This is a thread I started up for one of the Focus ST forums I belong to. Always tryin' to edjumicate.

I have written a few posts outlining a few things for those of you new-ish to the turbo scene. I'll keep the topics short and sweet, I'm hiding my math but I will show it upon request, and am trying to keep things general and somewhat light. My source of information are the classic turbocharging books (Turbochargers by Hugh Mackinnis, and Maximum Boost by Corky Bell), plus 15 years of farting around with old cars and slightly less old turbos, and a ton of auxillary reading on the subject. Another fine resource is the Borg Warner Matchbot (Performance Turbos | BorgWarner Turbo Systems) which is really fun to play with.

Again, this is intended to cover the basics and not get TOO crazy with technical or controversial information. I've tried my best not to make too many sweeping statements that can be misconstrued as absolute fact, and tried not to contradict myself or provide any information or opinion that is completely false. I've tried...

And having said that, here we go:

Heat versus Pressure:

Why does a turbocharger turn air into heat? It doesn’t, but it can seem like it. A turbo compresses air, and when you compress air by any means you also increase its temperature. A turbo is also not 100% efficient, so it has to work a bit harder to compress the air, and this extra work is absorbed by the air as heat. The more you compress the air, the higher the temperature. If you want to estimate the compressor discharge temp of your favourite turbo, all you need to know ahead of time is the atmospheric conditions you’re operating in and the manifold pressure you desire.

Common standard conditions are 14.7 psia (“a” for Absolute) and 68*F (I’m 43 and Canadian, so am somewhat fluent in both Imperial and Metric – I switch around from time to time but I’ll try to keep them all in American for your benefit). We typically need to think of our temperatures in absolute conditions, because most of the calculations require this, but we can retain the relative value of degrees F for the discussion. In our engines we’re running around 22psi of boost.

If a compressor were 100% efficient, compressing air from 14.7 psia ambient pressure to 36.7 psia manifold pressure, the calculation would be simple. (for this demonstration I’m keeping the maths on my spreadsheet) The compressor discharge temperature would be 223.9*F, for a temperature rise of 155.9*F. If a compressor is more realistically 75% efficient, the discharge temperature is now 275.9*F for a rise of 207.9*F.

Pretty hot, huh?

Sizing a turbo:

To make a quick pass at sizing a turbo, you need to assume a few more things. You need to estimate the rate at which your engine can swallow air, and this is done by picking a volumetric efficiency and calculating the airflow from that and from how fast the engine is spinning. Simply put, a 122 cubic inch engine draws in 61 cubic inches per revolution if it’s operating at 100% efficiency. When all the turbo books were written, a healthy VE was about 80-85%, but I believe that we’re running much closer to 100% nowadays. For this example, that’s exactly what I’ll assume. 61 cubic inches at 6000 RPM = 212 CFM. That part is simple.

You also need a compressor map – I will be using the GTX2867R map from ATPs website. Our earlier goal is 22 psi boost pressure, or a pressure ratio of 2.5:1, and if we look along the horizontal line that corresponds to a PR of 2.5, we see that the peak efficiency of this turbo is 78% for that pressure. We will use that in our calculation as well, and will tweak things to suit.

The first step is to estimate the discharge temperature for our pressure ratio, and the spreadsheet that I built tells me that this value is 268*F. To calculate the pre-turbo airflow we must apply a fairly simple calculation, which is (absolute ambient temp / absolute compressor discharge temp) x pressure ratio x post turbo airflow. Plugging in numbers, this is (527.67 R / 727.67 R) * 2.5 * 212 CFM = 383.4 CFM. Then we must convert cubic feet of air into a mass, each cubic foot weighing approximately 0.076 lbs at sea level. 383.4 x 0.076 = 29.14 lb, so our engine is consuming 29.14 lbs of uncompressed air every minute. Plotting that airflow on our map against the pressure ratio of 2.5, we find ourselves right around the 76% efficiency range. We would then use this to tweak our numbers, but the difference is going to be pretty much fractional: after tweaking the calcs with the new efficiency, the airflow is 28.93 lb/min.

Adding an intercooler:

The purpose of an intercooler is to reduce the temperature of the compressed air, because cooler air is denser and you can therefor pack more pounds of it into the small space inside the combustion chambers.

