I am from Nigeria. This is comprehensive. Recently, I have been talking to someone building solar power system for households that can afford(but he doesn't provide opportunities like Sun King). I have also been doing research on providing the system for cold storage for farmers(I have ideas but no funding). It's great to read that SunCulture is already providing for irrigation purpose. Insightful article.
Sorry for being dense, but I don't understand the maths in Chapter 3. How does "Pay $100 down, then $45-60 per month" work when you previously stated "a massive, seemingly insurmountable barrier: $120 upfront might as well be $1 million when you earn $2/day."
I had the same issue. Take a zero off of the down payment, to $10, and then it's fairly feasible, even if still pretty significant (5 days' pay in a place where most folks have minimal disposable income). $0.21USD/day is $1.47/wk.
EDIT: I cross-checked pricing for SunKing products. It's hard to actually know the pricing, since it's all going through resellers, and varies pretty widely.
I saw one seller, in December 2024 via a Facebook post, advertising a roughly $10.40 USD down payment, and then $1.04/week payments for 30 **weeks** - not months, as stated in the article - for their baseline home lighting system. That's 4 tube lights plus a 141 Wh battery plus a low-voltage battery charging system (2x 1.5 and 4x 12V DC). $38.90 paid in a lump sum.
A current online marketplace listing has - is priced at roughly $73. Extrapolating the same pricing model as that above for a lump-sum price of $73, you'd have a down payment of about $19.50, and $1.95/week for 30 weeks.
Regardless, the baseline system seems to be cheaper than what's listed in this article, for a system focused specifically on lighting and basic appliances (they also sell bundles with fans and small HDTVs for those with more disposable income), which is verifiable via online resellers who don't offer the weekly pricing option, but it's hard to tell what the weekly pricing really looks like without transparent pricing from SunKing.
That said, if this is the cost to install a solar pump system and drip irrigation, $45-60/month isn't unreasonable, it's just a very different customer from the person making $2/day.
I looked up Sun Culture's pricing myself, and it's actually ~$20-40 USD/mo per pump for 30 months, with the cheapest pumps covering one acre with no battery, and the most expensive covering 2 acres plus having battery backup, plus a $27 to $54 down payment.
That's pretty dang affordable. Warranty is 2 years on the drip irrigation pipe and the battery, and 3 years on the pump. I do really appreciate that their battery systems also have DC charging built in for phones and similar devices.
The "$14,000 per acre" quoted here has to be for fairly high value crops; we're not talking about corn or soybeans, but instead vegetables, flowers, fruit, etc. - and potentially the price for crops grown for export to higher paying markets.
Some quick Googling of crop yield values per acre in Kenya shows that this isn't a totally crazy figure - tomatoes could yield gross income of $3,000-6,000/acre in Kenya, so long as you can get them to market (and the companies ColdHubs and SokoFresh mentioned above are addressing that cold chain gap).
SunCulture claims it can boost your yields up to 5x, and your income up to 10x - which tracks, if you can grow higher-value but more fragile crops that need irrigation, and aligns with going from $600/acre to $6,000/acre USD.
Avocados, a key SokoFresh product, can potentially hit the $14,000/acre revenue target once established, and they are a thirsty crop that benefits from drip irrigation.
I think this is, in general, a case of either having outdated pricing info, or allowing Grok or similar to do the math. It's not currency fluctiations; the Kenyan Shilling to USD exchange rate has been pretty stable. The Nigerian Naira has been less stable, but has largely trended towards being more valuable, rather than less, which means the USD equivalents would have gone up.
There's also the danger of draining aquifers through making pumping water cheaper. That's a real downside risk in terms of the price of success.
"Handheld solar lamps ($50-120)" -> I can buy a name-brand, high end solar lamp in the US for less than this. $5-12 would be the cheap knockoff price, delivered in the US. If that's the price to outfit an entire house and also provide a solar system to charge them with, then maybe it's accurate, but for basic solar lights, and some solar batteries you can use to charge your phone, this seems high.
EDIT: checked pricing, and again, this article was off by an order of magnitude. Lump sum pricing from Jumia, ("The Amazon.com of Africa") has their lowest-cost solar lamp at about $5.41 USD, for a 50 lumen max lamp (25 lumens when not in "Turbo" mode).
Goes up to $22.60-29.80 for their higher-output lamps, which also have a small battery and solar panel to charge a phone.
I wonder how much a factor the carbon credits were. I have mixed feelings about them since there have been a few cases where they backfired spectacularly, but if this wouldn't have been as successful without them, it shows they can be a massive success too.
