Tech Focus: Cooling This month it’s everything you’ve ever wanted to know about cooling but were afraid to ask… Overheating is one of the biggest concerns of BMW ownership, so let’s take a closer look to fully understand how it all works, and, more importantly, how it can be improved… Words: Jamie King.
The internal combustion engine is a wonderful creation, but it isn’t exactly the most efficient thing in the world. Almost a third of an engine’s energy is wasted as heat output. This is hardly surprising really when you consider it is a ‘combustion’ engine, and when things combust they generally get very hot! This heat energy needs to be dissipated away from the engine, so cooling is absolutely vital.
When you say ‘engine cooling’ most people will immediately think of water and radiators, but it’s not just water temperatures that need be kept under control.
Oil, charge air in turbo applications, gearbox and diff oil, and in some cases power steering fluid, all absorb heat as they work and therefore need to be cooled to keep their temperatures within a safe operating window.
There are a number of variables that affect a cooler’s characteristics, and that’s what we’re going to look at here…
What is a cooler and how does it work?
Most cooling products, including water radiators, intercoolers, and oil coolers, all work in a similar way. The water, air, or oil to be cooled (working fluid) is passed through a cooler. This cooler is, or should be, situated in a place where it receives a constant flow of air as the vehicle moves forward.
The working fluid passes through the core of the cooler and transfers its heat to the (usually) aluminium core of the cooler itself. This heat is then dissipated to the surrounding air as the much colder airflow passes through the foil, or fins, of the cooler core. It’s a fairly simple principal but there are a number of factors that need to be considered in order to produce a cooler capable of doing what it should.
On the subject of different coolers, another type of cooler is a chargecooler. Its job is to cool the charge air on a forced induction engine. Unlike an intercooler, which is an air-to-air radiator, a chargecooler uses water to cool the charge air.
Essentially, it consists of a tube for the charge air to pass through, which is within a water-filled tank. As the charge air passes through the tube, the heat is transferred from the air to the surrounding water. The water is then pumped back into the engine cooling system and cooled by the water radiator.
The size of the cooler has a massive affect on its cooling properties, but contrary to popular belief it is not simply a case of ‘bigger is better’. The cooler needs to be correctly sized for its application. It is more to do with volume of the cooler and mass of the working fluid than the actual size of the cooler. For example, an air-to-air intercooler the same size as a tiny oil cooler wouldn’t be any good, likewise an oil cooler as big as a radiator wouldn’t work either. This is where things start to get a bit scientific. It’s all about the specific heat capacity of the working fluid, which relates to the amount of heat energy one kg of that working fluid can transfer. Air, for example, has a specific heat capacity of 1.01 kilojoules per kg. This means that every single kg of air can transfer 1.01kj of heat energy. Oil, on the other hand, has a specific heat capacity of 2.13kj per kg. Water is even higher at 4.18kj per kg.
These calculations are all based on mass (kg), so when you factor in the volume required to achieve this mass it starts to paint a clearer picture. One kg of water equates to one litre. One kg of oil is slightly more at 1.14 litres (oil is less dense than water). Whereas one kg of air fills a huge 1114 litres of space!
Yes, one thousand, one hundred, and fourteen! So you can see why air-to-air intercoolers need to be much larger than oil/water coolers in order to have the same cooling potential.
One other consideration affecting the size of a cooler is the temperature decrease required. Oil, for example, generally only needs to be cooled by 10-15 degrees, whereas charge air coming from a super hot turbocharger needs to be cooled by as much as 50- 60 degrees for best performance.
The optimum size of a cooler is a compromise between different factors. The most common factor is the physical amount of space available in the engine bay in which to fit it. Intercoolers also have to consider the fact that the more volume there is, the more laggy the engine will be because of an increased pressure drop across the cooler.
