Technology

ARBOK-DryTower

ARBOK-DryTower rejects heat by evaporating a working liquid under deep vacuum rather than by blowing air across finned surfaces.

Overview

ARBOK-DryTower rejects heat by evaporating a working liquid under deep vacuum rather than by blowing air across finned surfaces. The physical lever is the gap between specific heat capacity and latent heat of vaporisation: water carries 4.18 kJ/(kg·K) as sensible heat but 2,440–2,450 kJ/kg as latent heat under vacuum. One tonne of evaporated water therefore removes about 678–681 kWh of heat — enough to cool 584 tonnes of water by 1 °C, 58 tonnes by 10 °C, or 5.8 tonnes by 100 °C.

Against that, the stated internal energy cost is 2 kWh of electricity per tonne evaporated. The ratio gives 0.29–0.30% of the rejected thermal power, and the figure is nearly independent of the cooling range because it is set by the ratio of electrical input to latent heat, not by ΔT.

The equipment consequence is equally direct. Classic dry cooling buys its performance with structure: a 120–220 m concrete tower for natural draft, or banks of fans for mechanical draft, plus hundreds of finned heat exchange columns. ARBOK-DryTower replaces all of it with two horizontal vessels, a vacuum group, an internal vapour-transfer fan and a tube bundle. No high-rise civil works, no massive tower foundations, no large finned air-side surface.

Applications

Primary use cases:

  • Replacement of dry cooling towers and air-cooled condensers where the plot, the height or the civil works schedule is the constraint
  • Replacement of open wet towers where droplet aerosol, drift and Legionella exposure are the constraint
  • Thermal and nuclear plants facing summer curtailment on river temperature limits
  • Refineries, petrochemicals, metallurgy, pulp and paper, large compressor stations
  • Data centers and district cooling
  • Desalination heat rejection

Target scale: modular; the reference calculation is built on the 10,000 m³/h circulation loop used across the ARBOK cooling series (116 MW at ΔT 10 K, 186 MW at ΔT 16 K, 418 MW at ΔT 36 K).

Operating Principle

The working liquid is held under deep vacuum in a horizontal vessel. At the reduced pressure it boils at low temperature, and the latent heat of vaporisation is drawn from the liquid remaining in the vessel and from the tube bundle carrying the process water. The vapour is moved by an internal fan and the vacuum group maintains the operating pressure.

Because the mechanism is phase change rather than convection to air, the heat transfer coefficients are those of boiling and condensation — one to two orders above single-phase liquid convection and far above air-side convection. This is the structural reason the equipment can be small: it is not a better air cooler, it is a machine that does not depend on moving large masses of air.

Working liquid: water or alcohol of ARBOK's own manufacture. Alcohol is the engineering choice where the vapour-transfer fan governs the design, because at the same temperature its saturation pressure is several times higher and the vapour therefore several times denser — see §5.

Cycle boundary — confirmed closed. Vapour is condensed and returned; the working liquid is charged once and is not consumed. Makeup covers leakage only. Two consequences follow and must both be carried forward:

  1. Water balance is genuinely zero. No evaporative loss, no blowdown, no drift. The "70–95% water reduction" wording in ARBOK-CoolTower is superseded by zero and should be corrected there.
  2. No power is spent on the cooling itself. Heat rejection at the cold end is passive; the system regulates the degree of recuperation on its own. There are no cooling fans and no air-moving load, and the 0.72 kWh per tonne is therefore a net figure for the whole installation.

Scope note — what happens to the recovered heat is a bonus, not a constraint. Handling and monetisation of the heat leaving the cold end sits outside this card. It is covered by the adjacent ARBOK technologies — TERU, TEG Panels, Arbok-STEC, ARBOK-TELORISE, LB HEAT — and is treated as an added-value option on top of the cooling duty, selected per project. ARBOK-DryTower is specified, sold and compared as a cooler; anything obtained from the rejected heat is upside and is priced separately.

