Charge Air Cooling

[VerusEric]

In the intake manifold thread, there was a point where it got a little off topic and onto the topic of cooling. Thought it would be a nice thread starter so I took a week and wrote up what we know about charge air cooling. My day job is heavily involved in heat exchangers, so this was an easy-ish one for me to write about. I also have substantial first hand experience with methanol injection, lack of intercooling, and calibration challenges it comes with on my 93 MR2 Turbo. Please let me know if this brings any questions with it.

This will primarily be a discussion for forced induction applications; there isn’t much that can done for naturally aspirated cars other than place the inlet into the intake close to fresh air and away from the radiator/radiant heat.

Why does the air heat up?
Compressing air via a turbocharger or a supercharger will cause the intake air to heat up. This is due to the ideal gas law equation PV=nRT. If volume (V), amount of moles (n), and the ideal gas constant (R) all stay the same (which in our case, they do), as pressure (P) is increased, the temperature (T) has to increase as well to keep the equation equal. There is no way around increasing pressure without increasing temperature, this is physics.

What does this mean?

• Air at higher temperatures is less dense then lower temperature air. A denser (colder) charge increases the amount of air molecules that make it into the combustion chamber on the intake stroke, more air + more fuel = more power. So the opposite is true with hotter air, and thus reduced power and performance is realized.
• Disregarding that higher temps can cause pre-ignition/detonation and reduced timing as a result, for every 10 degree F that the intake air temp (NOT ambient air temp) increases or decreases, a resulting ~1% in horsepower is gained or lost.
• However, if we do consider that high temps can cause pre-ignition/detonation, then power can suffer significantly and engine damage could occur.

How can we keep the air cool?

Their are multiple systems that can cool this charge air. We will split these into convention and non-conventional systems. Conventional refers to the use of charge air coolers (CAC), commonly referred to as intercoolers; while non-convention refers to adding fluids to the intake charge to reduce the temps. Some fuels require less or no cooling as well due to the fuels ability to cool the intake charge. OEMs generally rely on the conventional tactics as they are safe, maintenance free, and highly effective.

Conventional Systems:

Air to Air (A2A) charge air coolers
These would be the large front mounts that we all love. Other packages have these located in side-pods, in the rear, and up top in air ducts. The charge air flows through passage ways through the core, and exchanges charge air heat, with the ambient air, similar to how a radiator works. These systems need ambient airflow flowing through the unit to work properly. There are two main variants of A2A cores, bar a plate, and tube and fin. Their are a multiple styles of each of these two as well, with different inserts, fins, and tubes used. Each of these offer its own set of benefits and drawbacks.


OEM Style tube and fin

Extruded tube and fin


Positives:

• Extremely simple design and operating principle
• Not much can go wrong with this system. Leaks would reduce performance but not cause anything catastrophic to occur.
• Costs are generally reasonable with the lower level construction cores.
• With higher level cores (think 4 figure pricing) and adequate airflow, near ambient temperatures can be reached.

Negatives:

• Impossible to cool temps below ambient
• Blocks airflow to the radiator and condenser.
• Some variants, typically the cheaper ones, are extremely heavy.
• The cheaper cores are generally not efficient, but still get the job done.
• Performance is highly dependent on airflow.

Air to Water (A2W) charge air coolers

These are the intercoolers found inside the intake manifolds of some OE cars, Crawford’s s/c, and Innovate’s s/c. With the use of its own cooling circuit, these can be placed in confined spaces and are generally used to cool the charge air temperatures from a supercharger, but can be used with turbo systems as well. This cooling circuit generally involves a low temp heat exchanger up front, the intercooler in the plenum, a water pump, a reservoir for filling, and the lines. Water flows from the low temperature heat exchanger, removing heat from the water, and then flows to the intercooler in the plenum, which exchanges the heat from the charge air and puts it back into the water.


ZR1 Intercoolers in the plenum


Heat exchanger up front.

Positives:

• Ability to be packaged in small spacing and placed within the intake manifold itself.
• Great for drag cars when used with ice, which can reduce air temps below ambient.
• These can reduce pumping losses from reduced tubing and bends.
• The intercooler portion itself can be much smaller than an A2A and still reduce air temps drastically.
• Less impact on the radiator cooling as the heat exchanger is generally very thin.

Negatives:

• These systems are more expensive as they involve more components.
• When used without ice, air temps will be higher than a properly sized A2A.
• While unlikely and uncommon, they can leak water into the engine.
• Less dependent on airflow as the heat exchanger can be placed remotely and a fan used to pull air through them.

