TL&DR Summary: yes, it really is that cheap. It’s not bullshit. It’s real. And it’s going to power the world. Well, not the whole world- but a very significant part of the world.
Those of you who read my articles will know that I have a small farm east of Toronto. We’ve built a couple small cabins there, and repaired a dilapidated building to become my workshop. There’s no well, no septic system, and no grid electrical supply. No, we don’t live there full time- but it does serve as a useful test bed for an off-grid system regardless.
Our power solution is pretty simple: we have solar, and in the few winter weeks where solar is interrupted by snow cover on the panels, we have a gasoline generator. That generator runs more often just to make sure that it works when we need it, than it ever does to make power.
Each cabin has a single panel, a small charge controller, a 300 W inverter (good enough for lights, a fan in summer, and cellphone charging), and a battery- generally an 80 Ah marine lead-acid battery, because they are best suited to sitting full most of the time. The breakdown of the cost of these systems is as follows:
(all costs in this article are in $CDN, not including taxes. As of publishing, the $CDN is about $0.75 USD)
12V 100W panel: $100 (Amazon) – you can get a more powerful 30V panel but then you need a better charge controller
PWM charge controller: $10 (AliExpress) Want a MPPT unit that can handle a higher voltage panel? Then you’re looking at $100
300 W 120 V inverter: $30 (AliExpress)
Wire and connectors: $50
80 Ah lead-acid marine battery: $200 (Canadian Tire)
Why bother with these little systems at all? Because the cabins are a long way from one another, and stringing wire through the woods is a pain in the butt, and ruins the aesthetic. Trenching is even more of a pain, and would sever the roots of numerous trees if we did. And a single panel and battery produces all the power we really need, with no ongoing operating costs other than some distilled water every once in a while to water the batteries.
You’ll note that the biggest cost in each of these systems is the battery. While “lead is dead” for real solar applications that cycle frequently or deeply, lead-acid’s one strong point is applications needing to sit at 100% state of charge for long periods of time, where discharge is at a low drain rate (low currents, draining the battery over a period of 20 hours or more). The capacity of lead acid batteries at high drain rates is very low due to the Peukert effect. Lead-acid batteries love to sit full, and need to occasionally be agitated to prevent sulphation- that’s why they last so well as vehicle starter batteries. In stationary applications, agitation is normally done by means of a “balance” or “equalization” charge cycle, where overvoltage is applied and the cell electrolyte is agitated with the resulting hydrogen.
If we ever need more power- to run power tools at one of the cabins for instance- we have a brick of four LFP prismatic cells with a 3000 W inverter mounted on top of it. This is our “silent generator”, and I use it for all my mobile construction projects. We recharge it using a DC/DC converter ($6, AliExpress) from the main pack in the workshop.
The workshop system is more complex, and more fully functional as a source of electricity. It consists as follows:
Four nominal 300 watt panels: I recently bought two 320 W Jinko panels for $135 each. These are brand new, just the older model. $0.42 CDN per watt of capacity…that’s insanely cheap. Total $540 for 1.2 kW of nameplate capacity.
Mounting structure: made of 2x4s sawn from fallen trees in our own forest. Previously, the other two panels were mounted on two ~ 4” diameter black locust trees that I had to fell because they were in the wrong place, which I just limbed, notched, and leaned up against the south side of the building (back during COVID when 2x4x8s were $10 each). It looked ugly and worked just fine, but now I have a nicer support system which allows me to tilt the panels to different angles in summer and winter. The structure cost me perhaps $10 worth of hardware and $10 worth of sawmill and chainsaw gasoline and tractor diesel and sawblade wear and tear combined, but if you bought the wood at Home Depot it might have cost $125 or so. The exercise I received is not charged for, but saves a gym membership.
Charge controller: EPEVER 40 A maximum power point tracking unit with buck stage, $200, Amazon. These things are pretty brainless and still don’t have proper factory lithium ion settings, but you can make them work especially with a BMS.
Batteries: eight 280 Ah LFP prismatic cells: $860 including delivery from China. These cells are guaranteed for 6000 cycles, and are about 90% efficient, i.e. they return 9 out of every 10 kWh you feed them. They come with the cell connectors to build a series pack. I built a box for them out of leftover plywood. The eight cells are arranged in series with a total nominal capacity of about 7 kWh.
Battery Management System: depending on what you’re doing, these can cost between $20 and $200 for a suitable unit. The 100 A 24V unit I am using at the farm cost $60 (AliExpress). The BMS monitors individual cell voltages. If any cell goes above 3.65 V, it shuts down charging. If any cell goes below 2.5 V, it shuts down the inverter output. The BMS also provides a shunt charging balance function to top-balance the cells, but with LFP such continual balancing is basically unnecessary.
Battery Status Gauge (“smart shunt”)- $60, Amazon (not strictly required, but helpful to know your state of charge, which with LFP cells you simply cannot tell via voltage readings except below 10% and over 90%). This device measures and displays currents into and out of the cells and keeps track of the battery state of charge.
