John asked for a feasibility look at putting part of the farm's compound land to use for solar power, battery storage, and Bitcoin mining. This is that look — built from public market data (equipment prices, Bitcoin's current economics, NORECO's published rates, real climate data for these coordinates), not a site survey, a contractor's bid, or a utility quote. Treat it as a solid starting point for a real conversation with installers, a cooling engineer, DAR (the Department of Agrarian Reform), and DOE, not as a final plan.

The numbers throughout are shown at three sizes — 1, 2, and 5 acres. That is not three options to choose between. It's a way to see how cost, income, and payback scale as the system gets bigger, so the actual decision on size can be made later, once real land measurements and contractor quotes are in hand. 1 acre is the starting point John originally asked about, and it's what the rest of this report focuses on in detail — 2 and 5 acres are shown mainly to make the scaling visible.

A handful of assumptions run through everything below, spelled out in full in the note at the very end: figures are planning-level, not quotes; ₱61 = US$1; the mining hardware assumed throughout is the water-cooled Antminer S21 XP Hyd, not the cheaper air-cooled version (why is covered in "The Mining Numbers"); and Bitcoin's market — which every mining figure in this report depends on — moves fast and could look meaningfully different by the time any of this gets acted on.

Before the costs and the numbers, it's worth being clear about what this system physically is. None of it is exotic — it's a known combination of parts, sized and adapted for this particular piece of land.

Almost none of this is sitting on a shelf in Tanjay. Most of it has to be ordered, shipped, and cleared through customs before anything gets assembled.

This report assumes the 1-to-5-acre site sits on the retained compound land near the farmhouse and shops, not the leased cane fields — everything below is priced and sized on that basis.

NASA's 20-year climate record for these exact coordinates puts Tanjay at 5.08 peak sun hours a day on average — a low of 4.23 in January, a high of 6.25 in April. This report uses 5.0/day as its planning number, right in line with that.

At this scale — well short of a true utility solar farm — a fixed-tilt ground-mounted array fits about 150 kW-DC per acre once access paths, fencing, and the equipment yard are accounted for. After inverter losses (~20%), that's roughly 120 kW-AC per acre, and at 5.0 peak sun hours a day, about 600 kWh per acre per day — 219,000 kWh a year at the panels. Battery charge/discharge round-trip isn't perfectly efficient either — knock off another ~5% for that — which is why every table further down works from 208,050 kWh/acre/year as the real, usable number.

Philippine solar installed cost has come down fast — from around ₱80–90 per watt in 2024 to ₱55–75/W in 2026. This report uses ₱55,000–60,000 per kW-DC installed, with a small discount at larger scale. Two or three quotes from Cebu- or Dumaguete-based installers would tighten this number considerably.

What's actually in "site, electrical & cooling": the shipping container that houses the miners, the wiring and switchgear connecting panels to battery to miners, the electrician and technician labor to install and commission all of it, land prep and security fencing, and the water-cooling tie-in covered in the next section. For 1 acre, that's roughly ₱1.5M for the container and its internal wiring/racking, ₱0.9M for install and commissioning labor, ₱0.65M for site prep and fencing, and ₱0.55M for the cooling tie-in — call it ₱3.60M total, in line with the chart above. Still a planning-level split, not itemized contractor quotes.

Reference hardware: Bitmain's Antminer S21 XP Hyd — the liquid-cooled version, 473 TH/s at 12 J/TH, drawing 5,676W, roughly ₱378,000 (~US$6,199) per unit at August 2026 street prices, paid in USD since this hardware is bought internationally. It costs more per unit than the standard air-cooled S21 XP (270 TH/s, ₱305,000), but it delivers so much more hashrate per unit that it actually works out cheaper per TH — about ₱799/TH versus ₱1,130/TH air-cooled — on top of being the better fit for this climate. That's the reason it's the reference hardware here, not just a climate accommodation.

At today's Bitcoin market, this hardware converts electricity into about ₱6.99 of Bitcoin revenue per kWh fed into it.

Separate from day-to-day price swings, Bitcoin's own rules cut the reward miners earn per block in half roughly every four years — next scheduled for 2028, then 2032, then 2036. That's not speculation, it's built into the protocol. If Bitcoin's price never moved from where it sits today, that schedule alone would shrink this fleet's income by half at each halving, and eventually turn it unprofitable on hardware bought now. Historically, price has tended to rise enough around halvings to offset the cut — but "historically" isn't a promise, and it's worth planning for both outcomes rather than assuming the second one.