Water radiators and oil coolers can also suffer from pressure drop and while the end result may not be as immediately noticeable as increased turbo lag, it can, however, be even more detrimental. The oil/water pump will have to work harder to force the fluid through the cooler, which can lead to cavitation with potentially fatal consequences for an engine. So, you want a cooler that is just big enough to offer the required temperature decrease, but not too big as you may start to run into other problems if you do.
Where the cooler is placed probably has the most influence on its performance. Contrary to common belief, it’s far more involved than cutting huge holes in the bodywork and assuming you’ve increased airflow and therefore optimised performance of your cooling system. It’s not just about how much can get to the cooler, but how effective it is, how quickly and easily it can pass through the cooler, and, in some applications, how much drag it creates.
Massive coolers may give better results on a test bed, but in reality, when they are mounted to the car, their performance is directly related to how much air can actually reach them through bodywork apertures, cooling ducts, and alike. A huge radiator will perform better than a smaller one on a test bed, but if in situ that particular application only has a letterbox-size opening through which it channels cooling air it won’t perform as well as a smaller radiator with a more efficient stream of airflow.
The speed at which the cooling air passes through the intercooler also needs consideration, and the cooler’s positioning can alter this. For example, an oil cooler mounted on the roof will receive plenty of unrestricted airflow, but the air will pass through it so quickly (especially at motorway speeds) that it won’t have time to absorb any heat from the oil within it, and therefore its effectiveness will be compromised. With the same oil cooler mounted within the confines of the engine bay, the speed at which the air passes through it will be considerably less due to the shrouding effect created by the bodywork. But obviously, if the air speed passing through the cooler is too slow, it won’t work efficiently either.
Another factor to consider, although somewhat less of a concern with mainstream road cars, is the drag created by a cooler. In race applications, especially single seaters where aerodynamics play a huge part in the car’s performance, the positioning of the various coolers can drastically affect the amount of drag the car produces.
The core is made up of two main parts; the tubes that the working fluid passes through, and the foil, or fins, which the surrounding air flows through as the vehicle moves forward. This ‘tube and fin’ design is the most common for most applications, but some front mount intercoolers use a ‘bar and plate’ design instead. The two are very similar in the way they work, but the stronger construction of a bar and plate design means it is capable of dealing with higher boost pressures more effectively and is more robust against things like stone chip damage.
The design of the core and the number of tubes used will have a huge effect on a cooler’s characteristics. For example, a core with four rows of 50mm tubes will have greater cooling potential than a single row core with a 20mm tube. The number and sizes of tubes used within a core depends on what application it is intended for.
A number of variables can be altered to give a core its particular properties. The first is the size of the tubes. A larger tube will flow more fluid than a smaller one, but the pressure drop across the core will be greater than that of a smaller tube.
This can be counteracted to some extent by the use of multiple rows of tubes. Two tubes will flow the same mass of working fluid at half the speed of a single tube of the same size. This results in a lesser pressure drop across the core.
However, by increasing the number of rows you also increase the thickness of the core. A thicker core will have a greater drag effect on aerodynamics and the speed that air flows through the core itself will decrease. This can lead to problems getting the air through the cooler and out the other side, causing detrimental effects to the cooling. One way to combat this is to use multiple smaller tubes, rather than one larger tube. For example a two-row core with two 25mm tubes will flow the same mass of working fluid at the same speed as a single row 50mm core.
Smaller tubes are also structurally stronger than larger items. The lower height/width ratio of the tube means it’s physically stronger and therefore less susceptible to ‘ballooning’ under pressure. The downside to using multiple smaller tubes is that they are much more fiddly, and therefore expensive to produce. There are at least twice the number of joints that need to be fused, which adds to the time taken to produce and the costs. Also multiple row cores tend to be heavier than single row ones.
You may think that the more rows in a core, the better it is at cooling. To an extent this is true, but there will come a point when too many rows, and a core that’s too thick, starts to have a detrimental effect. For example, a four-row core with four 20mm tubes will have better cooling potential than a core with only three rows of the same size tubes.
