Becoming self-sufficient,
from nothing Water · Solar · Wind · Casamance

At the start, there were the dunes, the sand, the sea and the birds. That is to say, nothing — except the essential: a pure, untouched, powerful nature. An idyllic setting. A nature so powerful that we gave up trying to tame it: it is impossible, and that is just as well.

It has to be said up front, because it explains everything else: in the beginning, there was no question of taking guests or building a business. That came much later. The first idea was simpler, and more brutal: to survive. To survive in a paradise. Because a paradise without water, without electricity, without the bare minimum of comfort quickly turns into hell. The sun itself, so generous, can kill you. So how do you become truly self-sufficient, from nothing?

The answer was not the one you would expect. Before energy, before light, before a single battery: water. Water first, because without it there is nothing — no life, no project, no lodge. Water, then something to eat. The rest comes after.

But there is one thing no high-season visit will teach you. Tourism in Casamance lives in the dry season: blue skies, calm sea, gentle evenings. You think you know this country. It is an illusion. You have to have been through a tropical rainy season to understand the truth of this place: in the dry season, nature leaves us in peace — then it unleashes. It strikes with lightning and kills. It floods, devastates houses, tears off roofs and uproots trees. You do not resist it: you work with it. You build around it, accepting in advance that there will be damage. To build here is not to defeat nature; it is to learn to bend without breaking.

And this is where the paradox that has guided every one of our decisions appears. To live under this sun, you need shade — so you plant. But here, a tiny shrub quickly becomes a tree over twenty metres tall: a filao grows at a speed you cannot imagine. Yet to produce your energy you need exactly the opposite: full sun on the panels, free wind for the turbine. Plant to survive, clear to produce — the same gesture and its opposite.

This guide tells that adventure: how, step by step and mistake after mistake, a survival project became a fully self-sufficient place — then a lodge. But it rests entirely on the lessons Casamance taught us from the very start.

It all began with a rope and a pulley. Not a pump, not a panel, not a motor. A rope, a pulley, a bucket — and arms. The well had given water, and it was good: it just had to be brought up. And at first, the only energy available was our own.

People tend to see this as a primitive stage, one you can't wait to leave behind. That is a mistake. Drawing water by hand was our best school. Before putting a single figure on paper, we learned the essentials in our shoulders: how many buckets a thirsty garden needs, how rare and precious water becomes when every litre is earned, what a hundred metres between the well and the furthest plantings really means, a watering can in each hand.

That is where we understood, physically, the questions we would have to solve. Because "watering the garden" is never a single action. It is a chain: you need to know what area you cover — for us, between 500 and 2000 m² depending on the season — how many watering points that requires, and therefore how much water leaves each day, especially in the hot season, when the sun drinks as much as the plants. And each answer called up another: if the garden consumes that much, you need a reserve; if you want pressure without electricity, that reserve must be up high; and if some plantings are more than a hundred metres from the well, the pressure must still reach them.

No catalogue whispered these questions to us. It was the rope and the pulley that posed them, one bucket at a time. The day we wanted to mechanise, we already knew exactly what we were trying to solve — and that is probably why we made fewer mistakes afterwards.

A detail first, which is not one. The first watering cans were ground to dust by the sun. The plastic whitened, cracked, then gave way. And that ordinary little failure opened our eyes to the true nature of the place.

The seafront is paradise. It is also a constant aggressor. Salt air, sea breeze, salt-laden humidity: all of it destroys, slowly, methodically, everything we put here. It is not the storm of a single night — it is a corrosion of every day, invisible, eating into metal, splitting plastic, swelling wood and rusting the smallest screw. Here, you do not build on solid ground: you live on a boat. And as on a boat, the choice of materials is not a matter of taste or budget. It is a matter of the installation's survival.

