An Engineering Look at the Basics

Turbo Power Output

Fig. 1-1. Porsche TAG turbo Formula 1 engine

The power-enhancing capability of the turbocharger has been most thoroughly demonstrated by the Grand Prix racing cars of the 1977 to 1988 era of Formula 1.

The comparison of the power output of atop fuel dragster with that of a Formula 1 race car will establish the turbo's credentials. Current output of top fuel сars with 500 cid (cubic inch displacement) engines is suggested to be in the 6000-6000 bhp range, which calculates to around 10 bhp per rid. These numbers are not favorable compared to the 1300-1400 bhp, 90 cid qualifying engines of the 1981 Formula 1 cars. These outputs represent 14 to 15 bhp per cubic inch. That the champion is crowned is obvious even to casual observers. However, for the potential street turbo user, large questions remain. Answers to these questions will indicate why turbocharging is equally useful to the fast car enthusiast who relies on his car for daily commuting, to the racer, and even to the outer fringe of street power freaks.

Why does the turbo produce more power than other forms of enhancement?

The power output potential of any supercharger is measured by the amount of airflow the device creates after factoring out the power required to drive it and the extent to which it heats up the air while creating the flow and pressure. While it might appear that the turbo does not drain power from the engine, since the exhaust energy is lost anyway, this is far from correct. Heat and airflow drive the turbine. When air is forced through the turbine section of a turbo, reduced-flow areas inherent in the design create back pressure. This causes a small loss in power that would not occur if the turbo had a power source other than the engine into which it is pumping. The power loss increases as the size of the turbo decreases, because the decreased size creates greater back pressure. Conversely, larger turbos create much less back pressure and therefore less power loss. The power loss inherent in a turbocharged engine is substantially less than the loss incurred by driving a supercharger with a belt or by some other means.

That an air pump always heats air it compresses is a thermodynamic fact with which we are stuck. Different kinds of air pumps heat air different amounts for the same flow rates and pressure ratios. These differences are due largely to the different efficiencies of various types of pumps. The classic Roots-type supercharger usually rates efficiencies of about 50% whereas the turbo runs efficiencies in the mid-70s. The higher the efficiency, the less the heating effect on the air. Efficiency is of paramount importance to the real power enthusiast, since heat in the intake charge is the enemy of performance. The density of an intake charge is less as the temperature rises; thus, an engine actually consumes less air at the higher temperature, even If the pressures are the same. A second problem is that higher temperatures promote detonation of the air/fuel mixture. Engines cannot withstand the thermal and pressure shocks of detonation for more than very short periods.

Withstanding Power Output

How can the engine structure withstand these huge power outputs?

To understand why the structure of an engine is not seriously affected by the increased power output permitted, within logical limits, by the turbo, it is necessary to look at the basic loads in an engine while it is in operation. Two basic loads are relevant to engine structure: inertia! load and power load, Inertial loads can be tensile (produced by pulling) or compressive (produced by pushing). Power loads can only be compressive. They must be understood both individually and in their interaction. This is necessary for a clear view of why the turbo does not send the crank south.

Inertial Load.

An inertial load results from an object's resistance to motion. To examine the inertial loads, it is convenient to divide a cylinder assembly into an upper half and a lower half. Imagine the two halves separated by an imaginary line called the center stroke.

Fig. 1-2, The relationship of engine loads to engine components has three significant piston/

crankshaft positions.

Fig. 1-3. Connectingrod inertial loads. Inertial loads applied to the connecting rod are closely approximated by the sine wave curve of load versus crank angle.

The piston always accelerates toward the center stroke, even when traveling away from the center stroke. In other words, when the piston is above the center stroke, it will always be accelerating downward. When it is below the center stroke, even at bottom dead center, it will be accelerating upward. Acceleration is greatest at top dead center and bottom dead center, when the piston is actually sitting still. When acceleration is greatest, the loads will be highest. Similarly, acceleration is zero and velocity is greatest as the piston passes the tenter stroke.

The size of the loads generated by these motions is proportional to the rpm of the engine squared. For example, if engine speed is increased threefold, the inertial load will be nine times as great. The action of the piston's being pulled (forced to accelerate) to a stop at top dead center and then pulled down the bore toward the center stroke will put a tensile inertial load into the con-rod/piston assembly. Similarly, as the piston is pushed to a stop at bottom dead center and then pushed back up the bore toward the center stroke, the inertial load will be compressive. Thus, any time the piston is above the center stroke the inertial load will be tensile, and below the center stroke, it will be compressive. The largest tensile load induced into a con rod is at top dead center on the exhaust stroke (because at top dead center on the compression stroke, the gas is already burning and creating combustion pressure to oppose the inertial load). The largest compressive load is generally at bottom dead center after either the intake or power stroke.