However, this difference will be smaller than you might think. And the downsides the extra ‘fourth’ row brings can in some cases outweigh the positive gains. The most effective row is the first one – the first part of the core that the cooling air comes into contact with. The second and third rows have an ever-decreasing impact, each being less effective than the row before it. Therefore the fourth row right at the back of our example cooler will only offer a small amount of cooling potential, but the increased thickness it adds will cause the speed of air flowing through the core to slow and offer more restrictions than a three-row core, so it may not be worth adding the extra row.
Welded and extruded tubes
There are two types of tubes used within coolers: welded and extruded.
Welded tubes are relatively cheap and easy to produce as they are simply flat sheets of metal that are folded into a tube and then welded together. It is also possible to insert something called ‘turbulators’ within welded tubes too. These are primarily designed for intercoolers where the increase in surface area is needed to transfer the heat.
Turbulators have a dual function, increasing the surface area available and also increasing rigidity to prevent ballooning under pressure.
An extruded tube is much stronger than a welded tube, but is significantly more expensive too. The process involves forcing aluminium through a specialist machine at very high pressures and temperatures to create a seamless tube that can cope with much higher pressures than a welded tube. However, due to the cost and practicality of producing varying sizes, an extruded tube is usually reserved for use is specialist applications where its strength and ability to cope with high pressures is required.
The foil, or fins, are the part of the cooler that the cooling air passes through. Heat from the tubes is transferred to these fins, where the cooling air rushing over them takes the heat away and therefore the cooler, and the working fluid within it, experiences a temperature drop.
The pitch of the foil used can alter the speed at which the air is able to pass through the cooler core, and therefore can affect the cooler’s performance. The more open the foil is, the less restrictions it will pose to the cooling air passing through the core. But the trade off is the number of fins (surface area) available within a given space is reduced, affecting the cooling efficiency. Tighter packed fins offer greater cooling properties, but come at the expense of increased drag and a slower speed of airflow passing through the core. That’s why the foil used needs to be carefully considered for the application and its intended use.
BMW cooling problems
The design, construction, and positioning of most BMW coolers is actually pretty good straight from the factory, but many models (especially the E36/E46) are renowned for poor cooling systems. How so? Well, in these cases it’s not actually performance issues that are the problem, but it is a more of a concern for reliability. Many standard cooling systems (like the E36/E46 engine cooling system) uses plastic components. These are cheap and easy to produce at manufacture, but after time can become brittle and cause leaks to the system.
It goes without saying that for any cooling system to work it needs to be sealed, and any leaks in that circuit will result in failure to keep the temperatures under control. Let’s take the engine coolant system as an example; any leaks will cause the volume of coolant within the system to fall. When this reaches the point that the volume of coolant left in the system isn’t sufficient to absorb the amount of heat the engine is producing, temperatures start to rise, rapidly. And they will continue to rise until something gives – usually a very expensive cylinder head warping!
To prevent this the cooling system, and all of its ancillaries, needs to be kept in tip-top condition. Replace items such as the thermostat housing, radiator cap, expansion tanks, bleeder screws, and even the radiator itself at regular intervals (every 75k miles) to avoid any disasters. Make a note of the dates/mileage these have been replaced at too, as a visual inspection simply isn’t good enough – something may look perfect today but then fail tomorrow, there is no way of telling.
You can, however, make a few upgrades to the system to aid both performance and reliability. Replacing the radiator with an all-aluminium performance item is a good start, as leaks can occur where the aluminium core and plastic end tanks meet on standard items. Likewise an alloy thermostat housing will not crack or break like the plastic item, but some of these are notorious for not sealing correctly with some owners reverting to an OE plastic housing. Replacing the standard rubber hoses, which can perish over time, with performance silicone items is another wise investment.
In short, don’t skimp on servicing your BMW’s cooling system or it will come back to haunt you…