Very few things truly resist salt. In practice, you always come back to the same ones: stainless steel, aluminium, wood — provided it is treated, because to the salt you must add termites and all those little creatures that, too, eat what you build. Everything else is living on borrowed time. Every screw, every panel mount, every frame, every pipe must be chosen by asking not "does it work?", but "how long will it last against the salt?"

And for roofs, the answer did not come from modernity, but from tradition: straw. It breathes, it insulates from the heat, it shrugs off salt. But it has its rules. It is held down with a cotton fishing net — which, too, will turn to dust within a few years: another consumable, another cycle to anticipate. And the straw itself is renewed at a pace that depends on the slope of the roof: the steeper the slope, the better the water runs off, the longer the straw lasts. Here again, nothing is laid down once and for all. Everything is thought through, watched, replaced in its time.

Here is the first real mistake, the one everyone makes: wanting to fill the reserve first. It is instinctive. You dig a well, set up a tank, and think "water reserve" the way you would think "stocking up": volume, enough to last. Except water is not a stock you pile up. It is a network that flows. And a badly designed network, however full, does not hold.

The example is simple, and we lived it. A watering tap a hundred metres away starts to leak. Or, more ordinary still, the gardener forgets to close it at the end of the day. If the whole lodge is on a single circuit, you have only one option: cut the water everywhere — the kitchen, the bedroom showers, all of it — to deal with a tap problem at the bottom of the garden. One night, and the tower has drained for nothing. Unacceptable in a place that welcomes guests.

The real reserve does not begin with the tank. It begins with distribution: splitting the supply into independent zones, each isolatable by its own valve. With us, the major areas are separated — the kitchen, the bedroom bathrooms, the garden. And the garden itself, the biggest consumer and the most exposed to leaks, is subdivided into independent watering zones. A section that leaks, you isolate; the rest goes on living. It is the logic of a boat: watertight compartments, so that one leak does not sink the whole ship.

And there remains the question that decides real comfort: pressure. Having water is not enough; it must arrive with force — for a proper shower as much as for effective watering a hundred metres away. The starting pressure comes from the height of the water tower, free, by gravity. But a bad network wastes it: the diameter of the pipes is decisive. Too narrow, or reduced over long distances, and the pressure collapses before it reaches the end.

There is a trick that changes everything, and that you only learn by tinkering with your own network. A classic watering line is fed from one end only: the water enters at one end, runs the whole length, and loses pressure as it goes. At the end of the line, where the furthest plantings often are, only a trickle is left.

The solution is not necessarily a bigger pipe or more pressure at the source. It is to loop the circuit. Rather than feed a zone from one side, you connect its two ends, so the water arrives from both ends at once. Every point is then fed from both sides: the losses largely cancel out, and the pressure rises noticeably — especially where it used to collapse, at the end of the line.

And this is where our zone valves reveal their double use. The same valves that let you isolate a zone in case of a leak also let you, conversely, connect two together to loop the circuit and gain pressure when you need it. The same architecture serves both: compartment for safety, mesh for performance.

This is no stroke of genius — it is exactly what city water networks do, always looped so as to hold pressure everywhere and have no weak point at the far end. We merely brought that principle down to the scale of a garden. But it is precisely the kind of detail that no theoretical plan gives you, and that only practice — a watering can spluttering a hundred metres away — ends up teaching.

The reserve itself remained to be chosen. And there, common sense says "big". Reality says "adjustable". We began with 1000 litres. Too small, very quickly. We moved to 2000 litres. That is the first advantage of a simple tank over a masonry structure: you can change the size. Nothing is set in concrete — literally. For a project you discover as you build it, that is precious: you don't need to know everything in advance, you just need to be able to correct.

But the real stroke of genius was not the size. It was the colour. We used septic-tank vessels — good black plastic, sturdy and cheap. And black, under the Casamance sun, heats up. Placed up high, in full sun all day, the black tank preheats the water on its own: a few degrees gained, for free, with no installation whatsoever. Turning a sanitation tank into a passive water heater is exactly the spirit of this place: you look at what you have, and make it do more than intended. (Real hot water came later, with a dedicated solar water heater.)