There is a feature phone that gets close to what FemtoTX is developing. The Nokia 225 4G uses a 22nm single core processor made by UNISOC which resembles or carries over the 2005 Symbian phones' IP and were designed to use one low power processor for both the OS and the antenna modem when the EKA2 kernel was released:
https://en.wikipedia.org/wiki/List_of_UNISOC_systems_on_chips Their T107 uses a 22nm processor, which is low enough for solar power, but runs at 1GHz, which is fine for the modern feature phone apps, but too much power for an easily rechargeable (instantly<1 minute in cloudy weather) solar phone (I grew up having a solar calculator, so it set high expectations. :))
What they haven't done yet is combine a 22nm or even 12nm processor with a 104 MHz Nokia 7650 with 4MB RAM for more basic text and voice features:
This article explains, similar to your flywheel analogy, how the compounded energy savings from smaller transistors, coupled with simpler, smaller (and often older, but not necessarily) software and operating systems allows for making solar mobile devices possible:
Refurbishing old phones is practical and I do the same with laptops, but the phonemakers typically add more features at the expense of battery life and rely on shrinking transistor nodes (3nm, 2nm TSMC iPhone) to get the most performance per watt each new generation. Unisoc is a big enough chip maker to have access to relatively new nodes such as 6nm (e.g. see 5G phones: https://en.wikipedia.org/wiki/List_of_UNISOC_systems_on_chips#5G_smartphone_processors but those processors are all 4+ cores and use a lot of power.
There has even been a resurgence in flip phones among youth: https://www.wsj.com/tech/gen-z-flip-phones-might-be-onto-something-c4744796 so I can definitely see something happening, eventually. Solar powered electronics does require some of the leading or state of the art technology, but also benefits when fused with some older tech concepts and software.
- The urbanisation rate in sub-Saharan Africa is approximately 43.5% in 2024.
- Only about 57% of urban households are connected to the grid (and that doesn't imply 99,999% uptime)
So short answer: not generally, but often yes.
Some thoughts:
- In dense, well-functioning urban cores with reliable grids, centralized power usually wins on cost and energy density, especially for AC, electric cooking, and elevators.
- In many African cities, large shares of households and street economies are unconnected or face unreliable service. For these users, distributed solar + batteries can beat the grid on availability, speed to deploy, and total cost versus generators or informal connections.
- Urban Africa is not all high‑rises: large township/slum areas and mobile street markets are better matched to portable or rooftop solar than to slow, capital‑intensive grid extensions.
- As battery costs fall, solar can cover more urban loads, but 24/7 high-power appliances still favor grid or microgrids.
Before the advent of mobile phones, Nigeria struggled for decades and never managed to have up to 100,000 land lines installed across the country. Phones were the exclusive privilege of the rich and connected. Then mobile phones came and now there are more than a hundred million of them. It sounds like a good thing but the reality is the country has never learnt how to deliver any kind of public good with a last mile function. You have the same problem with broadband and of course now solar. There is no such thing as leapfrogging. I know this is a futile complaint as nothing is going to stop the forces in motion right now but I wish the country actually learnt how to build a grid and deliver public services to people. This is why today for the vast majority of the population, they have never once in their lives experienced water delivered into their homes through pipes, something people in many other countries can take for granted.
There is no alternative to figuring out how to do this stuff
I remember talking about the potential for this kind of development twenty years ago.
Fantastic that all the pieces of the stack have finally come together and market development and penetration of the poverty wall getting razed is finally happening and building momentum.
Great story! I’m looking at how to apply these concepts to underserved communities in the developed world, in my case Texas, USA.
I particular I’m interested in how the example companies may be using Blockchain (NOT cryptocurrency) to lower transaction costs and provide secure transactions.transactions
Great article, thank you for this! I one have question - the 0.21/day are over lifetime (I assume) - but how does it work from a cash-flow perspective for the respective customer if they only make a couple of USD/day. Is this loan based?
One of @skanders best posts today. Bravo!! 👏
I am from Nigeria. This is comprehensive. Recently, I have been talking to someone building solar power system for households that can afford(but he doesn't provide opportunities like Sun King). I have also been doing research on providing the system for cold storage for farmers(I have ideas but no funding). It's great to read that SunCulture is already providing for irrigation purpose. Insightful article.
Thanks for the insights Seun!
Cold storage is a super interesting (and needed) usecase, have you checked out ColdHubs (in Nigeria) and SokoFresh (Kenya)?
Thanks for the insight. I will check.
Sorry for being dense, but I don't understand the maths in Chapter 3. How does "Pay $100 down, then $45-60 per month" work when you previously stated "a massive, seemingly insurmountable barrier: $120 upfront might as well be $1 million when you earn $2/day."
I had the same issue. Take a zero off of the down payment, to $10, and then it's fairly feasible, even if still pretty significant (5 days' pay in a place where most folks have minimal disposable income). $0.21USD/day is $1.47/wk.
