Heat reuse into Swedish district heating

Reading order for a municipal heat file

Read from the cooling loop outward, then from the municipal network inward. First write the temperature the hall can actually offer in winter and in summer. Then write whether a pipe or a valve exists on the plot. Then name the heat company — Stockholm Exergi, Falkenberg Energi, Falu Energi & Vatten, or another letterhead. Then classify the commercial state: prepared connection, letter of intent, signed heat contract, or dated delivery. Only the last state supports a delivery claim. A Stockholm Data Parks invitation is a programme page, not a delivery note.

Do not start with a national potential or with a homes-equivalent slogan. Exergi’s Sätra sentence uses “up to 10 MW” and “10,000 homes” as a utility comparison. That comparison is Exergi’s wording. It is not an annual measured volume in this register, and it is not a template for a smaller town whose network cannot take summer heat. A sentence that works beside a dense Stockholm pipe does not travel to a municipality that has no spare winter load.

Keep electricity counterparties in another column. EcoDataCenter’s press item about future 80 MW supply with Falu Energi & Vatten and Ellevio is an electricity sentence. It does not prove heat delivered. Ellevio appears in Stockholm Data Parks materials as a power partner. That is not a heat offtake. Svenska kraftnät’s indicative transmission map does not show district-heat headroom. Bidding areas SE1 Luleå, SE2 Sundsvall, SE3 Stockholm and SE4 Malmö move a wholesale power price. They do not move a district-heat molecule from Falun to Malmö.

• Loop temperature in winter and summer.
• Named municipal or commercial heat company.
• Prepared, intended, contracted, or delivered?
• Electricity counterpart kept in a separate cell.

Why Swedish municipalities are counterparties, not scenery

A Swedish district-heat network is usually owned by a municipal company or a mixed municipal-commercial owner. Stockholm Exergi’s own heat-recovery page describes a network of more than 300 Swedish miles of heat pipes and a much smaller district-cooling network, and it presents Open District Heating as a way for data halls and other surplus sources to sell heat at a market price. That is a Stockholm product description. Another municipality may have a smaller network, a higher summer surplus, or no interest in a low-temperature source. The developer does not own that decision.

Temperature is the first technical filter. Air-side rejection often leaves heat too cool or too scattered. Liquid loops and heat pumps can raise temperature so a utility will talk. Exergi’s Conapto page states that heat pumps lift residual heat from about 30 °C to about 70 °C before the heat enters the network. That pair of numbers is Exergi’s description of one contractual design. It is not a national temperature standard, and it is not proof that a hall in Luleå can sell outdoor-air reject heat into a pipe that does not exist on the cited Meta page.

Seasonality is the second filter. District heat is a winter-peaking product in most Swedish towns. A hall that rejects heat every hour of the year may meet a winter need and become a summer problem. A contract that is silent on summer curtailment, backup boilers and interruption is an incomplete commercial state. Stockholm Data Parks materials describe an ambition that halls using renewable electricity and heat recovery could support the city’s fossil-free aim. Ambition is not a measured year of delivered heat.

Planning and environment sit beside the flange. Outdoor heat-pump skids and dry coolers can become neighbour and noise objects. Boverket’s planning guidance may see them as part of the building. Naturvårdsverket’s Environmental Code pages appear if the plant changes an environmental investigation. Those tracks do not issue the heat contract. The heat company issues the heat contract. The municipality in its planning role is a different desk from the municipality as owner of a heat company. Write both desks.

What named primary sources show

Stockholm Exergi’s Open District Heating case page is the dated contractual colour for Stockholm in this set. It says the first Open District Heating agreement between Exergi and Conapto was signed in 2021, that Conapto then connected a hall and installed heat pumps, and that the first facility was enough to heat about 2,500 homes a year — Exergi’s comparison. It then says Conapto also contracted heat recovery from a new hall in Sätra inaugurated in August 2024, with capacity to recover up to 10 MW, compared by Exergi with 10,000 homes. This register stores those figures as utility-page claims. It does not store a measured annual megawatt-hour series, and it does not silently merge Sätra with the 5 September 2024 Conapto notice that opens Stockholm 4 South unless a source names both.

atNorth’s SWE01 Stockholm metro page describes heat recovery with Stockholm Exergi together with liquid cooling, a design PUE of 1.2 and 12 MW site power. Occupied IT load and heat actually delivered are not quantified on that page. Conapto’s Stockholm 4 South opening materials, read with related operator wording, mention recovery toward Stockholm Exergi. That remains operator heat wording until a delivery start or a measured year is published in a source used here.