And then there is the height: 12 metres. That is no random figure. At that height, gravity gives enough pressure to properly supply the bedroom bathrooms upstairs. Twelve metres of water column is a little over one bar of pressure at ground level: enough for a real shower on the first floor, and to push water to the far end of the garden. No pump runs for this. The pressure is stored in the height, available day and night, whether there is sun, wind or nothing at all.

A water tower needs upkeep: checking the level, cleaning the tank, inspecting the fittings. So at twelve metres, you have to be able to climb up — safely, and regularly. And there, the salt imposes its law once more.

It is a detail, in appearance. But it says everything: in this climate, nothing escapes the rule of materials, not even the ladder by which you reach the rest. The salt makes no exception; nor can we.

For energy, I did like everyone else. I searched online: water pumps, inverters, charge controllers, solar panels. You compare the watts, the prices, the efficiencies, and you build a first installation. Ours worked. It was afterwards that the real lessons began.

The first was a surprise: the most important part of an off-grid system is not the panel, it is the plant room. It is what houses everything — batteries, inverters, controllers, wiring — and all of it must be protected (from salt, humidity, dust, critters) and ventilated. Because all these devices give off heat, and heat is the enemy of batteries and electronics alike. We thought we were building an electrical installation; we were first building a room — secured, aired, designed so the equipment can breathe.

But the real blow, the biggest mistake of the whole project, was elsewhere. Invisible. Insidious. All these devices — inverters, controllers, pumps, electronics on standby — consume energy just to function. Not to produce anything: simply to be switched on, ready, waiting. This is the idle (no-load) consumption, and it is the black hole of off-grid living. Every box takes its little tithe, permanently. Multiply by the number of devices, by twenty-four hours, by the sunless nights… and you understand why the batteries drain while, to all appearances, nothing is running.

You always size an off-grid system by looking at what it produces. That is the mistake. You must size it by looking first at what it consumes doing nothing. The day I understood that, I stopped looking for more panels — and started hunting down every wasted watt. That is where, truly, our autonomy began.

Before the panels, before the batteries, before even hunting idle consumption, there is a question you cannot put off: at what voltage will the whole system run? 12, 24 or 48 volts. It looks like a detail. It is in fact the most committing decision of the entire installation.

Why? Because all the equipment is chosen around it. An inverter, a controller, certain pumps are designed for one precise voltage, and one only. And the day you would want to change your mind, you have to buy back part of the fleet. Unlike the water tank, which you enlarge for a few notes, voltage is the opposite: going back costs dearly. So you have to decide early, and decide right.

I went with 24 volts. For my needs at the time, it was amply enough. But there is a trap I had not anticipated. The bank voltage also governs the quantity of panels you can exploit. Since an MPPT controller is limited by its output current, and power is the product of voltage and current, the same controller takes twice as much solar power at 48 volts as at 24 volts. On my 24 V system, my MPPTs therefore capped at half the panel power they would have accepted at 48 V. As long as you stay modest, you don't feel it — but the day you want to grow, you hit that ceiling.

Back then, lithium did not yet exist — not for us, not here. The choice came down to two families: gel batteries, sealed, and open lead-acid batteries. And the choice imposed itself: sealed gel does not like heat, and heat is something Casamance has to spare. That left open lead. A constraint — the first of a long series where geography decided in my place.

Here, you don't pick your equipment from an endless catalogue: you take what exists within reach. It was Sukam, with lead-acid batteries of 12 V, 220 Ah. I bought four in Dakar to start, wired two by two in series for a 24 V bank. First blow on unboxing: of the four, two were already worn out. And impossible to return — the weight. Where the water tank could be swapped for next to nothing, the faulty battery, you keep. The lesson of remoteness, dry and final: here, a buying mistake is not corrected, it is endured.

