Rondo: the heat battery is also a steam contract

Bricks can shift electricity consumption in time. The harder commercial question is who guarantees the steam when the factory needs it.

Offaxis Research · Published 14 October 2026 · Primary sources checked 14 October 2026 · 6-minute read

A factory’s steam demand does not conveniently stop when electricity becomes expensive. That mismatch gives thermal storage its commercial opening. The useful question is how much of the factory’s energy bill can move into cheaper hours without moving its production schedule.

What the equipment actually stores

Rondo describes electric resistance heaters warming refractory bricks by radiation. Air carries stored heat out of the brick stack; heat can then be supplied to industrial processes, including as steam. Electricity enters, but heat is the stored product. This is a different comparison from a battery whose principal job is returning electricity to the grid. [1]

The distinction changes what an analyst should measure. A thermal-storage capacity in MWh is an amount of energy; a steam-delivery rating in MW is a rate. Neither, by itself, says how long a particular factory can run. The relevant boundary is usable heat delivered at the required temperature and pressure, including the losses and auxiliary loads of the chosen installation.

Buying a machine versus buying its output

Rondo’s product page offers equipment purchase or lease, with the customer procuring charging electricity. It also offers a heat purchase agreement: the customer buys delivered heat, typically steam, without upfront capital or managing battery charging. The stated pricing structures include a fixed heat price or a discount against gas-based heat. These are advertised options, not disclosure of any particular customer’s signed terms. [2]

Our interpretation: the second model makes power procurement and heat delivery part of the service being sold. It can reduce the customer’s coordination burden, but the contract still has to allocate availability risk, electricity-price exposure and remedies when steam specifications are missed. A financeable machine and a financeable service are related achievements; they require different evidence.

Two projects, two levels of evidence

Rondo’s Holmes Western case study labels a 100 MWh installation commercially operating from September 2025. It describes off-grid solar charging and continuous steam delivery alongside conventional boilers. Its 16 October 2025 announcement reported ten weeks of operation and efficiency above 97%. Those are supplier-reported operating claims; neither page provides the full time-series data needed for independent replication. [3] [4]

Covestro provides customer-side evidence for a different site. Its 19 January 2026 announcement records groundbreaking at Brunsbüttel, Germany, with commissioning planned for the end of 2026. It describes a 100 MWh unit intended to supply 10% of the site’s steam. Covestro’s 2025 annual report repeats that timetable. As of this review, the sources checked support a construction-stage project, not a verified commissioning result. The steam share is a target, not measured output. [5] [6]

Evidence ledger · checked 14 October 2026
Question	Evidence available	Still unresolved
Has it operated?	Holmes Western: supplier reports commercial operation since September 2025 [3].	Independently audited availability and full-year performance.
Is the German project online?	Customer documents target end-2026 commissioning [5, 6].	Commissioning confirmation and acceptance-test results.
What can customers purchase?	Equipment purchase/lease or a heat purchase agreement [2].	Project-specific prices, guarantees and financing obligations.

The margin lives between two energy bills

Our analysis: a useful comparison starts with the delivered cost of usable heat. Electricity purchases must include grid charges and any demand charges, then account for conversion losses. Equipment, financing, maintenance, backup and site integration also need a place in the calculation. The alternative is the cost of producing the same usable heat with the factory’s existing system, not the headline price of gas.

Cheap hours are valuable only if enough charging power is available during them. A constrained connection can prevent a site from capturing a short price window. On-site solar introduces another set of questions: seasonal generation, land, backup and whether the storage volume matches overnight demand. Thermal storage changes the timing problem; it does not remove the energy-supply problem.

The next useful evidence is therefore mundane: accepted steam specifications, metered charging and delivery, availability through maintenance, and the cost of retaining backup. A customer-confirmed acceptance test would strengthen a deployment claim. A disclosed contract boundary would clarify the business model. Neither can be replaced by a fundraising announcement.

Offaxis conclusion: Rondo is worth following as an industrial-heat supplier whose product links electricity scheduling to factory operations. The reviewed evidence includes a supplier-described operating installation and a customer-confirmed construction project. It does not establish universal cost savings, every project’s profitability or independent verification of efficiency.

Primary sources & further reading

Rondo: technical energy path — undated product explanation; manufacturer account.
Rondo: product and purchasing options — live page, checked 14 October 2026.
Rondo: Holmes Western case study — supplier-reported status from September 2025.
Rondo: operating announcement — 16 October 2025.
Covestro: Brunsbüttel groundbreaking — 19 January 2026; customer-side project evidence.
Covestro 2025 annual financial report: climate change — published 26 February 2026; see “Climate-Neutral Steam”.

This is original editorial analysis of public primary sources. Offaxis has not tested the equipment or audited the contracts. Technical and commercial unknowns are identified above.