Remote Radiator for Generator Sets: Types, Selection, and Installation Guide

When engine-room space, airflow, or noise constraints rule out a conventional cooling setup — this is how the cooling system is redesigned around a remote radiator.

1. What Is a Remote Radiator?

On a conventional generator set, the radiator sits directly in front of the engine. A belt-driven or motor-driven fan pulls air through the core, rejects the heat from the coolant into the surrounding air, and the cooled fluid returns to the engine in a continuous loop. Everything is packaged together — engine, alternator, radiator, fan — on a single base frame.

A remote radiator breaks that arrangement. The radiator is physically separated from the engine and installed at a different location entirely — on a rooftop, on an outdoor skid, in a dedicated plant room, or at any point where hot air can be discharged freely to the outside. The coolant still circulates between the engine and the radiator, but it does so through an extended pipework run rather than through a direct connection.

How this differs from a conventional setup

 

Core characteristics of a remote radiator system

● Physically separated from the engine: The radiator is not on the generator base frame. It may be metres away — or on a different floor or roof level.

● Fan is always motor-driven or hydraulic: Because there is no mechanical connection to the engine crankshaft, the fan must have an independent drive. Belt drive is not an option in a remote configuration.

● Extended coolant pipework: Hot coolant travels from the engine to the radiator and back through insulated pipework. Pipe run length, diameter, and routing all affect system performance.

● Elevated installation possible: The radiator can be installed above the engine — on a roof or elevated structure — as long as the coolant circuit is designed to account for static head and the expansion tank is positioned at the highest point of the system.

● May include auxiliary circulation pump: For long pipe runs or high static head, the engine-mounted water pump may not provide sufficient circulation pressure. A separate auxiliary pump is added to maintain adequate coolant flow.

2. When Do You Need a Remote Radiator?

Not every generator installation needs a remote radiator. But several common situations make the conventional engine-mounted configuration impractical — or impossible.

Insufficient engine room space or ceiling height

A conventional radiator requires clear air inlet and outlet zones around the genset — typically several metres of free space in front of and behind the unit. In basement plant rooms, tight mechanical rooms, or enclosed containers, that space is simply not available. The radiator cannot discharge hot air without recirculating it back to the inlet, which quickly drives coolant temperature above safe limits.

This is the most common trigger for a remote radiator project. Generac and other major OEMs specifically identify space constraints and inadequate engine-room airflow as the primary conditions calling for a remote cooling configuration.

Hot air discharge requirements

Indoor generator installations must reject heat to the outside. If the engine room is inside a building, ductwork and louvres can route air in and out — but the resistance added by the ductwork reduces airflow through the radiator, cutting cooling performance. For large gensets or high-ambient sites, the ductwork resistance cannot be managed with a conventional setup. A remote radiator positioned outside eliminates the problem by placing heat rejection where it belongs: in open air.

Noise isolation

Generator sets are acoustically treated to reduce noise — acoustic enclosures, anti-vibration mounts, silenced exhaust. A radiator mounted on the genset frame is mechanically coupled to that noise source. Moving the radiator to a separate structure, or to the building exterior, removes it from the primary noise path and simplifies the acoustic design of the installation.

Multi-unit shared cooling

Large installations running multiple generator sets — a data centre with eight 2,000 kW units, an industrial facility with multiple prime power gensets — can share a single remote cooling plant rather than running eight separate radiator and fan assemblies. Shared cooling also simplifies N+1 redundancy design: one additional cooling unit serves all engines rather than duplicating every individual radiator.

Original engine design without a belt-driven fan

Some engines — particularly larger gas and diesel units — were never designed with a belt-driven fan pulley. Others have been installed with heat exchanger cooling and are being retrofitted with a radiator. In both cases, there is no mechanical drive available for a conventional fan, and a remote radiator with a motor-driven fan is the natural solution.

3. Types of Remote Radiators

Remote radiators are not a single product type. They are categorised across four different dimensions, and a specific installation may involve combinations of these classifications.

