proves nothing. So the results are normalised: kilowatt-hours generated per watt of rated capacity (kWh/Wp).

That single figure is the fair comparison. It answers the only question that matters when you are choosing between panels: for every watt you pay for, how many units of electricity do you actually get back?

The second measurement is value. Dividing the panel price at the time by the energy it produced gives a cost per kilowatt-hour generated. That is where the results get interesting.

Results: output per watt

Ranked by kWh per watt of rated capacity. The final column shows each panel against the best performer in the test.

Top to bottom, the spread is 1.69 down to 1.48 — a difference of about 14%. That is real, and over twenty years it is not nothing. But look at where the panels cluster: twelve of the sixteen sit between 1.52 and 1.65, within roughly 8% of each other. On a typical household system that difference is worth a few hundred kilowatt-hours a year.

Results: cost per kilowatt-hour generated

The same sixteen panels, ranked instead by what each kilowatt-hour of output cost to buy. Panel pricing is as at January 2021; the percentage column compares each panel to the cheapest energy in the test.

Output plotted against price

The two tables above hold the same sixteen panels ranked two different ways. Plotting one against the other is the quickest way to see the relationship — or the absence of one. Output per watt runs up the side; what each kilowatt-hour cost to buy runs along the bottom. If paying more bought more electricity, the points would climb from left to right.

They do not. The points scatter almost at random, and if anything the line slopes the wrong way: all five panels in the best-value zone also sat in the top half of the output ranking, while the right-hand side of the chart — everything above $0.30 per kilowatt-hour — is occupied almost entirely by panels that finished in the bottom half. The single exception is the panel sitting alone in the top-right corner. The point nearest the top-left — most output for least money — is the LONGi 375.

What the numbers actually showed

2.4% more electricity per watt. 105% more money for every unit of it.

The panel that won on output was the worst value in the test. The LG 360 topped the yield table at 1.69 kWh/Wp. It also cost $0.44 per kilowatt-hour generated — more than double the cheapest energy in the array. You were paying a 105% premium to buy a 2.4% advantage over the next panel down.

The second-placed panel was the best value. The LONGi 375 produced 97.6% of the LG’s output per watt and did it at $0.22 per kilowatt-hour. On the numbers, it was the standout of the test: near-top yield at the lowest cost of energy in the array.

Premium branding did not predict performance. The three lowest positions on the yield table were held by two SunPower panels and a Q Cells Duo — all premium-priced names at the time. They finished 14th, 15th and 16th of sixteen. Meanwhile a Talesun 280, one of the cheapest panels in the array, finished 11th and returned the fifth-best cost of energy.

Rank on one table tells you almost nothing about rank on the other. Look down the yield-rank column in the second table: 2nd, 12th, 6th, 7th, 11th, 3rd. There is no pattern, because output per watt and price per watt were being set by different things. Efficiency is an engineering figure. Price is a marketing decision.

That is the finding, and it is the reason this page exists. If you are choosing between two panels and one costs 60% more, the extra money is buying you something — warranty terms, brand longevity, a degradation curve, an aesthetic — but on this evidence it is not reliably buying you more electricity.

What has changed since 2021

This test is a snapshot, and the market has moved underneath it. Four things in particular:

• LG has left the industry. LG announced its exit from solar panel manufacturing in February 2022 and wound down production that year. The panel that topped this yield table cannot be bought new.
• Panel wattages have roughly doubled. This array ran 280W to 400W panels. A standard residential panel today sits well above that, which means the same roof area now carries considerably more capacity than it did in 2021.
• Cell technology has changed. Most of the panels in this test were PERC monocrystalline. The mainstream has since shifted to N-type cell designs with better temperature behaviour — which matters more in Queensland than almost anywhere else in Australia.
• Prices per watt have fallen. The dollar figures in the value table are 2021 figures and should be read as ratios between panels, not as anything you would pay now.

So: the specific models are history. The method is not, and neither is the conclusion. If we ran the same array today with current panels, we would expect the same shape of result — a modest spread in output per watt, and a much wider spread in what that output costs you. The array itself has not stopped: it has now been running for over seven years, and nothing we have measured in that time has changed our answer.

Why we install LONGi

Everything above is the evidence. This is the conclusion we drew from it, and it is the reason a LONGi panel is what you will find on our quotes.

Note what this argument is not. We are not claiming LONGi makes the most efficient panel in the world — our own data says it was not even the most efficient panel on our own roof. We are claiming it is the best combination of output, cost, longevity and support we have been able to find: tested against fifteen alternatives under identical conditions, and confirmed by seven years of installing nothing else.

What this means when you are comparing quotes

Three practical things follow from the data.

Do not choose a system on the panel brand alone. The spread between a good panel and a great panel is small. The spread between a well-designed system and a badly designed one is not. Orientation, pitch, shading, string layout, inverter sizing and how honestly the installer has modelled your consumption all move your return further than the name on the frame.

Ask what the premium is actually buying. Sometimes it is a genuinely better product warranty, a manufacturer with a real Australian presence, or a degradation guarantee worth having over twenty-five years. Those are legitimate reasons to pay more. “It’s more efficient” is not, unless someone can show you the number and it is large.

Ask why they use the brand they use. Plenty of installers quote a single panel brand, and there is nothing wrong with that — we do it ourselves. What matters is the reason behind it. If the answer is a distributor agreement, or “it’s what we always use”, that is a commercial decision being presented as a technical one. If the answer is measured data and a service record, you can check it. We built the test facility so that ours would be the second kind.

If you want to see how this feeds into real pricing, our Brisbane solar installation cost guide sets out current system pricing, and the Queensland solar rebates guide covers what comes off the top.

Frequently Asked Questions

The test facility is one part of how we work. For accreditation, quoting process and coverage, see solar installers in Brisbane.

The results on this page were recorded in January 2021 at Connect Energy’s solar test facility, 4/38 Limestone St, Darra QLD 4076. Panel pricing shown is as at that date and is used to compare panels against each other, not as a current price guide. Several tested models have since been discontinued. Figures are the array’s own recorded output and are not manufacturer or laboratory ratings. Connect Energy is a Solar Accreditation Australia (SAA) accredited installer based in Brisbane. For a site-specific recommendation, ask for a quote and we will explain what we would specify and why.

← All guides