Everything above assumes the system just runs. It doesn't — someone competent needs to set it up, watch it, and keep it secure, and that's a real cost that belongs in this picture, not an afterthought. For context: an experienced outside specialist doing this properly — the kind of rate a North American security/IT consultant would charge for full-time setup, monitoring, and securing of a remote installation like this — runs on the order of US$200,000 a year (about ₱12.2 million). That's a useful upper bound, not what this actually needs to cost — a well-qualified local hire who knows the area and doesn't need to fly in could very likely do this for meaningfully less, and is probably the better fit for a farm-scale build anyway.

Not really. Higher-wattage panels and a tighter layout can push density to roughly 190 kW-DC per acre (up from 150), at the same ₱60,000/kW-DC installed cost used throughout this report — no exotic or premium-priced panels required, just a denser standard layout. That supports 6 miners instead of 5. Working through the same capex and revenue formulas used everywhere else in this report: solar ₱11.4M, battery ₱14.2M (507 kWh, scaled proportionally), miners ₱2.27M, cooling and wiring ₱2.49M, site prep ₱1.22M — ₱31.6M all in, against a net mining margin of about ₱1.60M/yr. Full-system payback: about 19.7 years versus 21. A real but modest improvement, not a game-changer — the extra solar and battery capex scales up almost exactly alongside the extra mining revenue, so packing more in changes the scale, not the payback ratio much. More battery doesn't help at all — the fleet already uses essentially the whole daily power budget, so there's nothing left to unlock.

What actually shortens the payback: Bitcoin's price recovering to a normal-cycle level cuts it to roughly 9.5 years on its own; claiming the RA 9513 tax break on the solar and battery portion (see "The Land") gets it to roughly 18.3 years; both together, roughly 8.4 years. Both are more realistic levers than trying to squeeze more hardware onto the same acre.

Everything above treats the array's output, net of storage losses, as if it all goes to mining. It doesn't have to. This same solar-and-battery system can run part of the farm's own operations — pumps, drying, housing, whatever's on the meter — and that's worth more per kWh than mining is: NORECO charges ₱9.54–12.22/kWh, well above the ~₱6.99/kWh mining earns at today's rate. Every kWh of real farm load this system covers is savings on top of the mining income already in the table above — it doesn't compete with the mining number, it adds to it.

This report can't put a peso figure on that saving yet, because it needs the farm's actual electricity bill history to size properly. But it's the single biggest lever for improving the whole-system payback shown above — potentially by a lot, since the ₱9.54–12.22/kWh farm-load rate beats mining's ₱6.99/kWh outright. Pull the last 12 months of NORECO bills and this section gets a real number instead of a placeholder — likely shortening that ~18–21 year full-system payback meaningfully, on top of the roughly 3.4-year payback the mining hardware already clears on its own.

A lot of the crypto-mining industry has spent the last couple of years switching from mining to renting out computing power for AI companies instead — in some markets, that trade has actually beaten mining outright. It's a fair question to ask before committing capital here, so worth addressing directly:

• Internet. AI computing needs extremely fast, dedicated fiber internet connections. Bitcoin mining needs almost none — barely more than a phone's data connection. Tanjay is well outside the fiber routes that run through Cebu and Manila; getting real high-speed internet out to a rural Negros Oriental site would mean a dedicated cable build most sites this size can't justify.
• Scale and customers. AI hosting deals are struck at a much bigger scale, with a specific big company renting the capacity. A 1–5 acre farm site has no path to landing that kind of deal, and no buyer even if the power were there. Bitcoin mining sells straight into a worldwide market with no sales process at all.
• The machines themselves. AI computers need much tighter climate control than Bitcoin miners and wear out faster. Bitcoin mining hardware is comparatively rugged and built for exactly this kind of remote, lightly-staffed site.

Worth another look in five or ten years if Negros Oriental gets real fiber infrastructure and someone lines up an actual hosting deal — but it's not something to plan around now. Bitcoin mining is the better fit for what this site has today.

Build the 1-acre version — but only after the paperwork, not before. Here's the order that avoids wasted money: