Multiple Radiator Fans

Caterham Twin Radiator Fans

Previously I’d done some sums that calculated my alloy race Caterham radiator could only cool a 4.0 (4031cc) V8. I was therefore fairly worried my 4.6Ltr Rover V8 would overheat. As these engines are famous for their overheating issues, I’ve added a second Radiator fan.

How to improve your cooling system

There is a lot talked about all the improvements you can make to stop a car over-heating. I’ve read dozens of articles. Some are very technical, comparing laminar flow to turbulent flow and various arguments are very hotly debated, so I decided to do some simple measurements using a few cheap tools.

I boiled a kettle, plugged up the radiator hoses and measured how long it took for water to drop from 60°C to 20°C. The reason I didn’t pick 100°C was because it took time for the radiator fins to heat up and stabilise. Without a fan it took approximately 190 seconds to drop 40°C. It was approximately 14°C in the workshop.

I now have two 12v slimline fans

Speed: 2100rpm
Power Supply: 12v DC, 7A, 80w
Air Flow Rate: 1130 cbm/hr, 2150cfm

When I held the first fan against the radiator, without a fan shroud, the air-flow was barely noticeable. After straightening all the radiator fins, the air flow was slightly better, but still not as high as I expected.

I used a cheap anemometer and a simple sum to get Mass Air Flow. You can’t just stick one of these in front of a fan as air speed and mass would be different across the length of the blades. Therefore I made a 3 foot long cardboard tube, slightly funnelled to get an average turbulent flow.


m = ρ*V*A

m = Mass flow rate
ρ = Density of the liquid or gas(kg/m3) – 1.225,
V = Velocity of the liquid or gas(m/s)
A = Flow Area of the Liquid or gas(cm2) – 706.85834715

One Fan – No Radiator

Without a radiator interrupting flow, a 12″ 80watt curved blade fan moved air very quickly.

Direction V (m/S) m
Sucking 3.8m/S 3290.43
Blowing: 4.3m/S 3723.38

If you put your face in front of one of these 12″ fans, it’s like sticking your head out the car window on a motorway. Put a radiator in the way and the air flow is similar to opening the window whilst parked.

One Fan, A Radiator a no Fan Shroud

The radiator dropped the air Velocity considerably. Without a shroud it was obvious air was going in all directions apart from through the radiator. In fact 87% of the air mass (m) didn’t go through the radiator!

Direction V (m/S) m 60°C to 20°C
Sucking 0.5m/S 432.95 121 secs
Blowing: 0.8m/S 692.72 75 secs

One Fan, A radiator & a Fan Shroud

I’d made sure that the shroud was sealed air tight to the radiator with silicon rubber edging. Without the edging you could hear the air whistle through the tiny gaps and clearly the air would rather escape sideways than pass straight through the radiator.

With the fan shroud, air speed was clearly up. But what was really incredible was the speed the water now cooled. To get the same 40°C temperature drop took 1/3rd the time. I wasn’t expecting that sort of improvement!

Direction V (m/S) m 60°C to 20°C
Sucking 1.6m/S 1385.44 38 secs
Blowing: 1.8m/S 1558.62 33 secs

Fan Shroud

The first fan I fitted, was sucking air through the radiator from behind. The fan didn’t seem to perform quite so well sucking as blowing, so I checked to see if there was enough room to mount a fan in front of the radiator and blow. If I made another fan shroud 8mm shallower, then yes.

A sucking fan would be behind the radiator, this would not restrict the incoming flow when turned off.

Fan shroud

2 Car Radiator Fans with Shrouds

Now, having two fan shrouds, I bought a second fan and mounted them both. One pushing air, the other blowing. As you’d expect, with double the fan power, the improvements were far more significant than making sure the shrouds were air tight.

Two fans now pushed air close to the velocity of one uninhibited by a radiator. The air flow with two fans didn’t just flutter a rag, it blew tools off the bench! As you turn the second fan on, you can hear the first fan accelerate as well, making both fans perform better.

Direction V (m/S) m 60°C to 20°C
Sucking & Blowing (12volt parallel) 3.3m/S 2857.47 21 secs

Twin Radiator Fan Shrouds

The noise of air blasting through the radiator is very noticeable. I’m hoping that when mounted in the car, the engine will drown it and the bodywork will muffle it.The weight increase may have been a problem if I was building an ultra light racer, but this is a 300bhp V8 4×4 with custom bodywork.