The intercooler absorbs heat from the boosted air, but the efficiency with which this heat is transferred from air to aluminium depends on the turbulence within the intercooler. A big thick cast aluminium pipe will absorb some heat and will therefore cool the boosted air slightly, but it’s very inefficient in gathering that heat because very little of the air actually comes in contact with the pipe, and it’s very inefficient in shedding that heat to the cooler atmosphere for the same reason. Passing that air through several smaller tubes increases the surface area and the chance for air/aluminium contact, and if those smaller tubes have turbulators inside of them there’s a much greater chance for contact. The more aluminium, the more mass, the more thermal capacity the intercooler has, but one must also understand that the bigger the intercooler is, the slower the air moves through it and the less actively the air transfers heat to the cooler.

Once the intercooler metal heats up, it starts transferring that heat to the atmosphere. The end tanks do some of this but the bulk of the shedding happens by the same principle as what’s happening inside the intercooler. Note that the intercooler only starts shedding heat once the metal temperature has risen above the ambient temperature, and the charge temperature must be hotter than this otherwise it wouldn’t be able to transfer heat to the intercooler. The intercooler should be looked at as a HEAT SINK primarily, with the external fins as an added means of shedding that heat.

Note that an intercooler does the majority of its heat shedding during the time that it’s NOT being used. It’s gained temperature after a heavy boosted run, and it sheds it in two ways: The first is the way you’d expect – by passing ambient air over the cooling fins – but there is another and that is by the engine drawing air through it. Without boost, that air through the IC core is darn near at ambient temp, and is doing a ton to cool that intercooler back down. This is another reason to have your intercooler sized properly – to keep the internal activity up so that hot air gives its heat to the core, and so that the core gives its heat back to the cold air passing through it.

All that activity within the intercooler comes at a price, and that price is pressure drop. An intercooler that is efficient at removing heat from the boosted charge will generally have a higher pressure drop, but that pressure drop will either reduce manifold pressure or increase compressor discharge pressure, which increases heat and puts more load on the whole system. It’s unavoidable, but the intercooler is an integral part of a turbo system that operates above about 6 psi boost.

Many of the intercoolers on the market have at least some data collected by their owners on this forum, so we have a bit more to go on. To keep things simple, and ideal, we’ll use the CP-E intercooler for its massive size and the ability to absorb practically 100% of the charge temp for at least one run. For the record, this is not my intercooler of choice, but for this example we can use that nice round figure of 100% (initial) efficiency. For reference, an intercooler that allows a 20*F increase in charge temps over ambient is still 90% efficient at the pressure/temperature we’re talking about.

So what happens when you reduce the boosted charge temperature from 268*F to 68*F? Assuming a 1 psi pressure drop across the intercooler (so a compressor discharge pressure of 23 psi, manifold pressure of 22), the engine can now ingest 40.49 lb/min of air. This is a pretty big increase over the 28.93 lb/min of a non-intercooled example, and plotting this on the GTX2867R map we see that the compressor is still well above 75% efficient.

Well sized, ATP!

Changing environments:

This is fine and good for a spring day by the Oceanside, but what happens when you change a few things?

If the ambient temperature is 100*F at sea level, the compressor discharge temperature rises to 325*F from 268*F back on that 68*F day. Assuming the intercooler can keep up at the same rate as before and cool the air back down to ambient, bulk flow only drops from 40.49 to 40.17 lb/min.

Where I live (3500 feet) the air pressure is about 90% of sea level. This changes the pressure ratio of the system, which raises temperatures and alters the flow so that now my pressure ratio is 2.67, my compressor discharge temp is 297.5*F, and my bulk flow is down to 38.51 lb/min.

Back at sea level again, on that hot day: If the intercooler performs at a more reasonable level than 100% - say, 85% once it’s good and hot – the post-intercooler temp is up to 129*F and the bulk flow is down to 38.19 lb/min.

Free ride? Hardly! (Misconceptions, etc.):

One of the more common misconceptions about a turbocharger is that it converts heat into rotational energy. This is just not true – it converts exhaust flow into rotational energy, and because heat makes exhaust gas more vigorous it makes sense to keep as much heat in the exhaust gas as possible before the turbine. Because air loses heat when pressure is released, the turbine discharge flow is much cooler than the turbine inlet, and it appears as though the turbine is "converting" that heat into motion. Nope, it's just the ideal gas law at work.