Key Parameters

Thermodynamic base

| Parameter | Value | Origin |

|—|—|—|

| Latent heat of vaporisation, water under vacuum | 2,440–2,450 kJ/kg | source document |

| Heat removed per tonne evaporated | 678–681 kWh, taken as 680 | source document |

| Equivalent sensible duty, 1 tonne evaporated | 584 t of water by 1 °C / 58 t by 10 °C / 5.8 t by 100 °C | calculated, 2,440 ÷ 4.18 |

| Electrical input | 2 kWh per tonne evaporated | source document |

| Heat rejected per kWh of electricity | ≈ 340 kWh | source document |

| Auxiliary load as share of rejected heat | 0.29–0.30% | 2 ÷ 680 |

Auxiliary load against the market

| System | Auxiliary load | Scope of the figure |

|—|—|—|

| ARBOK-DryTower | 0.29–0.30% | vapour-transfer fan only — scope to be confirmed, see §13 |

| MVM EGI Heller, natural draft | 0.7–0.8% | complete cooling system |

| Modular dry coolers, market (DCX 1 MW) | 2.6% | complete unit, 32.4 kW on 1,236 kW |

| Modular dry coolers, upper range | 2.7–5.4% | complete unit |

Capital cost

| Parameter | Classic dry cooling | ARBOK-DryTower |

|—|—|—|

| Full specific cost, turnkey | 120–150 $/kW(th) | 21–32 $/kW(th) |

| Equipment-only supply price quoted by MVM EGI | 8–15 €/kW(el) = 5.3–12.5 €/kW(th) | not applicable |

| Concrete tower 120–220 m | yes | no |

| High-rise foundations | yes | no |

| Large finned air-side surface | yes (238–264 Forgó columns per tower) | no |

| Modularity and erection speed | low | high |

| Phased capacity growth | limited | good |

Vapour volumetric flow — why the working liquid matters (calculated, ARBOK SIR, 30 °C basis)

| Liquid | Saturation pressure | Vapour density | Volumetric flow at 186 MW |

|—|—|—|—|

| Water | deep vacuum | 0.030 kg/m³ | ≈ 2,515 m³/s |

| Methanol | deep vacuum | 0.277 kg/m³ | ≈ 578 m³/s |

Alcohol reduces the volumetric duty on the vapour-transfer fan roughly fourfold at equal thermal power. The counterpart is mass: latent heat of the higher alcohols is 582–708 kJ/kg against 2,440 kJ/kg for water, so 3.5–4 times more mass must be evaporated for the same heat. Whether that mass is recovered or lost is the open question of §13.

Internal circulation of the medium (calculated): at 186 MW the circuit evaporates and recondenses ≈ 274 t/h on water, or ≈ 1,030 t/h on a higher alcohol at 650 kJ/kg. This is internal throughput, not consumption — the loop is closed. It sizes the vessels, the transfer fan and the return line; it does not appear in the water balance.

Vapour-transfer fan power is fluid-independent (calculated, ARBOK SIR, 2026-08-12). For a given temperature lift the work is Carnot work and does not depend on the medium: the denser vapour of an alcohol needs a proportionally higher pressure rise per kelvin, and the two effects cancel exactly.

| Medium | Δp per 1 K | Volumetric flow per MW | Fan power at η 0.7 |

|—|—|—|—|

| Water |deep vacuum/K | 13.59 m³/s | 4.71 kW/MW |

| Methanol | 1,deep vacuum/K | 3.11 m³/s | 4.71 kW/MW |

| Butanol |deep vacuum/K | 56.49 m³/s | 4.71 kW/MW |

Ideal Carnot work for a 1 K lift at 303 K is 3.30 kW/MW. Reading the stated consumption backwards through this relation:

| Stated consumption | Share of rejected heat | Implied temperature lift |

|—|—|—|

| 2 kWh per tonne (gross) | 0.296% | 0.63 K |

| 0.72 kWh per tonne (net) | 0.107% | 0.23 K |

The figures are internally consistent and describe a very-low-lift vapour mover, not a heat pump. The machine transfers vapour at essentially the same temperature; it does not raise the heat to a higher grade. Consequence to hold: the heat sink must already sit at close to the required cold-line temperature.