Using the crude equation, from an ancient car mag, I found I needed to make similar air flow improvements a spacious engine compartment would bring. My radiator was close to being big enough but still inadequate.

With two 12″ radiator fans, a large oil cooler and 7″ Fans, I’m confident that air into the engine bay is about as good as it will ever be. I now have concerns that the super snug engine compartment will prevent the air from exiting smoothly.

Oil Cooler Fan

Originally I had the oil cooler behind the radiator with the fan sucking air and directing it towards the engine. I’ve moved it to the side of the engine compartment blowing air out of the engine bay. I have another 7″ fan for the other side too. The combination of fans pull air in, then push it back out.

I’m going to heat wrap the exhausts, but I’m also going to scoop outside air in from both sides of the engine compartment and funnel that over the exhausts.

Diameter blade to blade 7″
Depth 2.5″
CFM ratio 1550 cfm
Maximum fan CFM: 1,730 cfm
Maximum fan RPM: 2,250 rpm
Amp draw: 12.70 amps
Overall Size: H : 220mm W : 245mm

Oil Cooler Fan Shroud

When I start the bodywork, I will construct another shroud to make sure all air entering the front grill is funnelled to the radiator. The simple edge sealant trick, proved that even when assisted by a fan, air would rather take any path than pass through a radiator. Anything than can be done to force air to take radiator route should help my cooling worries.

Without measuring a similar set-up I’m about as confident as I can be. The air flow is definitely on par if not significantly more than the viscous fan fitted to Rover SD1’s. 

Do Multiple Radiator Fans cool better than one?

Although I did a few sums, it was the measurements that proved more air flow does, in my situation, improve cooling dramatically.

Getting that air flow is down to two things. Most importantly it’s down to directing air through the radiator and not allowing it to pass around the sides.

Is Fast Air Flow better than Slow?

In my set-up, I proved that making the air flow faster also increased the Mass of Air passing through the radiator. Doubling the air speed produced an 83% Air Mass increase and a 35% reduction in cooling time.

What would be interesting to test, given a radiator wide enough, is whether 2 side by side fans cooled better or worse than the same fans sucking and blowing from opposite sides. This would give a correlation between mass of air and speed of air. Is a high Mass Air flow but slow air speed better or worse than fast air but low Mass Air Flow?  I suspect Mass of Air is the key, If the mass is the same, cooling rate would be the same. However, depending on the set-up, getting that extra Mass may require higher air speed. – but who knows without testing.

This is my theory….
If air within the engine bay was trapped, near the exhausts would become super heated, whilst air away from the engine would stay cooler. Circulating air (turbulent flow) would even out the temperature distribution but ultimately the same amount of trapped heat energy would remain. If the air wasn’t trapped and changed on a constant basis that energy could be dispersed outside the engine bay. The Mass of Air would enable better dissipation.

With engine cooling, it is not only important that air cools the water in the radiator, it is also important that air passing over the engine, in particular the exhausts, isn’t already hot. Static air would simply heat up to the same temperature as the engine and give no cooling benefits. A high turn-over of air ensures surface cooling of the engine and exhausts. Slow air flow would firstly get heated passing through the radiator then again by the engine.

Other Flow restrictions & considerations

If the radiator was crammed up right to the engine there would be another flow restriction. It is important to have low pressure air behind the radiator. An under-tray between the radiator and engine could help with this.

Exhausts without heat shrouds will dissipate heat into the engine bay and not along the length of the car.

When engine run efficiently they also run cooler. Retarding ignition timing begins the combustion process later in the cycle and makes heat, therefore advancing timing can reduce heat.

A thin radiator with a large surface area is going to cool better than a small thick one with the same capacity.

Importance of a Fan Shroud

A simple fan shroud, doesn’t weigh much and doesn’t draw huge amounts of power. At cruising speed, ensuring all air through the grill passes through the radiator does far more than a fan.

Whilst stationary, a thermostat controlled fan is vital. If you’ve ever owned a vintage car, you’ll know that cars don’t run too well from cold, with the top speed halved. Getting to temperature quickly not only reduces wear it helps performance. As a car overheats, the power drops and it begins to rattle.

A second fan does indeed speed up cooling, however the relationship between air mass and cooling times are non linear. Therefore the improvements from each additional fan are diminished.

A single fan knocked 153 seconds of the cooling time. A second fan only knocked a further 17 seconds off. However at high temperatures that difference may be essential.