GleSYS’s 1 October 2020 release says surplus energy from the new Falkenberg hall had begun to become district heat for Falkenberg Energi and lists 3 MW utility power at completion. The later operator page states 5 MW, PUE 1.2 and district-heat recovery. Measured annual megawatt-hours after October 2020 are not in the cited pages. EcoDataCenter describes recovery with Falu Energi & Vatten. The 80 MW future-power sentence with Falu Energi & Vatten and Ellevio stays an electricity row. The 2025 annual report’s 41 MW and 65 MW campus sentences stay separate capacity rows. Heat delivered versus heat-recovery design remains a stated knowledge gap for Falun.

Stockholm Data Parks is a municipal-and-utility invitation naming the City of Stockholm, Stockholm Exergi, Ellevio and Stokab. It is a programme door, not a national heat census. Meta’s Luleå page describes outdoor-air cooling and hydropower and does not, in the cited text, document a district-heat sale. evroc’s 3 February 2025 Arlandastad notice designs for up to 16,000 GPUs and does not name a heat buyer in that notice. No source in this set publishes a national total for data-centre heat delivered in Sweden.

Questions a municipality can actually ask

A planning officer and a heat-company officer should not share one undifferentiated inbox. The planner asks whether the detaljplan allows the outdoor plant, whether noise and traffic files exist, and whether the use is a data hall or a generic industry. The heat company asks whether it wants the temperature, the volume and the interruption terms. A welcome from the first desk is not a offtake from the second.

Useful questions from the heat company include: which season the hall can guarantee a temperature; who pays for the heat pumps; who owns the flange; what happens in a network outage; and whether any measured year already exists. Useful questions from the planner include: which operating cases the acoustic file covers, and whether heat-rejection equipment is inside the building envelope described in the bygglov. Neither desk should be asked to certify a PUE.

Common misreads

The first misread is to treat Exergi’s 10 MW Sätra capacity sentence as delivered heat or as Conapto’s IT load. It is a utility recovery-capacity claim. The second is to treat “2,500 homes” or “10,000 homes” as a register metric. Those are Exergi comparisons. The third is to copy Falkenberg 2020 onto every Swedish coast. It is one dated start in one town. The fourth is to add Falun’s 80, 41 and 65 megawatt electricity rows into a heat figure.

The fifth is to publish a national data-centre heat-market size from a handful of operator and utility sentences. This set has one documented delivery start in Falkenberg, utility contract colour in Stockholm, and recovery wording in Falun. That is not a market size. The sixth is to treat heat reuse as a PUE certificate, or a design PUE as proof of sold heat.

What to ask next

Ask the heat company for the latest paper — drawing, letter, contract or delivery note — with a date. Ask whether any measured megawatt-hour year is published. Ask who pays for heat pumps and who takes summer interruption. If the buyer is unnamed, the commercial state cannot be higher than intended.

Then open the parent heat-reuse page for the four-state vocabulary, the cooling page for the plant that creates the heat, and the noise-and-permit-timelines page for outdoor equipment. Ask the municipality whether the detaljplan actually allows the skid. If the operator will not name a volume, record the gap rather than estimating the town’s take.

1. Named heat company and named hall.
2. Temperature, season and flange owner.
3. Prepared, intended, contracted, or delivered?
4. Any measured MWh year after a start date?
5. Electricity MW kept out of the heat cell.

What this does and does not prove

This page proves that Swedish heat reuse is a municipal-network problem with named counterparties and different evidence states. It proves that Exergi published 2021 and 2024 Conapto-related contract colour, that GleSYS published a 2020 Falkenberg delivery start, and that Falun recovery wording is not a measured heat year in the cited pages.

It does not prove a Swedish data-centre heat-market size. It does not convert homes-equivalents into register megawatt-hours. It does not merge Sätra with Stockholm 4 South without a source that names both. It does not treat Meta Luleå or evroc Arlandastad as heat-delivery case studies on the cited pages.

Unknown delivered heat stays unknown. Point to the named source gap. A later primary source may add a contract date or a measured year. Until it does, prepared and delivered remain different words, Exergi comparisons remain Exergi comparisons, and a national delivered-heat total remains unpublished here.