Another common misconception (based on the above) is that the turbo is just along for the ride. It’s not true either – a healthy street turbo has an EBR (Exhaust Backpressure Ratio) of about 2:1. This means that there’s 2 psi of pressure between the exhaust port and the turbine wheel for every 1 psi of boost pressure. On a poorly designed turbo (like the classic T3/T4 hybrid) it could have an EBR of 3:1 or greater. Choking the flow to this extent is terrible, but it aids boost response. A turbo with a large turbine section can have a very low EBR (even below 1:1 if designed properly) but this isn’t a practical turbo if you want to see any kind of response from it on the street. These are usually for continuous high speed use, or are spooled up with nitrous on the strip.

Edit - the bottom line result for EBR is what happens when the intake valves open. The pressure in the combustion chamber will be equal to or slightly greater than the pressure within the headifold and turbine housing, and the exhaust valves will still be open when the intake valves crack (overlap condition). If there's 20 psi in the intake manifold and we have an EBR of 2:1, that means there's still 40 psi in the combustion chamber. Obviously this difference in pressure will force exhaust OUT of the intake valves until the exhaust valves close and the piston moves down far enough to lower the pressure enough that air moves the right direction. **hint hint - this is where the crud on the intake valves comes from in a GDI engine** If you have a low EBR, lower than 1:1, the boosted air is actually at a higher pressure than the stuff still in the combsution chamber, and the engine will actually scavenge and will run at greater than 100% volumetric efficiency. Lots of power to be had here...

Yet another misconception is that a larger turbo produces less heat than a smaller turbo. Wrong: if the smaller turbo has an efficiency map better suited for the engine in question, the smaller turbo WILL produce less heat from the compressor section. Look at the compressor map of a wide range of turbos and have a look at their numbers – most of them peak between 76 and 79% efficient, but the efficiency islands vary greatly in shape, size, and what airflow/pressure ratio they are most efficient at. A huge turbo on a small engine WILL make more heat than a properly sized turbo, even if it is smaller.

Size is everything?

We see ads for intercoolers that are “600hp capable”, but what does this mean? Just like a turbo that is “600hp capable”, it really needs to be qualified.

A large capacity, high compression engine can be turbocharged by using a large turbo with a peak efficiency in the high flow, low boost zone. Because of the low boost, the volumetric flow of discharge air will be very high relative to the bulk flow, so all of the downstream components must be sized to handle this volumetric flow. The piping, the intercooler, and the throttle body must all be sized to allow many hundreds of CFM of airflow. Because of the low pressure ratio in the turbo, there won’t be a lot of heat generated in the compressed air, and the intercooler doesn’t need to absorb or shed very much heat. In this instance, a thick and squat core intercooler with a large cross-section for boosted airflow is beneficial.

A small, low compression engine can take a ton of airflow too, but it will have to be boosted a lot to fit into the small combustion chamber space. The compressor map will have to be very broad at higher pressure ratios, with the high efficiency island way up and to the right, but the surge line will be very important to watch out for. The turbo is taking a lot of air in but is squeezing it to high pressure and low volume, so the volumetric flow is much lower than the larger, low boost motor of equal horsepower, even though the bulk flow is the same. The intercooler piping and throttle body requirements will be much smaller because the air is crammed into such a small space, and the intercooler needs to be completely different. On the smaller engine the intercooler can sacrifice some cross section for the same reason as the piping and throttle body can be smaller, but instead of being thick and squat the intercooler should have as much frontal area as possible, because it’s dealing with a lot of heat and needs to get rid of it as easily as possible.

Head porting, header size, manifolding, etc – there are no rules that apply both to the larger and the smaller engines in this example, not that can be derived from horsepower alone. 500 hp is NOT 500 hp when a boosted engine is in the picture. A 500 hp 2 litre generally doesn’t need a throttle body or intake piping any bigger than a naturally aspirated 2 litre – in fact, probably smaller. A 500 hp 4 litre needs to move twice the volumetric flow as the 2 litre, so everything must be twice as large or twice as plentiful. The turbos and intercoolers would be completely different as well.

Awesome read Matt - thanks for putting this here. I've read many of the same books and managed to get less out of them due to lack of real experience. One thing that has always puzzled me (or at least the physics part of me) is that people seem to think that turbo's are producing more volume for their engines. This is also in all the advertising - company's 6 cylinder with a turbo is like an 8 cylinder from yesteryear. Your piping discussion points to the fact that the compressor is increasing the density of the air, not the volume. An engine can't gain more volume as the pistons still only sweep a given amount per revolution. So for a given engine, the sizing of things like IC pipes, manifolds etc have to be sized to the motor as you're only changing the density of the air, not the volume.