And yet we held on for more than a year with that lopsided rig. The rest was of a piece: three 400 W panels (1200 W in full sun) and a 1200 VA Sukam inverter for the house. It was water that derailed it all. I had installed a surface pump — possible because, after the rains, the rice-paddy water table is only a few metres down. But that pump drew about 1200 watts, and at every start-up the inverter went into alarm and cut out. We ended up adding a small generator dedicated to the pump, just to be sure of having water. The irony, for a project meant to be 100% solar.

That is the great lesson of this first installation. Idle consumption is a permanent, silent leak; the start-up surge is a brief, brutal violence. Both share one thing: the catalogue does not announce them. An off-grid system is not calculated on what devices consume when all is well, but on their extremes: what they take doing nothing, and what they demand the instant they switch on. All of it shouted the same thing: we needed a thought-out architecture, not one assembled in a hurry.

To regulate the charging of the batteries, we needed a controller. And there again, geography dictated the choice before the technology. Solar gear is expensive in Senegal; right next door, in Gambia, a free-trade zone makes it noticeably more affordable. So you cross the border, and come back with your gear under your arm. Autonomy here often begins with geographic resourcefulness: knowing where to buy matters as much as what to buy.

I came back with a Morningstar TriStar TS-45. A good device, rugged — but let's be honest: a PWM controller, and not a true MPPT. A PWM simply connects the panels to the batteries by chopping the charge; it wastes part of what the panels could give. An MPPT goes looking for the panels' maximum power point and actually pulls more watts from the same sun. But the TS-45 was there, affordable, working: at that stage, you move forward with what you have.

It is more than thrift: it is a way of thinking. In a place where every part has been dearly acquired, transported, sometimes carried across a border, you don't abandon a device that works — you find it the post where it will still be useful. And it carries us, almost naturally, towards this place's other source of energy: the wind.

The generator for the pump did not last long — and good riddance. The noise, the upkeep, the petrol, the smell: all of it was the very negation of what we wanted to build. An ecolodge purring on diesel just to have water is a contradiction.

The solution was not to add power everywhere, but to separate the uses. Instead of a single inverter that had to absorb everything, I installed two, dedicated. One for the house electricity — lights, fridge, gentle loads. The other for the pump and the garden — a rough post, made of big start-ups. So when the pump surges and demands its peak, it no longer brings down the house lights. It is, on the electrical side, exactly what we had learned on the water side: you compartmentalise. The watering zones isolated leaks; the separate inverters isolate the surges.

And from that principle flows another, perhaps the most important when you live far from everything: redundancy. Here, a piece of equipment that fails is not replaced in an afternoon: you have to order, ship it in, sometimes cross a border again, and wait. The only answer is to plan for failure before it happens: a fallback, a spare, a degraded path to hold out while you repair. Better two modest devices than one perfect one.

I needed power for the pump. I thought I had found a bargain. Online there were inverters claiming 3000 W continuous and 6000 W peak, for a fraction of the price of specialist gear. I ordered. A fatal mistake.

The thing burned. Not tripped, not shut down safely: burned. The "6000 W peak" evaporates at a pump's first real surge, the components heat up, and one day it lets go. When you depend on that device for water and light, hours from any repair, it is not a saving: it is a bet lost in advance.

That failure is what led me to discover the real off-grid systems, built for low consumption and reliability: Studer, the Swiss, and Victron, the Dutch. The first reflex, faced with the prices, is to back away. For a long time I thought it was a luxury. It is the opposite: it is the cheapest option, provided you know how to count.

The reasoning is counter-intuitive. A cheap inverter is not only fragile: it is also greedy at idle. And that idle consumption has to be paid for — not in a bill, since you are self-sufficient, but in production equipment. Every watt wasted permanently is extra battery capacity to get through the night, and extra panels to recharge. And batteries and panels cost a great deal.

