03 · An Réiteach — The policy option
Let flexible demand use the surplus
Why it matters, whatever your priority
Bills
Every megawatt-hour used instead of turned away is one that may not need compensating — and those payments reach every bill.
Climate
Wind farms that can sell power they now lose are easier to finance, which helps the build-out the 2030 targets depend on.
Public money
None of the options needs a subsidy, levy or guarantee. Private investors would carry the risk.
The problem and the counterpart
Compensation for turned-away wind
≈ €55M
Modelled range €28M–€84M for 1,266 GWh in 2024 — ≈ €30 per household
The same energy, mined at today’s network†
≈ €110M
≈ 1,500 BTC gross revenue, before costs — the net result is below
† Bitcoin figures · Modelled gross revenue at the BTC price (€75,800) and network hashrate on 4 October 2026 (stored snapshot — live data unavailable), before hardware, power and running costs. Not financial advice.
Does it pay?
Revenue is only half the picture. A fleet that runs only on surplus has to be big enough for the surplus when it comes, and sits idle the rest of the time. With surplus available for 2,000 hours a year, absorbing the 2024 volume takes a fleet of ≈ 570 MW running 23% of the year.
Under our central assumptions
At today's price the fleet would not cover its costs: ≈ €110M of revenue against ≈ €170M of costs a year. It breaks even at ≈ €110K per BTC, or with cheaper hardware, more hours of surplus, or no network charges behind the meter. Change the assumptions below.
Net result a year: BTC price against hours of surplus
The 2024 volume (1,266 GWh), every other assumption at its central value. Modelled; € a year after hardware, site, operations, network charges and payments to generators.
| BTC price | 1,000 hours | 2,000 hours | 4,000 hours | 6,000 hours |
|---|---|---|---|---|
| €50,000 | ≈ -€220M | ≈ -€96M | ≈ -€32M | ≈ -€11M |
| €75,800(today) | ≈ -€190M | ≈ -€58M | ≈ €6.5M | ≈ €28M |
| €100,000 | ≈ -€150M | ≈ -€21M | ≈ €43M | ≈ €64M |
| €150,000 | ≈ -€74M | ≈ €54M | ≈ €120M | ≈ €140M |
Orange: covers its costs. Fewer hours of surplus mean a larger fleet idle for more of the year, which is why hours matter as much as price.
Net result, a year†
≈ -€57M
Does not cover its costs under these assumptions. Breaks even at ≈ €110K per BTC.
- Fleet size
- ≈ 570 MW · runs 23% of the year
- BTC earned (after pool fee)
- ≈ 1,503 BTC
- Gross revenue
- ≈ €110M
- Hardware and site, a year
- ≈ €130M
- Operations
- ≈ €9.1M
- Network charges
- ≈ €23M
- Paid to generators
- ≈ €11M
- Net
- ≈ -€57M
† Modelled. Block subsidy only (fees excluded); network and price held constant for the year. Not financial advice.
How it compares with the alternatives
| Option | What it does | Time to deliver | Who pays | Limits |
|---|---|---|---|---|
| Grid reinforcement | Removes constraints at the source | Years to a decade | Billpayers, through network tariffs | Slow to plan and build; does not remove system-wide curtailment |
| Batteries | Shifts surplus by a few hours | 1–2 years | Developers, recovered through markets and capacity payments | Hours, not days — long windy spells overwhelm them |
| Interconnectors | Exports surplus to Britain and France | Years | Billpayers and developers | Neighbouring grids are often windy at the same time |
| Electrolysers (hydrogen) | Turns surplus into hydrogen | Years; early stage in Ireland | Developers, often with public support | Low round-trip efficiency; needs hydrogen users and storage |
| Heat and demand response | Moves existing demand into windy hours | Months to years | Customers and suppliers | Limited by how much demand can move |
| Flexible mining | Adds demand that runs only on surplus | Months | Private investors | Volatile revenue; hardware waste; noise; needs connection rules |
These are complements, not rivals: the grid needs reinforcing whatever else happens. Flexible mining's distinguishing features are speed and private funding; its weaknesses are volatile revenue and public acceptance.
Three policy options
A
Publish dispatch-down data by node and hour
EirGrid publishes annual totals. Hourly figures by grid location would show investors — of any flexible load, from batteries to electrolysers to mining — where and when the surplus occurs, and let the public check the cost.
B
Define interruptible flexible demand in connection policy
Create a connection category for demand that agrees to run only when the system has surplus and to stop on instruction, with faster, cheaper connections in return. Loads that cannot meet the terms do not qualify.
C
Pilot co-location at a constrained renewable site
A time-limited pilot behind the meter at a wind farm with high constraint, privately funded, with published data on hours run, energy used, response times and the effect on compensation.
Options A and B are technology-neutral: they help batteries, electrolysers and demand response as much as mining. Option C tests the claims on this site with published data.
Objections, answered
“Bitcoin mining wastes energy.”
The option here uses only power that is already being turned away. If the rules allow it to run at any other time, it would add demand like any other load — which is why option B makes interruptibility a condition of connection.
“It will raise emissions.”
Not if it runs only on surplus wind and stops when the system needs the power: at those moments the extra demand is met by wind that would otherwise be turned down. Enforcement is the point of options B and C.
“It will compete with homes and industry.”
A load that must stop on instruction cannot take power others need. The test is whether the instruction is followed, quickly and every time — something a pilot can measure.
“It doesn’t make money.”
At today’s price, under our central assumptions, a fleet that runs only on surplus does not cover its costs (see above). That is for investors to judge; the policy options cost the public nothing either way.
Looking ahead: a 20-year scenario
A scenario, not a prediction: renewable capacity follows Ireland's published targets, and dispatch-down rises with it on a grid that is not reinforced. Move the sliders to test the assumptions.
Read this first
Dispatch-down, and the share flexible demand could use
Gross mining revenue from that energy, before costs†
The steps are the halvings (2028, 2032, …): each one halves the new bitcoin issued per block, so the same energy earns less unless the price rises.
Milestones
| Year | Wind + solar | Dispatch-down rate | Dispatch-down | Block subsidy | Gross revenue† |
|---|---|---|---|---|---|
| 2026 | 7 GW | 11% | 1.7 TWh | 3.125 BTC | ≈ €90M |
| 2030 | 22 GW | 23% | 11 TWh | 1.563 BTC | ≈ €170M |
| 2035 | 33 GW | 31% | 22.3 TWh | 0.781 BTC | ≈ €85M |
| 2040 | 43 GW | 40% | 37.4 TWh | 0.26 BTC | ≈ €24M |
| 2045 | 53 GW | 48% | 55.4 TWh | 0.098 BTC | ≈ €6.6M |
Want to help?
Policymakers, renewable operators with constrained sites, and anyone who wants the data published — there is a way to get involved.