3.1 By Installation Structure

Vertical radiator

The fan axis is horizontal — the fan blows air horizontally through the core. Vertical radiators have a small footprint and are suitable for narrow sites, wall-adjacent installations, or locations where horizontal airflow is preferable. Commonly used in container installations, compact plant rooms, and sites where the width of the installation is constrained.

Vertical remote radiator — horizontal airflow, compact footprint

Horizontal (flat-mounted) radiator

The fan axis is vertical — the fan blows air upward through the core. Horizontal radiators offer a larger heat exchange surface area and higher thermal efficiency per unit of plan area, since the full core face is exposed to rising air. Widely used for rooftop installations and large open-air sites where vertical clearance is available.

Horizontal remote radiator — vertical upflow, rooftop installation

V-shape cooler

The core is arranged in a V-configuration with two angled sections. This arrangement provides significantly more heat exchange surface area in the same plan footprint as a flat core — making it the preferred choice for high-capacity dry cooler systems where space is limited but cooling load is large. Common on data centre cooling plants and large power generation facilities.

V-shape cooler — increased surface area for high-capacity applications

3.2 By Cooling Method

Air-cooled dry cool

The coolant circulates in a closed loop and heat is transferred to the ambient airstream by convection across the fin-and-tube core. No water is consumed in the process. Dry coolers are the standard choice for generator set applications — reliable, low-maintenance, and effective across a wide range of ambient temperatures.

Performance degrades at high ambient temperatures because the temperature differential between coolant and air decreases. A properly sized dry cooler accounts for the maximum expected ambient at the installation site.

Evaporative / adiabatic assisted cooler

A dry cooler with an evaporative pre-cooling system added. In high-ambient conditions (typically above 35°C), the incoming air is wetted by spray nozzles or wetted media pads before it reaches the core. Evaporation of the water cools the air temperature — sometimes by 10°C or more — before it contacts the heat exchange surface. This restores or improves the temperature differential and significantly increases cooling capacity under peak conditions.

Adiabatic cooling is the preferred solution for installations in the Middle East, sub-Saharan Africa, and other high-ambient environments where a standard dry cooler would need to be impractically large to handle peak loads. It does consume water during the spray periods, which must be factored into site water supply planning.

3.3 By Coolant Circuit Configuration

Single circuit

The radiator handles the engine main coolant circuit only — the high-temperature (HT) jacket water loop. Simpler to design and install. Suitable for smaller diesel or gas generator sets without charge air cooling, or where the charge air cooler is separately managed.

Dual circuit / multi-circuit

Larger diesel and gas generator sets — particularly high-output turbocharged engines — operate two separate coolant circuits:

○ High-temperature (HT) circuit: Jacket water cooling. Typically 85-95°C operating range.

○ Low-temperature (LT) circuit: Charge air cooling (intercooler/aftercooler). Typically 40-55°C operating range.

A dual-circuit remote radiator handles both loops independently, with separate core sections or separate cores operating at different temperature levels. This configuration is standard on large Cummins QSK, MTU 4000 series, Perkins 4000 series, and CAT 3500 series engines in prime and continuous duty applications.

3.4 By Fan Drive Type

Electric motor-driven

The fan is driven by an AC motor (fixed speed or variable frequency drive). This is the most common fan drive for remote radiators — the motor provides independent speed control, can be operated immediately at full speed regardless of engine speed, and has no mechanical coupling requirement to the engine. VFD control (variable frequency drive) allows fan speed to be modulated based on coolant temperature, reducing noise and power consumption during light-load operation.

Motor IP rating must match the installation environment. Minimum IP54 for standard outdoor locations; IP55 or higher for coastal, offshore, and mining environments. See Section 5.5 for environmental protection requirements.

Hydraulic motor-driven

The fan is driven by a hydraulic motor supplied from the engine hydraulic circuit or a dedicated hydraulic power unit. Used in applications where electrical power supply to the radiator location is not available, where ATEX (explosion-proof) requirements prohibit standard electric motors, or where very high fan power is required in a compact motor package.

Related reading: For a detailed comparison of electric vs belt-driven fan configurations and selection parameters, see Belt-Driven vs. Electric Motor-Driven Fans: How to Choose the Right Fan for a Generator Radiator.