Through mistakes, the system eventually found its shape. Not the perfect shape — there is none — but a thought-out architecture, where every choice answers a lesson paid for in cash. Bit by bit, everything moved to Victron. First because, in the region, it is the serious brand that is easiest to find. Then, and this clinched my decision, for its remote monitoring.

Because an off-grid system has to be watched: knowing at any moment the battery charge, what the panels produce, what the house consumes. Victron's VRM system does exactly that, from a phone, wherever you are. It is no gadget: it is the difference between suffering a failure and seeing it coming.

The balancer deserves a word: in a lead bank, it is always the weakest battery that drags the others down, and an ignored imbalance means the early death of the whole bank. Balancing is protecting your most expensive investment. As for the Raspberry Pi, the contrast makes me smile: a computer worth a few tens of euros overseeing years of investment. Everything in this installation is redundant — because the rule of remoteness has never changed: here, what has no backup will, one day, leave you in the dark.

One last revolution remained to be tamed: lithium. LiFePO4 arrived — lighter, more enduring, able to take the deep discharges that kill lead. The temptation to replace everything at once was strong. I held back. Faithful to a method this place taught me — never bet big without testing for real — I first fitted out a small beach house as a proof of concept: a LiFePO4 Must LP6000 battery, with a Victron controller and MPPT.

The wind, I will say it frankly: it is complicated. Far more than solar. And I am not even certain, even today, that the benefit is always worth it. If I write it here, it is precisely because everyone sells dreams about wind, and the truth is more nuanced.

Solar is almost insolently simple: no moving part, it produces the moment there is light, it almost never breaks down. Wind is the opposite. A mechanical machine, which wears, vibrates, must be braked, secured, maintained — all of it perched high up, exposed to the salt and the rages of the rainy season. Where a panel merely exists, a turbine demands attention.

And yet there is one serious reason to take an interest, only one, but it is sound: the wind blows when the sun does not shine. At night, at dawn, on the grey days of the rainy season. And the idle consumption never sleeps. If wind merely manages to cover that nightly leak, to offset what the system consumes while solar rests, it will have done its job. I ask no more. That would already be a lot.

This last point is the problem peculiar to wind: what do you do with the energy when the batteries are full? A panel, you disconnect without risk. A turbine, no: if you cut it while it spins, it races and destroys itself. You must always give it something to do with its current — hence the diversion, useless but life-saving.

We have spoken of production, of storage, of big equipment. There remains the end of the chain, the part you actually see in the evening: the garden lighting. And it is there, in this seemingly trivial last link, that all the project's lessons come together.

Low voltage is also the right energy choice. To step the 24 V bank down to this 12 V lighting, you need a 24→12 V DC-DC converter — and Victron offers two families: isolated and non-isolated. An isolated converter electrically separates input from output, but consumes more at idle. The non-isolated one is more frugal. So, wherever galvanic isolation is not essential, I choose the non-isolated. A few milliwatts? Yes — but multiplied by every converter, every night, all year, it is again that silent leak that ends up costing one more battery and one more panel.

You can size a battery bank perfectly, choose the most frugal inverter, hunt down every idle watt — and watch the whole balance wobble because of a variable no controller commands: people. The day you welcome guests, the self-sufficient system becomes a tacit contract with those who live in it.

Two worlds meet. There are the guests who understand. To whom it is enough to explain once — "here, the electricity comes from the sun, it is precious" — for them to adapt their gestures. What we ask is minimal, almost symbolic: turn off the light when leaving the room. Nothing more. No timed shower, no comfort sacrificed. Many even leave charmed by this quiet sobriety, happy to have lived a few days to the rhythm of the sun.

And then there are the others, for whom a socket is just a socket, with infinite current behind it. They arrive with a 2500-watt hair dryer and plug it in without a thought. But 2500 watts at once, on a self-sufficient system, is a sledgehammer — the kind of power draw that vividly recalls the pump that used to trip the early inverter.