Module capacity in a standard 20-ft container (calculated). Container 6.058 × 2.438 × 2.591 m; plot 14.77 m²; internal cross-section 5.62 m². Vapour duct Ø1.6 m (2.01 m²) at 50 m/s.

| Cold line | Methanol: pressure / capacity | Butanol: pressure / capacity |

|—|—|—|

| 30 °C | deep vacuum / 33 MW | deep vacuum / 2.6 MW |

| 45 °C | deep vacuum / 62 MW | deep vacuum / 6.6 MW |

| 60 °C | deep vacuum / 109 MW | deep vacuum / 14 MW |

| 75 °C | deep vacuum / 181 MW | deep vacuum / 29 MW |

| 90 °C | deep vacuum / 285 MW | deep vacuum / 53 MW |

| 105 °C | deep vacuum / 428 MW | deep vacuum / 90 MW |

| 120 °C | deep vacuum / 619 MW | deep vacuum / 144 MW |

Reference point: one 20-ft container ≈ 30 MW on methanol at a 30 °C cold line. For comparison, DCX places 1.24 MW of complete dry cooler, air side included, in the same container.

Containers required for the 186 MW reference duty, and the resulting plot:

| Cold line | Methanol | Butanol |

|—|—|—|

| 30 °C | 6 containers, 89 m² | 71 containers, 1,049 m² |

| 60 °C | 2 containers, 30 m² | 13 containers, 192 m² |

| 90 °C | 1 container, 15 m² | 4 containers, 59 m² |

| 120 °C | 1 container, 15 m² | 2 containers, 30 m² |

A Heller tower of the same 186 MW duty occupies 4,500–6,000 m².

Fluid selection follows the temperature level, not preference. Capacity scales with vapour density, so the medium must be matched to the cold line: methanol where the cold line is low, butanol only where it is high. At 30 °C butanol gives 2.6 MW against methanol's 33 MW — a factor of 13. Above 90 °C the gap closes and butanol becomes the better choice on safety and material grounds (flash point 35–37 °C against 11 °C, compatible with aluminium, water separates as a second phase and can be decanted out).

> Scope of the figures above. These are vapour-transfer capacities of the container — how much thermal power one module moves. The cold-end package is sized per project and quoted separately; see the scope note in §4.

Architecture and Components

  1. Two horizontal vessels — evaporation and vapour handling under vacuum. No concrete tower, no basin, no high-rise structure.
  2. Tube bundle for the process water being cooled.
  3. Vacuum group — maintains operating pressure and removes non-condensable gases.
  4. Vapour-transfer fan — internal, moves vapour between vessels.
  5. Piping, instrumentation and automation.

Cost structure is dominated by vessels, tube bundle, vacuum pumps and the transfer fan. The most expensive items of classic dry cooling — the tower, its foundations, the high-rise erection package and the large finned surface — are absent.

Module capacity, vessel dimensions, weight, vacuum level, fan pressure rise, tube bundle area: [требует уточнения из базы]

Advantages

Energy. Auxiliary load of 0.29–0.30% of rejected heat, set by the ratio of electrical input to latent heat and therefore almost independent of cooling range. Against the best classic dry system on natural draft (0.7–0.8%) this is 2.5–2.7 times lower; against ordinary modular fan-driven units it is 9–18 times lower. Subject to the scope confirmation in §13.

Capital. 21–32 $/kW(th) against 120–150 $/kW(th) for turnkey classic dry cooling — a factor of 4–7. The saving comes from removing the concrete tower and everything attached to it, not from cheaper heat exchange.

Operating. No permanently running high-power fan banks and no large finned surfaces to clean. Classic dry towers require washing once a year in spring and twice where dust loading is high.

Dynamics. Fast start-up and fast response to load change. This inverts the competitor's own claim: MVM EGI presents the thermal inertia of several thousand m³ of circulating water as a strength, which for load-following generation and for data centers with load steps is a defect.

Sanitary. No open basin, no droplet drift, no salt aerosol, no visible plume. The Legionella habitat is removed architecturally rather than treated. MVM EGI holds the same property and does not market it — the position is unoccupied.

Siting. No high-rise civil works removes the geological, seismic and height-clearance constraints that govern where a 120–220 m tower can be built at all.

Fluid flexibility. Water or alcohol of own manufacture, selected against the temperature level and the vapour-fan duty.