4. How Does a Remote Radiator Cooling System Work?

Understanding how the full cooling circuit connects — and where the potential failure points lie — is essential before specifying a remote radiator. The system is more complex than a conventional engine-mounted setup, and the additional components each introduce design decisions that affect performance.

The basic cooling circuit

The fundamental loop is unchanged from a conventional system: the engine generates heat, the coolant absorbs it, and the radiator rejects it to the ambient air. What changes is the physical path between engine and radiator:

1. Hot coolant exits the engine at the thermostat outlet — typically 85-95°C depending on the engine and duty cycle.

2. The hot coolant travels through insulated pipework to the remote radiator location.

3. Inside the radiator, heat transfers from the coolant through the tube walls and fins to the airstream generated by the fan.

4. Cooled coolant (typically 65-75°C) returns through the cold pipe to the engine water pump and re-enters the engine circuit.

5. The cycle repeats continuously.

See system diagram above: red lines indicate hot coolant flow; blue lines indicate cooled return flow. Dashed sections indicate extended pipe run through building or structure.

When the engine-mounted pump is not sufficient

The water pump built into the engine is sized to circulate coolant through the engine block and a radiator positioned close to it — not through 20 or 30 metres of pipework. As pipe run length increases, so does hydraulic resistance: friction losses in long pipe runs, additional bends and fittings, and vertical head all add pressure drop that the engine pump must overcome.

When the total system pressure drop exceeds what the engine pump can sustain at the required flow rate, coolant velocity drops, heat transfer deteriorates, and engine coolant temperature rises above the target range. This threshold varies by engine and pump specification, but as a general guideline:

● Pipe runs up to ~10-15m: Engine-mounted pump is usually sufficient with correct pipe sizing.

● Pipe runs beyond ~15m, or with significant elevation change: An auxiliary circulation pump is typically required.

● Dual-circuit systems: Each circuit may need independent pump sizing.

The auxiliary pump is sized based on the required coolant flow rate and the total system pressure drop calculated from the pipe run geometry. It is typically positioned on the cold return line, upstream of the engine water pump.

Heat exchanger as a circuit break

In some system designs — particularly where the remote radiator uses a different coolant chemistry than the engine, or where isolation between the engine circuit and the external pipework is desirable — a heat exchanger (brazed plate or shell-and-tube type) is installed between the engine circuit and the remote radiator circuit.

This creates two independent loops: a primary loop (engine coolant) and a secondary loop (remote radiator circuit). Heat transfers between them through the exchanger. Advantages include circuit isolation (glycol chemistry can be optimised independently for each loop), corrosion protection (the external circuit can use inhibited water without affecting engine coolant specification), and simplified maintenance access. The trade-off is an additional approach temperature difference across the heat exchanger, which must be accounted for in the thermal design.

5. Design Parameters

The following parameters define the cooling system and must be established before a remote radiator can be correctly specified. This section covers each parameter in detail and explains the design decisions that follow from it.

5.1 Cooling Capacity

The starting point for any cooling system design is the engine heat rejection rate — the quantity of heat the coolant must remove per unit of time, expressed in kilowatts (kW) or BTU/hr.

This figure is not the same as the generator's rated electrical output. A 1,000 kW generator set may have a jacket water heat rejection rate of 450-550 kW — depending on the engine model, fuel type, and duty rating. Engine manufacturers publish heat rejection data in their technical datasheets for each engine model at each rated duty point (ESP, PRP, LTP, COP per ISO 8528-1).

For dual-circuit systems: Heat rejection must be established separately for the HT circuit (jacket water) and the LT circuit (charge air cooler). These circuits operate at different temperature levels and may require separate core sections within the remote radiator.

The remote radiator's rated cooling capacity must meet or exceed the engine's total heat rejection at the specified ambient temperature and airflow conditions — not at a standard reference condition that may not match the actual site.

Related reading: For a detailed explanation of how engine heat rejection maps to cooling system load, see Where Does the Heat Go When an Engine Runs?.

Related reading: For cooling capacity requirements across standby versus continuous duty ratings, see Standby vs. Continuous Duty Radiator Rating: How to Select the Right Generator Radiator.

