Radiators: Copper or Aluminium?
Submitted by John Faubion
Reproduced courtesy of the KartWeb.com
As Mr. Denman knows I'm one of the very few kart racers that run a copper radiator. Anyone that has seen my kart probably thinks I must not know anything since I run such a small radiator. However they also do not know that I have a hard time getting the water temp over the 100-degree mark! I use a copper heater core for a radiator. This system is a single pass core with the water entering and exiting from opposite ends. I run a Honda CR-125 with more compression than the law should allow. I made 36 hp with this engine after running it for a full season with the same piston and ring! Yes, you read that correctly, I ran an entire season on one piston and ring and still pulled 36 hp from it. However to get back to the subject at hand, one of the keys to this success is the copper radiator. Why?
Lets face it the goal of any radiator is to transfer excess heat from the engine right? The amount of heat a radiator can remove is related to three things, the efficiency, the surface area, and the temperature delta of the mediums. Lets consider the efficiency of the radiator first.
Copper will transfer 384 watts/meter-degree centigrade (w/m-įC) while aluminum only transfers 155 w/m-įC. Therefore copper is nearly two and half times more efficient at transferring heat than aluminum. In fact only silver is a better conductor of heat. But considering the weight and cost, a silver radiator is probably not a good idea. But isnít copper is heavier than aluminum?
Granted the density of copper is 3.3 times that of aluminum so your might think that the copper radiator would weight 3.3 times as much but it doesnít. Copper is 2.8 times stronger than aluminum as well so the copper doesnít have to be as thick. So a copper radiator will only be slightly heavier than the same size aluminum radiator. However since copper is 2.5 times more efficient than aluminum its surface area can be 2.5 times smaller to transfer the same amount of heat! So a correctly sized copper radiator could actually be lighter than the aluminum radiator it replaces.
Since the copper radiator is smaller we can install it perpendicular to the airflow to minimize the airflow restriction. The larger aluminum radiator usually has to be mounted at an angle. This causes us to force the air to change direction to pass through it. Aerodynamically this is very bad because it also causes air turbulence behind the radiator. Ideally we duct the air into and out of the radiator to better control the air turbulence. I currently run ducting into the radiator but not out of itÖ yet! We have covered the efficiency and the surface area now lets get to an area most kart racers have never considered, the temperature delta!
If a radiator has an efficiency of 75% and we pass water at 140 degrees through it what should we expect the water temp to be at the outlet if the air temp is 80 degrees? If you said 95 degrees then you donít need to be reading this anyway! For everyone else, the temperature drop across the radiator is the difference or delta between the water temp (140) and the air temp (80) times the efficiency (75%) or (140-80)*.75=45 degrees. This is the temperature drop across the radiator and since we started with 140-degree water our outlet temp is 95 degrees. If we change the temperature delta we can take more heat from the system. As an example, lets say the ambient temperature is 60 degrees while all other factors stay the same. We will now drop the temp by (140-60)*.75=60 degrees with a resulting outlet temp of 80 degrees. But we usually donít have control of the ambient temperature do we?
So what typically happens if we are not removing enough heat from the engine? The water temp rises and our delta changes. Letís look at another example. Lets say the inlet temp rises to 160 with our original conditions. We are now dropping the temp by 160-80*.75=60 degrees so our outlet temp is now 100 degrees. Our inlet temp increased by 20 degrees but our outlet temp only increased by 5 degrees. This should help explain why during a race the water temperature sometimes goes higher that we saw during practice and then levels off. As long as we do not exceed the pressure limit of the radiator cap the system will eventually stabilize.
But there are some problems with using a copper radiator. The worst of these problems is corrosion. By running a copper radiator with an aluminum cylinder we create a battery effect. This galvanic action will corrode the aluminum faster than the copper. You may be thinking that you can prevent this from happening by using de-ionized water but you canít. You see once it begins the aluminum will be releasing ions into the water, which will no longer be de-ionized. It may help to slow it a bit initially but the galvanic action will occur.
This same action occurs in cars with iron blocks and copper radiators. Some manufacturers placed a sacrificial anode in the system to protect the radiator and engine blocks. They usually made this anode from zinc. So why not add a chunk of zinc to the system to protect the cylinder? Well zinc is actually nobler than aluminum and would therefore cause the aluminum to corrode faster. While there are a few metals that are less noble than aluminum, most of them (like sodium, potassium, and lithium) react quite violently with water. The only one that could possibly be used would be magnesium. However when hot water is passed over magnesium it releases hydrogen gas. So using magnesium is probably not a good idea for our use.
So what can be done? The only way to completely eliminate this effect is to use all aluminum components through out the system. This would include the sensors as well. Most of us are going to have to deal with it to an extent anyway so the best we can do is to reduce the action as much as possible. We can do this by using some kind of anti-oxidant in the system. The best that is allowed for most of us to run is Red Line's Water Wetter. Water Wetter has been proven to greatly reduce this effect in cooling systems with only a 5% mixture. Another measure that can help is to drain the system between races so that the galvanic action does not have an electrical path between the dissimilar metals. You can also use a Calcium, Lime and Rust (CLR) remover periodically to eliminate any corrosion that has occurred. Just be sure to flush the system thoroughly after the CLR. This is due to the mild acids in the CLR.
Another problem is finding the correct heater core to use as a radiator. Ideally what we want is a large single pass heater core with the inlet and outlets on opposite ends. In fact opposite corners would be even better. This would allow us to mount the radiator for a cross flow and to take the water off the bottom to the water pump inlet. Of course mounting the radiator is another problem since a heater core usually has no mounting provisions since they are usually captured inside a heater box. I am currently working on a design for a ducting system that will incorporate the mount, ducting, and protection screens into a single unit. I know that John has been working on a way of mounting a heater core in the side pod of a kart. His method should provide for quite a bit of protection to the radiator while eliminating some frontal area of the kart.
This article should give you some food for thought during the off-season.
Below are some references used in writing this article.