EDIT: I cross-checked pricing for SunKing products. It's hard to actually know the pricing, since it's all going through resellers, and varies pretty widely.
I saw one seller, in December 2024 via a Facebook post, advertising a roughly $10.40 USD down payment, and then $1.04/week payments for 30 **weeks** - not months, as stated in the article - for their baseline home lighting system. That's 4 tube lights plus a 141 Wh battery plus a low-voltage battery charging system (2x 1.5 and 4x 12V DC). $38.90 paid in a lump sum.
A current online marketplace listing has - is priced at roughly $73. Extrapolating the same pricing model as that above for a lump-sum price of $73, you'd have a down payment of about $19.50, and $1.95/week for 30 weeks.
Regardless, the baseline system seems to be cheaper than what's listed in this article, for a system focused specifically on lighting and basic appliances (they also sell bundles with fans and small HDTVs for those with more disposable income), which is verifiable via online resellers who don't offer the weekly pricing option, but it's hard to tell what the weekly pricing really looks like without transparent pricing from SunKing.
That said, if this is the cost to install a solar pump system and drip irrigation, $45-60/month isn't unreasonable, it's just a very different customer from the person making $2/day.
I looked up Sun Culture's pricing myself, and it's actually ~$20-40 USD/mo per pump for 30 months, with the cheapest pumps covering one acre with no battery, and the most expensive covering 2 acres plus having battery backup, plus a $27 to $54 down payment.
That's pretty dang affordable. Warranty is 2 years on the drip irrigation pipe and the battery, and 3 years on the pump. I do really appreciate that their battery systems also have DC charging built in for phones and similar devices.
The "$14,000 per acre" quoted here has to be for fairly high value crops; we're not talking about corn or soybeans, but instead vegetables, flowers, fruit, etc. - and potentially the price for crops grown for export to higher paying markets.
Some quick Googling of crop yield values per acre in Kenya shows that this isn't a totally crazy figure - tomatoes could yield gross income of $3,000-6,000/acre in Kenya, so long as you can get them to market (and the companies ColdHubs and SokoFresh mentioned above are addressing that cold chain gap).
SunCulture claims it can boost your yields up to 5x, and your income up to 10x - which tracks, if you can grow higher-value but more fragile crops that need irrigation, and aligns with going from $600/acre to $6,000/acre USD.
Avocados, a key SokoFresh product, can potentially hit the $14,000/acre revenue target once established, and they are a thirsty crop that benefits from drip irrigation.
I think this is, in general, a case of either having outdated pricing info, or allowing Grok or similar to do the math. It's not currency fluctiations; the Kenyan Shilling to USD exchange rate has been pretty stable. The Nigerian Naira has been less stable, but has largely trended towards being more valuable, rather than less, which means the USD equivalents would have gone up.
There's also the danger of draining aquifers through making pumping water cheaper. That's a real downside risk in terms of the price of success.
There are some other wonky numbers as well.
"Handheld solar lamps ($50-120)" -> I can buy a name-brand, high end solar lamp in the US for less than this. $5-12 would be the cheap knockoff price, delivered in the US. If that's the price to outfit an entire house and also provide a solar system to charge them with, then maybe it's accurate, but for basic solar lights, and some solar batteries you can use to charge your phone, this seems high.
EDIT: checked pricing, and again, this article was off by an order of magnitude. Lump sum pricing from Jumia, ("The Amazon.com of Africa") has their lowest-cost solar lamp at about $5.41 USD, for a 50 lumen max lamp (25 lumens when not in "Turbo" mode).
Goes up to $22.60-29.80 for their higher-output lamps, which also have a small battery and solar panel to charge a phone.
That stopped me cold as well. Any replies?
I had the same doubt.
This is clearly mostly written by an LLM. The style is dead on. Not surprising the math is off.
I wonder how much a factor the carbon credits were. I have mixed feelings about them since there have been a few cases where they backfired spectacularly, but if this wouldn't have been as successful without them, it shows they can be a massive success too.
Mixed feelings here too (see https://www.climatedrift.com/p/carbon-offsets-explained-part-3), although they can be a nice additional revenue source (and here it feeds the flywheel nicely)
Not to focus in on the wrong subject but anyway:
"That’s not a charity. That’s a fucking rocketship.
Okay, this is where it gets really spicy."
Did you write this with grok?
He did retweet saying vibewrote this week.
So yes
Amazing stuff
Thanks Kevin!
Agreed @Marco Morawec !
Thank you Eleanor!
Fascinating insights
Thanks Bethan!
awesome post. I wrote a lot about this in 2022. Working on developing a solar powerable cell phone & laptop: https://ei2030.github.io/FemtoTX/#about
Super interesting.
Sunking now also goes strongly into selling and refurbishing phones.