These inertial loads are huge. A large-displacement engine running 7000 rpm can develop con-rod inertial loads greater than 4000 pounds. (That's like m Cadillac sitting on your rod bearing.)

Power load.

A power load results from the pressure of 1 he burning gases applied to the piston. An example would be the compressive load put into a connecting rod as the burning gases force the piston down the bore of the cylinder.

Fig. 1-4. Burning gas induces a compressive load in the connecting rod.

Pressure created by the expansion of the burning gases applies a force to the top of the piston equal to the area of the bore times the chamber pressure. For example, a cylinder with a bore area of 10 square inches (3.569-inch bore) with 800 psi of pressure would be subjected to a compressive power load of 8000 pounds.

Fig. 1-5. Combined power and inertial loads. Note that power and inertial loads generally subtract from one another

The peculiar relationship of the inertial and power loads is of most interest in the upper half of the power stroke. Here we have the odd circumstance that the two loads acting on the con rod are doing so in different directions. Remember that an inertial load is tensile above the center stroke, while a power load is compressive in all cases. Power load peaks at the torque peak and fades a little as rpm increases but is generally greater than the inertial load. The difference between these two loads is the real load in the con rod (fig. 1-5).

Clearly, the inertial load offsets some of the power load. It is further apparent, as indicated above, that on the exhaust stroke, when the con-rod/piston reaches top dead center and is unopposed by combustion pressure (because both valves are open), the highest tensile load is reached. This load is the most damaging of all, because tensile loads induce fatigue failure, whereas compressive loads do not. For this reason, when a designer sits down to do the stress analysis on the con rod and con-rod bolts, the top dead center and bottom dead center inertial loads are virtually the only ones he is interested in knowing.

Fig. 1-6. Torque input into the crankshaft versus crank angle at approximately two atmospheres of pressure. Note that for the turbo engine, maximum pressure occurs at about 20 º ATDC, yet only about 20% of the mixture will have burned. Even with high boost pressures, the small amount burned will not. result in large maximum pressure changes. As the burn nears completion, the greater mixture density can raise the pressure three- to fourfold at crank angles near 90º such that torque input to the crank at that position can be twice as great.

The thought of doubling an engine's torque (doubling the power at the same rpm) easily gives one the idea that the power load will double. Thank goodness this is not true. To show how power can double without the combustion chamber pressure's doubling is much easier done graphically. Any significant design load changes would be based on peak pressure in the chambers, and it can be seen in figure 1-6 that with twice the mixture in the chamber, peak pressure is up only about 20%. There are two reasons for this disparity.

Long-Term Durability

First, power is a function of the average pressure over the entire stroke of the piston, not just peak pressure. The average pressure can be dramatically increased due to the much higher relative pressures near the middle or end of the stroke, while the peak does not gain significantly.

Second, peak pressure is generally reached after only 18-20% of the mixture has burned. If the mixture quantity is doubled, 18-20% of it, too, will have burned by the time peak pressure is reached. Since the total chamber pressure consists of the compression pressure plus the burning gas pressure, it is impossible to double the total pressure by doubling only one of its constituents. (Clearly, mother nature has a soft spot in her heart for con rods and con-rod bearings.)

A careful study of figure 1-6 will show that at crank angles nearing 90 º, chamber pressure is perhaps three to four times as great when operating under boost. This is, however, noticeably less than peak pressure. Therefore, it does not create a damaging load. The part of the power stroke near 90 º is where the real turbo engine power increases take place. If a physics type looks at the graph, he will tell you that the area under the respective curves represents the power. Thus, the difference in the two areas represents power gain due to the turbocharger. It certainly is a neat deal that we can double the power but not the load!

The preceding discussion establishes that the increased combustion chamber pressure due to a turbo, and thus the power load, will have only a moderate adverse effect on the structure of the engine.