Inverter: 3000 W peak (1500 W continuous) true sine wave unit, 24V nominal input $108 (AliExpress)
Miscellaneous: wire, connectors, disconnects, terminals etc.: perhaps $200 all in
Ignoring my labour (and why shouldn’t I- this was all good fun!) that’s about $2000 for the whole shebang. That’s not just solar, but also includes storage. More than enough storage for our needs- but we’ll soon have even more, as I augment the existing pack with batteries taken out of my electric car conversion projects after 10 years of service. At the lower C rate of a solar application, these LFP cells will likely last ANOTHER ten years.
Some of you will say, “Oh, isn’t it nice to be an engineer, to know how to do all this stuff?!” Well, I’ve got one word for you: YouTube. Visit one of the numerous sites such as Off Grid Garage, and you’ll soon know what’s involved. It ain’t rocket science, or brain surgery, much less rocket surgery! As long as you stick with either lead-acid or LFP with a BMS, and stay at 48V or less, you’re going to be just fine. And yes, you can pay more for simpler solutions- LFP batteries packaged with their own internal BMS for instance so you don’t have to trouble yourself over them any more than you would with a lead-acid battery. You can always push the “easy button”- it just costs you a bit more. Of course you could push the “more money than brains” button and buy a Tesla Powerwall, but I definitely wouldn’t recommend that! That product is massively over-priced. Think of it as the “Apple” of battery storage systems.
How much electricity can my workshop system produce in a year? Well, the panels will generate about 1,300 kWh per year per kW of installed capacity, so about 1,560 kWh per year. And yes, they generate a surprisingly large number of kWh in winter, despite the snow- tilted at the correct angle, they tend to shed snow surprisingly well. But of course I won’t use all of those kWh, so it’s hard to make a proper cost comparison. In summer, most of the kWh go to keeping a fridge and freezer cold while we’re not there- but it’s nice to have cold drinks and to not have to go into town every time you need something. In winter, the system more or less does nothing when we’re not there.
The important thing is this: the system is sized for our winter needs. In summer, we make excess kWh but they cost us nothing. When the batteries are full, the charge controller or the BMS simply disconnects the panels. The cost of curtailment is zero.
But when we’re there, that system does everything we might expect to do with a power line connected to the grid. If we lived there full time, we’d likely go up to 10 kW of panels (because they are VERY cheap) and quadruple our battery storage too, going with higher output hybrid inverters etc. etc., but that would be more about convenience, reliability, and recharging our electric car, than about anything else. That system would definitely be up to the job of running a heat pump for cooling and shoulder-season heating. We’d still heat the place primarily with wood, because with 28 acres of woodlot, we could never burn all the deadfall.
Costs per kWh and Other Metrics
You could perhaps add up 16.5 years worth of the panels’ production (daily cycling of the batteries to achieve the guaranteed 6000 cycles takes 16.5 years- the panels etc. should all last much longer than that) and do a ratio, i.e. $2000/(1,560 x 16.5) kWh = $0.08/kWh, but that would assume that I use every kWh I make.
Remember, that includes storage. That’s for fully functional solar power.
But a better comparison is this: what else can you buy for $2000?
You can buy the installation of one (1) power pole. And I’d need fourteen (14) poles installed to bring power to my site from the nearest utility pole. That doesn’t include wire, or labour to install it.
You can buy 4.5 years worth of monthly account charges to have an account with the local electrical utility, whether you use any kWh or not…
Are you agreeing with me yet?
Solar power is just that cheap. Solar plus batteries are of course even cheaper in Australia where capacity factor is such that each kW of panels generates 2.7 times as many kWh per year, and where snow cover on the panels is never a worry.
And panel and battery prices are still falling- as are the prices of all the various bits of kit needed to build an off-grid or “behind the meter” system. Sodium ion is just coming on stream now, and has the twin advantages of low materials cost (nothing in those batteries is anything other than earth abundant), and the ability to be shipped fully discharged, i.e. no hazardous goods shipping charges, which are a significant fraction of the cost of buying LFP cells today.
I can clearly see where this is going. I’ve seen it for years now. Can you?
Yes, I know that my own little system isn’t a pin for pin replacement for the grid connection you have to your house, condo, flat, or castle. I know you “can’t” and really wouldn’t build a whole grid like this. I know there are reliability needs that literally mean life or death, that must be satisfied. I know every human in a developed country is now conditioned to think that power can be supplied in any quantity with no more thought than it takes to flip a switch. I know that building a fully decarbonized energy system is going to take decades, and cost trillions, and require major changes to how consume energy and hence how they live. But if you don’t see that solar is going to play a major role in energy supply in a decarbonized future, I’m sorry- you either haven’t thought about it carefully enough or you’re delusional. Energy this cheap is just too good an opportunity to pass up.
Disclaimer: this article was written by a human, about his own little experiments with solar power. If I’ve made a mistake, as humans are wont to do from time to time, correct me and I’ll be grateful.
If however you don’t like what you’ve read because I’ve threatened your own pet idea, then you can contact Spitfire Research Inc. who will be more than happy to tell you to piss off and write your own article.