Less Common, Non-conventional Systems

Water Injection

By injecting a small amount of water into the intake tract, the water (preferably vapor) can turn to steam and with it, take a significant amount of energy out of the air charge. Water’s high latent heat of evaporation makes it a great fluid to absorb the energy and thus heat from the inlet air. This change in phase (water vapor to steam) is the key to its success. This system can be used in cases when an intercooler is not enough to cool the intake charge.


Coolingmist kit

Postives

• Increased knock resistance
• Reduced inlet charge temps
• Can produce slightly more horespower

Negatives

• Maintenance
• Cost of the system
• Have to remember to fill the reservoir
• Possible failure point (wouldn't be significantly destructive if it fails though)

Methanol/Alcohol Injection
Methanol exhibits two main differences from water when being injected into the intake stream. One, methanol is a fuel with the octane rating of ~109 RON, which will greatly increase the knock resistance. The other difference being methanol evaporates at a much lower temperature then water. Place a small amount of rubbing alcohol on your hand and you can feel it become cool to the touch, more so then water. Water will remove more heat per unit through its latent heat of evaporation due to its physical properties; however, we are restricted by the amount of water we can inject before it will bog the engine. On the other hand, with methanol being a fuel, we are not limited by the amount of liquid we can inject. We can inject a large amount, which in turn can actually lower the intake charge significantly, all while increasing the knock threshold significantly through the increase in octane.


M1 Methanol

Positive:

• Increased octane, reduced knock threshold
• Ability to run more timing
• Ability to run more boost pressure
• Reduced air temps

Negatives:

• Added costs (system and the fuel)
• Remembering to fill the reservoir
• Potential failure point that will ruin the engine
• Calibration challenges

Non-Intercooled Applications:

Some vehicles, generally drag cars, do not need intercoolers due to the primary fuel that is used. These vehicles run on either alcohol, methanol, or nitro-methane. For reasons previously discussed, these fuels can cool the intake air beyond that of an intercooler, rendering the use of one obsolete. Ethanol (E85) reduces the need for intercooling but does not offer enough of a cooling effect to abandon the use of one in most cases.

Conclusion:

We recommend the use of the conventional systems listed above as we share a similar mindset to the OEMs. These systems work well and if designed properly, you won’t need the use of chemical (fuel) intercooling. When chemical intercooling is used and it is not the primary fuel (methanol injection), this can lead to a very dangerous condition and could lead to the demise of an engine. If used with the right precautions though, it is a viable alternative for those willing to take the risk for the reward.

[ATL BRZ]

Great post Eric!

I'll add to this with a pic of the only AWIC turbo FA20 I've ever seen. Owned, built and tuned by Doug at DBW Motorsports here in Atlanta.

[VerusEric]

Figured I would expand on this some more.

So one of the biggest mistakes I see in the industry is huge FMIC. I see this a lot in the Honda drag cars, I don't see it too often on this platform yet.

In some cases, this is necessary due to lack of space and the want to keep charge air pressure drop across the core minimal (a good example would be the BMW F30 stock CAC, which consequently I am working on one at work because the thing won't stay cool on the track). If you can grow core face and instead of going deep, go taller, you will net better air temps in every condition but a drag race, and even then, you may do better than a 5" thick unit. A2A CAC's rely on ambient airflow and when the airflow can't get through the core, it's just a huge heat sink.

[gramicci101]

With air-to-water, can you use other liquids than water? Is there another liquid that would be better at absorbing the heat from the air?

[cdrazic93]

Originally Posted by gramicci101

With air-to-water, can you use other liquids than water? Is there another liquid that would be better at absorbing the heat from the air?

Specific heat values are your friend

http://koolance.com/cooling101-heat-transfer

[VerusEric]

Originally Posted by gramicci101

With air-to-water, can you use other liquids than water? Is there another liquid that would be better at absorbing the heat from the air?

My day job ran quite a number of fluids that supposedly increased heat transfer, and to this day we have not found a fluid (that you would run in a cooling system) with more ability to remove heat then water. Obviously you'll have to worry about freezing and boiling, but the closer to pure water you can get, the better.

Originally Posted by cdrazic93

Specific heat values are your friend

http://koolance.com/cooling101-heat-transfer

Never knew diamond was so high up there... I think I'll design a diamond heat exchanger next.

Side note: Although copper has a higher thermal conductivity, aluminum cores have surpassed copper cores a few decades ago. Copper cores are brazed with a lead based clad, which significantly reduces the heat transfer from tube to fin, and thus fin to air. Aluminum cores use another aluminum as clad, and their heat rejection is either equal or greater than a copper counterpart at a significant reduction in weight.