5.2 Piping Distance and Pressure Drop

The hydraulic design of the coolant circuit must account for the total pressure drop through the system — from the engine outlet, through the supply pipe, through the radiator core, and back through the return pipe to the engine.

Pressure drop accumulates from:

● Straight pipe friction losses: Proportional to pipe length, velocity squared, and the Darcy-Weisbach friction factor. Longer runs at higher flow rates produce significantly higher losses.

● Fittings and bends: Each elbow, tee, valve, and fitting adds equivalent pipe length. A complex routing with many bends can double the effective pipe length.

● Radiator core pressure drop: The coolant-side resistance of the radiator itself, dependent on core design and flow velocity.

● Static head: Covered in Section 5.3 below.

Pipe diameter is selected to limit flow velocity to a range that balances pressure drop against pipe size and cost. Typical design targets are 1.0-2.5 m/s for coolant in carbon steel or copper pipe.

All pipework must be insulated to minimise heat loss from the hot supply line and prevent freezing on the cold return in low-temperature environments.

5.3 Static Head and Elevation

If the remote radiator is installed at a different elevation than the engine — above it on a roof, or below it in a basement — the height difference creates a static head effect that must be accounted for in the hydraulic design.

● Radiator above engine: The pump must overcome the static head of the coolant column in the supply pipe. This directly adds to the total system pressure drop that the pump must deliver against.

● Radiator below engine: The static head assists the pump on the supply side but opposes it on the return — net effect depends on the circuit layout.

For rooftop installations — a common configuration for data centre and commercial building generator sets — the radiator may be 10-20m above the engine. At a typical coolant specific gravity of 1.05 (water-glycol mixture), a 10m elevation adds approximately 1.03 bar of static head that the pump must overcome. This is a significant load that can exceed the capacity of the engine-mounted pump alone.

Design requirement: The expansion / surge tank must always be positioned at the highest point in the system. See Section 5.4.

5.4 Expansion / Surge Tank Location

The expansion tank serves two functions in a remote radiator system:

● Volume compensation: As coolant temperature cycles between cold start and operating temperature, its volume changes. The expansion tank accommodates this volume change without pressurising the system beyond its design limit.

● Air elimination: Air trapped in the coolant circuit accumulates at the highest point of the system. The expansion tank — positioned at that highest point — provides a venting point for air to escape and be removed from the circuit.

Critical rule: The expansion tank must be the highest point in the coolant circuit. If the remote radiator is on the roof, the expansion tank must be at or above the radiator level, not at engine level. Positioning it lower than the radiator creates an air lock that cannot be vented.

Tank sizing is based on the total coolant volume in the system — engine water jackets, pipe volume, radiator internal volume — and the expected temperature range. A system with long pipe runs has significantly more total coolant volume than a conventional engine-mounted setup, and the expansion tank must be sized accordingly.

5.5 Ambient and Environmental Conditions

The operating environment of the remote radiator determines both its thermal performance and its material and protection specifications. These are not independent — an environment that pushes performance limits often also imposes harsher material requirements.

Ambient temperature

The radiator must be sized for the maximum ambient temperature at the installation site — not an average or standard reference temperature. High-ambient sites (above 35°C) require either a larger core, higher airflow, or evaporative pre-cooling to maintain the required temperature differential between coolant and air.

For the Middle East and parts of Africa, peak ambient temperatures of 45-50°C during summer months are not unusual. At these conditions, a radiator sized for a standard 25°C or 35°C reference point will be significantly undersized.

Altitude

Air density decreases with altitude. At 1,500m above sea level, air density is approximately 83% of sea-level density — meaning a fan delivering the same volumetric airflow carries less mass of air and therefore removes less heat per unit time. Both fan selection and radiator sizing must account for site altitude, typically by increasing airflow or core size to compensate.

Dust and particulate loading

In high-dust environments — mining sites, desert installations, construction sites — airborne particulate progressively blocks the fin matrix of the radiator core, increasing air-side pressure drop and reducing airflow. Fin pitch selection must account for dust loading: a finer fin pitch that maximises clean-air heat transfer will block faster in dusty conditions than a coarser pitch with more open area.

