HackiNews

• turtlebits
  The nice thing about solar is that you can start small and add incrementally.

  I dont know what the $4500 includes, but you can fairly easily get 1kw of solar and 5kWh battery and inverter for 1k USD.

  • thelastgallon
    The cost of solar is 1X - 20X depending on location. US has the highest costs, followed by maybe Western Europe.

    The panels are dirt-cheap[1] straight from China, but your friendly govt adds tariffs.

    The balance of the system is super costly in some locations, ridiculously cheap in other locations.

    In Pakistan, they are laying out the panels on the roof without any structure.

    In most of the third world, homes are concrete built, they already have structural elements and it isn't a big deal to add panels. It is a massive project in US, because of shingled roof. Labor is also available, available immediately, and permitting is either easy or not necessary.

    I'd assume the cheapest total install cost would be (cheapest to costliest): China, Pakistan, Bangladesh, middle-east,, Africa, South East Asia, Central America, Mexico, India, South America, Australia, Eastern Europe, Western Europe, Canada, and finally US.

    I'm ignoring the cost of land which varies a lot in each country. We also don't need land, just add vertical solar on every structure, fence and wall.

    [1]or as cheap as plywood which a HN comment pointed out

• thelastgallon
  For off grid residential (which will be ~1 billion people), I wonder if it can be a lot cheaper to go all DC, avoid inverters, losses converting DC<->AC<->DC.

  Looks like it is possible: https://solarpoweredge.com/2026/05/10/pure-dc-off-grid-solar...

  • mindslight
    Yes, no conversion losses. No quiescent draw from needing to create a standing AC waveform. No need to size an inverter for maximum load. My gut says you'd still want some kind of MPPT/charge controller, but the article you linked didn't use one. I'd be tempted to go two battery banks for dual-ended +/- 50 volts (same trick as split phase), but 100V DC is starting to get up there so maybe not. (And if you did you'd probably want some converter to be able to balance loads between rails).

    My main concern would be creating such a bespoke system that you couldn't find anyone else to work on it. But quite irrelevant when we're talking about a whole country going that direction.

    • BizarroLand
      The real issues with a full DC system is finding compatible electronics to meet your wants.

      Want to play a video game? Gotta either convert your DC to AC so the AC to DC converter can swap it to the right voltage. Either that or you need a DC DC buck/boost converter that you can trust to not trip out and fry your system.

      The systems exist for local full DC power, but they are likely more expensive, designed for vehicle/RV use (and thus smaller than the typical home use item of the same variety) and more difficult to repair because they are made smaller.

      Not saying you can't do that, only that there are plenty of tradeoffs to take into consideration due to the differences in scale of devices built for AC vs devices built for DC.

      • mindslight
        Fully agree, but I would think this a bit less of a concern with a less developed market, due to just having fewer devices. Network effects aren't going to be as strong.

        Like I'm at the point I have no idea how many USB chargers I have. They're everywhere. If I switched to 48V, I would become immediately aware of how many I had, and what I used them for (0, until I bought/made some).

        Plus in general once you convert (or don't have) high power draw devices (eg do electric clothes dryers or automated dishwashers even exist in this context?), you can get by with a much much smaller inverter to use the handful of low-draw things you have left and just pay the efficiency tax when you're actually using one. You probably don't even need a true sine wave to drive most of the rando AC-DC adapters we're talking about.

        (Also smaller dwellings mean you could also probably get by with one good quality trusted 12V/5V buck and catch 90% of devices, without having to run more sets of thicker cables to distribute those voltages)

• toomuchtodo
  https://archive.today/ivm8t

  Assuming ~50M households in Nigeria, and the current cost of ~$4500 to equip a home with solar, the math is still daunting.

  - The equipment needed to keep a three-person home running will pay for itself in as little as 16 months, BloombergNEF research shows — but, at $4,500, getting set up costs about seven times as much as buying a new petrol generator. (It’s the generator’s fueling costs that will eventually bite, at roughly $200 a month.)

  Related:

  Chinese EVs Make Inroads in Nigeria as Gasoline Prices Rise - https://www.bloomberg.com/news/features/2025-02-26/lagos-ev-... | https://archive.today/lMAtD - February 25th, 2025

  (if you're looking for a startup idea for YC, consider a way to finance these systems for consumers; go to the bond market for capital, manage credit risk and accepting payments; your TAM is all of Africa's energy needs)

  • toomuchtodo
    Related:

    Ember Energy: The take-off in African solar that official statistics can’t yet see - https://ember-energy.org/latest-insights/the-take-off-in-afr... - August 26th, 2026

    - Africa installs a record 17 GW of solar in 2026 – three-quarters of it is distributed solar, which is largely invisible

    - 100k solar panels installed across Africa every day during 2026

    - 19 countries have seen year-on-year growth of over 100%, including 544% in the Democratic Republic of the Congo (DRC), 282% in Zimbabwe, 176% in Egypt and 117% in Zambia. Solar growth used to be dominated by South Africa, but its share of African installations in 2026 will be below 20% for the first time since 2019, as other countries surge. Solar is now the single biggest addition to grid capacity in many African countries – Senegal will add solar capacity from 2023 to 2026 equivalent to almost 80% of its entire 2023 grid capacity, with DRC and Kenya adding more than half.

  • metalman
    I am very happy to see Nigeria establishing such a sensible tradition. solar and batteries will outlast a generator, and have significantly higher uptime, also quality of life around a generator is reduced, along withtime to start/stop and get the fuel, which requires fuel to get, so any reasonable comparrison would need to get hard numbers on all of the costs. I live off grid, and can proffer this, that solar is slightly more boreing than watching paint dry, as the paint does change it's coulor ever so slightly. Sodium batteries are going into giga scale production now and will offer significantly better economics ,for small business and farms solar is the thing that will smooth out economic shocks and give the versatility to adapt
  • silverFork
    You need to buy fuel to use a generator. No fuel-> No electricity. The sun though rises everyday.

    • toomuchtodo
      Solar panels also last at least 25 years, batteries anywhere between 10-20 depending on chemistry and cycling. Stocks vs flows.

      (tangentially, China's solar PV manufacturing capacity is double current global demand, so capacity to produce is not the issue, but getting the clean tech into the hands of those who need it)

      • thelastgallon
        Yes, 92% production warranty at the end 25 years. Which is the longest product warranty of anything sold. At 92%, they could last a few more decades.

        Newer panels are offering 40 year warranty.

        40-year warranty covers the panel and its performance. And in the unlikely case you’ll ever need it, we make the return process simple and easy. With a historical return rate of only 1 in 20,000 panels under warranty: https://www.maxeon.com/maxeon-solar-panels

  • telebub
    Great application of cross continental research by Blomberg. A typical Nigerian family of 3 (already atypical) is probably not spending one parents entire salary on electricity.
  • ejanus
    Any link to BloombergNEF research, not pay wall?