[ChuckD]

Hi Eric,

I've done some heat management experiments using extruded aluminum heatsinks in electronics. I found that anodizing the heatsinks helped remove more heat from the metal core PCB and that black anodizing worked best. Just curious to you have tried that in an automotive situation (I'm sure someone has).

Thanks
ChuckD

[VerusEric]

Originally Posted by ChuckD

Hi Eric,

I've done some heat management experiments using extruded aluminum heatsinks in electronics. I found that anodizing the heatsinks helped remove more heat from the metal core PCB and that black anodizing worked best. Just curious to you have tried that in an automotive situation (I'm sure someone has).

Thanks
ChuckD

Chuck,

What you're witnessing is a change in the emissivity of the part. I'm not good at descriptions, but my understanding of emissivity is the ability to absorb or radiate heat. Think about wearing a white shirt vs. a black shirt in the sun, you're hotter in the black shirt. Well, this same principle applies to releasing heat as well.

Black radiates more heat then that of plain aluminum, so painting it black will reject more heat. Here's where it gets tricky, this only happens when stationary! This is because the amount of heat lost due to radiation is minimal when compared to the amount of heat lost due to conduction.

My friends and I actually had a large discussion about this about 6 months back, and we came to this conclusion. I happened to stumble upon this during my search for answers. [ame="https://www.youtube.com/watch?v=f1QL9veQaNg"]Black Intercoolers Mythbusted - YouTube[/ame] while it's goofy, it really does show the differences well and they do a good job with the experiment.

Thanks,
Eric

[ChuckD]

Eric

Yeah didn't feel like talking alot about emissivity this morning like i said figured people have looked at it and all the situations I deal with is in a stagnet environment with very little air movement.

Take care
Chuck

[VerusEric]

Originally Posted by ChuckD

Eric

Yeah didn't feel like talking alot about emissivity this morning like i said figured people have looked at it and all the situations I deal with is in a stagnet environment with very little air movement.

Take care
Chuck

Chuck,

In your situation it makes a ton of sense to paint/anodize it a dark color then! I think emissivity is pretty neat, it isn't exactly intuitive. However, go stand in the sun with a black shirt and a white shirt, and you get a quick lesson with emissivity! Pretty cool... but I'm a nerd!

Updated main post with not-broken picture links.

Thanks,
Eric

[NyC Zn6]

Originally Posted by ATL BRZ

Great post Eric!

I'll add to this with a pic of the only AWIC turbo FA20 I've ever seen. Owned, built and tuned by Doug at DBW Motorsports here in Atlanta.

Off topic. What are those two blue cylinders on both sides of the strut brace?

[Tokay444]

Originally Posted by NyC Zn6

Off topic. What are those two blue cylinders on both sides of the strut brace?

External reservoirs for the front dampers.
On topic, I use an Aquamist HFS4 V3 methanol injection system in my built, big turbo Mazdaspeed3. It's a extremely reliable system, high quality system, build for direct injection applications which references injector duty cycle, boost, and maf, and has a built in failsafe. The nozzles are plumbed in at the throttle body with a spacer. On top of the knock prevention associated with methanol, I'm also using it for about 150whp worth of fuel, as the factory fuel system taps out around 350-380whp.

[Spartarus]

Originally Posted by VeloxEric

Black radiates more heat then that of plain aluminum, so painting it black will reject more heat. Here's where it gets tricky, this only happens when stationary! This is because the amount of heat lost due to radiation is minimal when compared to the amount of heat lost due to convection.

In the interest of scientific accuracy, advection or conduction would both be correct terms, depending on the scale of observation, but convection does not occur in or around an intercooler.

Some may see that as pedantic, but I think it's an important distinction. There are so many misconceptions out there. You could spend a day just explaining how intercooling (in the ideal sense) manages to be an isobaric process, and why pressure loss across an intercooler does not correlate to the temperature drop.

[VerusEric]

Originally Posted by Spartarus

In the interest of scientific accuracy, advection or conduction would both be correct terms, depending on the scale of observation, but convection does not occur in or around an intercooler.

Some may see that as pedantic, but I think it's an important distinction. There are so many misconceptions out there. You could spend a day just explaining how intercooling (in the ideal sense) manages to be an isobaric process, and why pressure loss across an intercooler does not correlate to the temperature drop.

You're correct, I meant conduction but clearly put convection. Thanks for catching that.

Thanks,
Eric