Where dust loading is severe, the radiator design may incorporate cleanable core geometry, access panels for pressure washing, or pre-filtration. Maintenance intervals for fin cleaning must be specified as part of the system design.

Humidity

High relative humidity accelerates corrosion on exposed metal surfaces. For the radiator core, this affects the choice of fin material and protective coating. Copper and cupronickel fins resist humid-environment corrosion better than bare aluminium. For the radiator frame and structure, hot-dip galvanising or high-performance paint systems are required.

Coastal and salt spray environments

Coastal and offshore installations are exposed to salt-laden air that aggressively corrodes standard carbon steel and unprotected aluminium. The corrosion category for these environments is typically C4 (high) or C5-M (very high, marine) per ISO 12944.

Material and coating requirements for C5-M environments:

● Fins: copper, cupronickel, or aluminium with epoxy coating

● Tubes: copper or cupronickel

● Headers and tanks: brass or stainless steel

● Frame: hot-dip galvanised steel with C5-M certified topcoat

● Fan motor: minimum IP55, preferably IP66 with marine-grade paint

Related reading: For more detail on corrosion protection specifications and coating standards, see Manufacturing Process of Engine Radiators.

Corrosive industrial atmospheres

Industrial sites with chemical processing, sulphur compounds, ammonia, or acidic emissions require site-specific material assessment. Standard corrosion categories may not capture the specific chemical exposure. Sinrui's engineering team can advise on material selection for specific industrial environments based on site atmospheric data.

6. Typical Application Scenarios

Data centres — rooftop installation

Data centre generator sets are typically installed in basement plant rooms or at ground level inside the building envelope. Space and airflow constraints almost always rule out engine-mounted radiators. Remote radiators are positioned on the building roof, connected to the engines via insulated pipework routed through the building services risers. For large data centres running multiple generator sets, a shared remote cooling plant with N+1 or N+2 redundancy is common.

Fan noise at the roof level must comply with planning conditions. Variable frequency drives on the fan motors allow noise to be managed during off-peak hours.

Oil and gas — offshore and coastal platforms

Offshore platform generator sets operate in C5-M corrosion environments with high humidity, salt spray, and potentially ATEX classified areas. Remote radiators are positioned at the platform edge or on elevated structures where clean sea air can be used for cooling. Hydraulic fan drives are often preferred in ATEX zones where electrical motors require expensive Ex-rated equipment.

All materials — fins, tubes, frames, fasteners — must be specified for marine exposure. Routine inspection and cleaning intervals are defined at the commissioning stage.

Mining — enclosed facility or surface plant

Mine surface facilities and enclosed workshops present the same space and airflow constraints as industrial buildings. Remote radiators are positioned on the building exterior or on a dedicated cooling skid adjacent to the genset facility. High dust loading in mining environments requires coarser fin pitch, cleanable core geometry, and defined cleaning intervals.

For underground mining applications, the remote radiator may be located at the shaft head or ventilation point, with long coolant pipe runs to the underground generator sets.

Industrial facilities — multi-unit prime power

Large industrial facilities running generator sets at prime or continuous power duty often operate multiple units. A shared remote cooling plant — serving two, four, or more generator sets from a central radiator bank — simplifies maintenance access, allows centralised redundancy, and reduces the number of fan assemblies to maintain.

Dual-circuit remote radiators are standard in this configuration, since large continuous-duty engines invariably use separate HT and LT coolant circuits.

7. Remote Radiator Selection Checklist

The following checklist covers the information required to specify and design a remote radiator system. Providing complete data at the enquiry stage allows the engineering team to produce an accurate technical proposal without extended back-and-forth.

Not all parameters may be immediately available. Send what you have — our team will identify any gaps and follow up with targeted questions.

 

Stay updated on generator cooling systems, industrial radiator design, and power generation insights — subscribe to the SINRUI  LinkedIn Newsletter for technical articles, project case studies, and product updates. Have a cooling application to discuss? Reach out to our engineering team at sales@sinruiradiator.com.