On start up the fans seemed to draw huge amounts of current. My ammeter only measured up to 20Amps, but I’m suspecting current draw on start up could be close to 80Amps. It takes almost 15seconds for the current to drop to ~10amps. The 7Amps quoted by the manufacturer seems a little optimistic.  If you have multiple fans it makes sense to stager the temperature they kick in at, otherwise you’ll be looking at misfires. 2 Fans take similar current to a starter motor, hence I’m using 80Amp relays.
12v 80A 5pin Car relay

To offset the massive start-up current draws, I’m using 2 different fan switches. 79°C / 88°C  (INTERA033866). and 76°C / 86°C (50250). Hopefully those 3°C offset start-up by several seconds. Blow @ 76°C, suck at 79°C. I have an 80°C water thermostat.

As well as offsetting the start times, I’m using an automatic fan control system that starts the fans at half voltage. This seems to considerable reduce start up current to somewhere around 40Amps.

With two fans I might have to upgrade my 80amp alternator. I’ll also have to make sure all the fan wires are short and thick.

Hopefully the air under the bonnet will be pretty turbulent as the temperature across the engine bay will be more constant.

Automatic 2 Speed Fan Control

When two similar 12v fans are wired in series, each fan has 6 volts across it. Each fan will run at reduced speed. Two fans sucking and blowing, running at 6 volt each (series), produce 22% more air flow than one fan but the noise is considerably reduced from when running at 12 volt (parallel).

Direction V (m/S) m 60°C to 20°C
Sucking & Blowing (6 volt each) 2.2m/S 1904.98 28 secs

To get the fans to run at full speed, they must be wired in parallel with 12volts across each. This can be done with a series of relays.

Starting a stationary fan will draw huge amounts of current, but increasing the speed of one already running will take much less current.

When just the low temperature switch is active both fans run at reduced capacity. When the higher temperature fan switch kicks in, both fans run at full capacity.

If the low temperature switch fails then we can add a diode to ensure both fans kick in at full speed with the high temperature switch.

Automatic 2 Speed Fan Wiring

A good article is here:


Caterham Radiator Maximum Cooling Capacity

Caterham Race Radiator Fan ShroudI'd purchased a cheap, second hand custom made, aluminium Caterham radiator designed for a full race 2Ltr Duratec engine. Although it had seen some stone damage it was essentially in good order. I had a cheap eBay radiator but I wanted to use this on another Hot Rod project and ...

Aluminium Radiators - Galvanic and Electrolytic Corrosion

I'd bought a super cheap aluminium kit car radiator from ebay. Was it big enough? What about Aluminium radiator corrosion? I had done my sums to make sure it was big enough for my V8. See - What size radiator do I need for my car? Start with 2 cubic inches ...

What size radiator do I need for my car?

Lots of people; including Westfield; use the VW Polo MK2 1043cc (63.6ci) radiator (1983 - 1990), but is it big enough for a V8 and is there anything else I could do to improve cooling? The Polo radiator fits inside a Westfield nose-cone almost perfectly. see link here. They can be bought ...

Rover V8 Engine Numbers

If it ain't broke don't fix it! Unfortunately, In this case it may need some fixing. I've got a Rover V8 for measuring and set-up purposes. I'll strip it down, see what it needs and if it's not too far gone, rebuild it. It looks like a late SD1 block ...

IVA problems

Can you imagine getting all this Haynes Roadster through the IVA test? With all these parts sticking out at the front, there would be no chance of getting this through an IVA exam. With the decision to build a completely new chassis for my other build, I now have a ...


Front suspension design has been complicated by the 4x4 running gear. It quickly became obvious that it would be impossible to use the ubiquitous Ford Mk2 Escort steering rack as it would try to pass straight through the Sierra Front diff. The push rod for the coil-over shock also tried ...

Sierra Tripod Differential

The Sierra Diff was quite hard to measure and draw up. I started with the push in shaft 7" 'Tripod' version. Most of the time drawing up this Ford Sierra Differential was taken during the measurement stages. When there are so many dimensions to be taken you almost need a 3D ...

Chassis Collection

Having debated building a 'super seven' style car for decades, when a chassis came up on ebay with a reserve of only £0.99, temptation and curiosity took hold. The auction finished with only 3 bids at £56.55. Surely this was a bargain??!? A 550mile round trip would answer that. The unfortunate answer ...

Translate »

Web Design by Go Web Solutions