There is a feature phone that gets close to what FemtoTX is developing. The Nokia 225 4G uses a 22nm single core processor made by UNISOC which resembles or carries over the 2005 Symbian phones' IP and were designed to use one low power processor for both the OS and the antenna modem when the EKA2 kernel was released:
https://www.theregister.com/2005/12/30/singlechip_rtos_phones/
https://www.gsmarena.com/nokia_225_4g_(2024)-12968.php
https://en.wikipedia.org/wiki/List_of_UNISOC_systems_on_chips Their T107 uses a 22nm processor, which is low enough for solar power, but runs at 1GHz, which is fine for the modern feature phone apps, but too much power for an easily rechargeable (instantly<1 minute in cloudy weather) solar phone (I grew up having a solar calculator, so it set high expectations. :))
What they haven't done yet is combine a 22nm or even 12nm processor with a 104 MHz Nokia 7650 with 4MB RAM for more basic text and voice features:
https://www.gsmarena.com/flashback_nokias_first_camera_phone_was_also_the_first_symbian_s60_smartphone-news-45239.php
This article explains, similar to your flywheel analogy, how the compounded energy savings from smaller transistors, coupled with simpler, smaller (and often older, but not necessarily) software and operating systems allows for making solar mobile devices possible:
https://semiengineering.com/the-next-big-thing/
https://bit-tech.net/reviews/tech/cpus/intel-solar-powered-cpu
Refurbishing old phones is practical and I do the same with laptops, but the phonemakers typically add more features at the expense of battery life and rely on shrinking transistor nodes (3nm, 2nm TSMC iPhone) to get the most performance per watt each new generation. Unisoc is a big enough chip maker to have access to relatively new nodes such as 6nm (e.g. see 5G phones: https://en.wikipedia.org/wiki/List_of_UNISOC_systems_on_chips#5G_smartphone_processors but those processors are all 4+ cores and use a lot of power.
There has even been a resurgence in flip phones among youth: https://www.wsj.com/tech/gen-z-flip-phones-might-be-onto-something-c4744796 so I can definitely see something happening, eventually. Solar powered electronics does require some of the leading or state of the art technology, but also benefits when fused with some older tech concepts and software.
Does solar beat grid in urban areas? Because urbanising has benefits beyond electricity access, and Africa is urbanising faster than anyone.
Hi Drea! Good question!
Did quick research on it
- The urbanisation rate in sub-Saharan Africa is approximately 43.5% in 2024.
- Only about 57% of urban households are connected to the grid (and that doesn't imply 99,999% uptime)
So short answer: not generally, but often yes.
Some thoughts:
- In dense, well-functioning urban cores with reliable grids, centralized power usually wins on cost and energy density, especially for AC, electric cooking, and elevators.
- In many African cities, large shares of households and street economies are unconnected or face unreliable service. For these users, distributed solar + batteries can beat the grid on availability, speed to deploy, and total cost versus generators or informal connections.
- Urban Africa is not all high‑rises: large township/slum areas and mobile street markets are better matched to portable or rooftop solar than to slow, capital‑intensive grid extensions.
- As battery costs fall, solar can cover more urban loads, but 24/7 high-power appliances still favor grid or microgrids.
Before the advent of mobile phones, Nigeria struggled for decades and never managed to have up to 100,000 land lines installed across the country. Phones were the exclusive privilege of the rich and connected. Then mobile phones came and now there are more than a hundred million of them. It sounds like a good thing but the reality is the country has never learnt how to deliver any kind of public good with a last mile function. You have the same problem with broadband and of course now solar. There is no such thing as leapfrogging. I know this is a futile complaint as nothing is going to stop the forces in motion right now but I wish the country actually learnt how to build a grid and deliver public services to people. This is why today for the vast majority of the population, they have never once in their lives experienced water delivered into their homes through pipes, something people in many other countries can take for granted.
There is no alternative to figuring out how to do this stuff
Very interesting, however I feel like it was written with Claude or another LLM, at least partially
I remember talking about the potential for this kind of development twenty years ago.
Fantastic that all the pieces of the stack have finally come together and market development and penetration of the poverty wall getting razed is finally happening and building momentum.
Thanks for bringing the hope and godspeed.
Great story! I’m looking at how to apply these concepts to underserved communities in the developed world, in my case Texas, USA.
I particular I’m interested in how the example companies may be using Blockchain (NOT cryptocurrency) to lower transaction costs and provide secure transactions.transactions
Please stop writing articles with LLMs
Great article, thank you for this! I one have question - the 0.21/day are over lifetime (I assume) - but how does it work from a cash-flow perspective for the respective customer if they only make a couple of USD/day. Is this loan based?
I didn't read it as suggesting that. It is repayment of a loan, I think.
Once paid, cash flow improved.