question

Hi Gary,

Nominal Voltage: 12-volt
Load: 3-amps constant

That is a fairly decent load, 72ah a day.

Battery Amp-Hours: ?

You are looking at drawing 50ah out of it with long winter nights. What battery chemistry? Maybe a 200ah agm, or 100ah lithium (minimum size with no days of autonomy)

Solar Watts: ?

Assuming a mppt charger and Sydney weather. According to the Bom, Sydney gets an average of 8.6MJ/msq in June. Divide by 3.6 to get 2.38kWh/msq.

If we pick a 250w panel, with an area of 1.6msq, and a module efficiency of 15%, you would harvest 47ah a day in June (average).

Hi Gary. Please pardon me for asking, but as a Victron Service Partner, are you 'market researching' the great-unwashed-end-users? Not being critical, just wondering?..

"how would one determine the optimum solar configuration that would "reasonably" guarantee power 24/7 all year round?"

Defining 'reasonably' is the big issue, and is the thing so hard to nail down, given the vagaries of weather. Related of course to 'days of autonomy', a term which I dislike generally, but sorta indicative of the need to find something which can cope to some degree with multiple poor solar days in sequence.

I don't supply nor install, but often asked to recommend (another freeby arranged for this weekend). Ya just know, but I quite like the Victron calculator to check and explain to those in need.. https://www.victronenergy.com/markets/off-grid

Fooling with the sliders there is fun to show how this stuff, and sizing it, fits together. I've never managed to master selecting the location there though (blame my browser?).

Not a direct answer, just pondering the wider picture. When you say optimum, it makes me think about how is optimality defined in your problem, and what the constraints are.

Lets say that your potential customer is an informed one, and asked you to provide the system that you asked about (12V, 3A constant draw) at the lowest possible investment (including replacement cost for panels, fuel to run genset, etc) over X years. And as John suggested in his answer, you are also given some metric like days of autonomy or maximum acceptable probability of outage.

Then you have enough information to construct a whole of life cost model (with ability to insert secondary power sources, change battery/panel ratio, solar conditions, etc).

e.g. W dollars for panel, replaced Y times during X years => $Z investment

Thing can be as complex as you want, and is entirely dependent on validity of numbers.

And then you answer the optimality question by finding the lowest cost solution (within some limits on time expended). And if the customer doesn't give you specific constraints needed for the model, you find the most likely based on market research and offer that up as part of the proposal.

I can only imagine that such a model exists, although likely to be proprietary!

edit: there's probably some interesting trade offs in there in terms of lowest possible investment over X years may mean a higher upfront cost that a customer may baulk at...

Given the below power requirement, how would one determine the optimum solar configuration that would "reasonably" guarantee power 24/7 all year round?

Given the given location of Sydney, one can conclude that;

You can expect consecative days where the solar irradiation will average about 1kWh/msq. This would determine the sizing of the solar array to ensure a reasonable yield to power the device and charge the battery.

You can also expect random days where the solar irradiation is below 0.5kWh/msq. This is not enough to wake up the mppt controller, thus you could not expect any useful yield on such a day. You would have to dimension the battery so that it would have enough storage for a night, an unproductive day, and a night.

There is a point where it is no longer viable to over dimension an array, because double of nothing, is still nothing.

@Gary Pacey It may be worth looking for someone with a reference site to get a realistic real world view as well. I run 8 x 300W mono panels with a 77Ah Lithium totally off grid. near Cape Agulhas, SA. A few things. Choice of tech. Mono cells are sensitive and by 10am the system can be charged up ... that is long before the panels even see direct sun. There has not been a single day throughout the year the battery did not fully charge. You may be just be surprised what you can get on an overcast day. Secondly, Lithium charges very fast so as in my case, unless there are other loads, all that energy remains unharvested after 10am. Were there to be loads while charging like our washing machine or dishwasher, the MPPT will up the power output to supply it first. I now leave the fridge on when the house is unoccupied to keep the battery "exercised" So, while all the calculations are vital, I'd suggest not to overlook the mentioned details and also that the system is a consumer too.