At heart, it is consistent with all the rest. You do not tame the sun, you work with it; you do not tame the rainy season, you adapt to it; and you do not tame people's habits either — you set a clear limit, explain the why, and trust. Autonomy here is not only an architecture of panels and batteries. It is also a small shared culture, kept for the length of a stay.

One thing must now be said, at the close of this account, and said frankly: total autonomy does not exist.

You can dig your well, raise your panels, balance your batteries, hunt every watt — and believe you are nearly there, soon "100% self-sufficient". It is an illusion. A beautiful illusion, of the kind we tell ourselves for reassurance in a world where resources are counted. You always depend on something: on the sun and the wind, which do as they please; on the rain; on an inverter made on the other side of the world; on a border to cross to find a part; on the sea, which slowly takes back all we leave it. Perfect autonomy does not exist. Well-managed interdependence does — and that is already a great deal.

Even the ecological balance, which we would like spotless, slips away when you look it in the face. I did the maths. To reach a carbon balance of −80%, counting also all the carbon it took to make, ship and haul this equipment — its grey debt — I would need more than fifteen years of operation. And fifteen years is precisely when everything will have to be replaced. On the very day I reached my balance, the debt starts again from zero. You never cross the line. Here too, the reassuring figure is a mirage.

Why this vertigo? Because there is a word for it, and it concerns us all. Predation is consuming without the slightest idea of what you consume. And man, by default, is a predator — not from malice, but from blindness. That is exactly what this guide has never stopped tracking, page after page: the cost you do not see. The idle consumption, the surge you don't anticipate, the carbon debt you forget to count, the label that lies about the true price of things. My whole craft as an off-grid builder will have consisted in making visible what we consume without seeing it.

There is even a paradox that makes me smile, and that points at me as much as anyone. Four of my best guests in five arrive because they typed "ecolodge" into Google. And four in five of them have not the faintest idea of what that word really covers. They are looking for an image, a reassuring label — like the "100%" or the "−80%" I have just dismantled. My own search ranking, what feeds me, rests on a word most people do not understand. I write it without bitterness: it is simply true.

And yet. If the figure is an illusion, the direction is not. Reducing one's needs remains right, even when the absolute zero is out of reach. The role of a place like this is not to display a score, nor to judge anyone. It is humbler: to make visible, for the length of a stay, what we consume without seeing. A light switched off on the way out. Water that rises by gravity alone. Electricity that sleeps when the sun goes down. Not to shame — to open eyes.

In the beginning, there was a well and some seeds. The wish, simply, to survive in a paradise too powerful to be tamed. I know today that I have never tamed anything — not the sun, nor the rainy season, nor the salt, nor the sea. I have only learned to live with almost nothing, attuned to a place greater than myself. This is not the failure of autonomy. It is perhaps its only honest definition.

For there is not only the concrete, the rational, the measurable. There is also all the rest — what no multimeter will ever read. We are in Casamance, a mystical land by essence, a patchwork of cultures and ethnicities where the invisible has its full and rightful place. At Le Papayer, I placed fetishes to protect the place from evil spirits and from jealousy. Not as folklore: without them, I would never have found anyone to work with me. The most rational installation holds only if you also respect what cannot be calculated. Working with nature, with matter, with people — and with their beliefs. It is, to the very end, the same gesture of humility.

There remains a handful of travellers — the rare among the rare — who truly understand, and leave moved. For them, what gratitude. They are the ones who prove the original idea was a fine one, and that none of this was in vain. Each brings what they can. Like that little bird of the legend who, faced with the wildfire ravaging the bush, shuttles tirelessly to throw onto the flames the few drops of water its beak can hold. The other animals watch, incredulous: "You're mad, you'll never put it out." And it answers: "I know. But I am doing my part."

A guest who turns off their lamp on the way out, a traveller who leaves with new eyes: it is little, and it is everything. My part, I do it.