Integrations

  • ARBOK-SONAR — 24/7 monitoring of the working medium with automatic protection.
  • TERU — recovery of parasitic heat from temperature differences.
  • Plant side: condensers, process heat exchangers, in place of the classic wet or dry tower circuit.
  • Vacuum competence: shares the platform logic of ARBOK-VC (Vacuum Cracking) and Vacuum Osmosis.

Interfaces to plant SCADA/DCS, retrofit sequencing: [требует уточнения из базы]

Deployment & Operation

Delivered as modular equipment; capacity scaled by module count rather than by construction. Reference economics across the ARBOK cooling series are built on 8,000 operating hours per year.

Installation duration, commissioning procedure, staffing, maintenance interval, working-liquid makeup regime: [требует уточнения из базы]

TRL

Not assigned. The source document presents thermodynamic calculation and comparative economics; no pilot installation, reference site or third-party validation is stated.

Benchmark for the gap: MVM EGI holds over 20,000 MW of installed Heller capacity, 147 dry-cooling references since 1954, and heat exchangers from 1987 and 1989 reused in new units in 2017 — a demonstrated service life of 40–55 years. In power generation this weighs more than a CAPEX difference.

Market Potential

Cooling towers and cooling fields exist at thermal and nuclear plants, refineries, chemicals, metallurgy, pulp and paper, compressor stations, data centers and cold-storage terminals. Demand drivers: tightening water abstraction rules, Legionella liability legislation, summer river temperature restrictions, rising industrial water prices, and the capital and schedule burden of high-rise cooling structures.

Segment guidance. Direct competition with MVM EGI on 1,000 MW power blocks is decided by references, not by parameters. The addressable near-term segment is industrial closed-circuit heat rejection — refineries, chemicals, metallurgy, pulp and paper, compressor stations, data centers — where procurement is faster and 1954 references are not demanded.

Market size in monetary terms: [требует уточнения из базы]

Typical Project Economics

| Parameter | Value |

|—|—|

| Specific CAPEX | 21–32 $/kW(th) |

| Competitor turnkey CAPEX | 120–150 $/kW(th) |

| CAPEX advantage | factor of 4–7 |

| Auxiliary load | 0.29–0.30% of rejected heat, scope pending |

| Reference loop | 10,000 m³/h, 8,000 h/yr |

| Reference duty | 116 MW at ΔT 10 K / 186 MW at ΔT 16 K / 418 MW at ΔT 36 K |

Working-liquid makeup cost, dry residue handling, module replacement, insurance: [требует уточнения из базы]

Risk Factors

Closed on 2026-08-12: the cycle is closed, vapour is condensed and returned, the medium is not consumed. Water balance is zero and the working-liquid cost argument is void — the charge is one-time. What this does *not* remove is the need to reject heat to the environment; it relocates the question to the condenser.

Open questions that must be closed before any external communication (raised ARBOK SIR, 2026-08-12):

Also closed 2026-08-12: no power is spent on the cooling itself — heat rejection is passive, and the degree of recuperation is regulated by the system. The consumption figure of 0.72 kWh per tonne is net.

Structural risks:

  • Deep vacuum with a flammable working liquid admits air on any leak; ingress creates a combustible mixture inside the equipment and non-condensable gases degrade condenser performance. Vacuum pumps run automatically on pressure rise; inert blanketing to be confirmed.
  • No pilot, no reference, no third-party validation against a competitor with 20 GW installed.
  • Cold line is delivered as a configured guarantee rather than a fixed physical limit (§4a). The internal approach of the exchange stage itself still has to be measured and stated — the table in §4a assumes 11 K, which is the best figure the competing school reports. If the real value is better, every line in that table improves; if it is worse, they all shift.

Related Technologies

Inside the solution (ARBOK's own means of delivering the specified cold line — internal, not offered or priced separately):

DRY CLOUD · Arbok-Criojet · ARBOK-SONAR · TERU

Adjacent:

ARBOK-CoolTower · ARBOK-ClosedTower · ARBOK-VC (Vacuum Cracking) · Vacuum Osmosis · ARBOK Digital Twin · ARBOK Low-Carbon Water · TSM Ceramic