** RULE: Power toads generally won't tickle the engine structure's tummy.

Long-term durability: Is it there, and how is it attained? The answer to 'Is it there?' is relatively easy to show by citing a few examples. Someone at Porsche once stated that a racing mile was about equivalent in wear and tear to 1000 street miles. Porsche's turbocharged race cars have won so many twenty-four-hour endurance races that only a racing historian can keep up with the number, These cars generally cover over three thousand miles in such races. A street car with three million miles on it may seem to be stretching the point, but the idea doesn't fail to impress. To stand along the banking at Daytona when a Porsche 962 turbo comes whistling by in excess of 200 mph can easily leave one aghast to think that these things are going to do this for twenty-four hours. The violence and speed can give the initial impression that nobody will finish this race. Yet, chances are a turbocharged racer will Lake the checker first. This book is primarily about street turbocharging, not race cars, but the problems are the same, even if different in magnitude. Street cars, by comparison, are a piece of cake, Chrysler even put a 70,000-mile warranty on some of its turbo cars.

How durability is attained is not quite so easy to answer as is the question of whether it exists. In a broad sense, durability boils down to the control of heat in the engine/turbo system, Each aspect of the system in which heat plays a part is a candidate for the Achilles' heel. Far long-term durability, each of these factors must be optimised. They include turbo compressor efficiency, intercooling, control of end-gas temperatures, turbine bearing temperatures, and many others, and will be discussed in the following chapters. We should call the answer to the entire heat problem 'thermal management,' In reading this book, it will prove useful to keep uppermost in mind that virtually the entire success of a turbo/engine system lies in thermal management.

Power Gain

Where does the power gain come from? What is the equation for the power of any given engine, and how does the turbo influence that equation? (Don't let equations scare you offthese are both neat and easy.)

It is revealing to examine the simple equation that relates power to the parameters describing the internal combustion engine.

Power = PxLxAxN

P is brake mean effective pressure, or bmep. An easy way to imagine bmep is as an average pressure pushing the piston down the bore.

L is the length of the stroke. This tells you how far the pressure is going to push the piston.

A is the area of the bore. This is, of course, the area the pressure has to work on.

N is the number of putts the engine makes in one minute. This represents how fast the engine is running and how many cylinders it has.

N = number of cylinders x rpm/2

(For a 4-stroke engine, the rpm is divided by 2 because each cylinder fires only on alternate revolutions.)

Now, there are several interesting relationships here! For example, take the P and multiply by the A and you have a pressure times an area, which is nothing more complicated than the average force pushing down on the piston. Now multiply the PA (force) by the length of the stroke, L (distance), and you have a number that represents the torque output of the cylinder Then take this figure and multiply by the N (how fast the job is getting done), and the result is

Power, the thing we are really after.

Please note that this means

Power = torque x rpm

Fig. 1-7. 'PLAN' is the key to the source of all power output.

Since the whole purpose of this exercise is to get more power, let's examine what this PLAN gives us to work with.

First, let's check out what working with the N can yield. There are two ways to get more putts per minute; add more cylinders or rev the engine higher That leaves little to work with, as the whole field of endeavor known as blueprinting: is almost solely for the purpose of allowing higher rpm with some degree of safety. Consider that those nasty inertial loads go up with the square of the rpm increase. That means that at 7200 rpm, the inertial toad will be 144% greater than at 6000 rpm. Wear and tear lies up there. Ultimately, it is neither cheap, pleasant, nor durable long-term to increase power output by increasing the N, Since we cannot, for practical reasons, increase power significantly with N, the only remaining choice is to increase torque by doing something with the PLA,

So we must go back and look at the PLA a bit more. We can change the A. Bored, it's called, but how much does it help? Change A by an eighth of an inch and maybe you'll gain 10%, Not worth the trouble. We can also change L. Stroked. Another 10%, maybe. Obviously, then, if we're pursuing real power, the A and the L don't hold much promise. Changing P becomes our only hope.

How to successfully change P is the crux of this hook. P can be changed by factors of 1.2, 1.5, 2, 3, 4, 5. .. The real potential is not known, since engineer types push the envelope every year. The Grand Prix racing cars of the '87 season took turbo development to the highest levels ever achieved, with power outputs of nearly 15 bhp per cubic inch. Suffice it to say, then, that doubling the power of a street engine, while not exactly child's play, is well within our reasonable expectations.

It is essential here to make clear the fact that we are dramatically increasing power without changing rpm. Therefore, it is torque (PLA) that we are really changing.

** RULE: Turbos make torque, and torque makes fun.

Drivability Limitations

What are the drivability limitations of a turbo charged engine?

The nice drivability of most cars today is something wo have grown to expect under all conditions. Get in, turn on, drive off smoothly, Nothing else is acceptable anymoreexactly as the situation should be.

Fig. 1-8, One typical example of the difference in torque curves for a turbo-charged and an atmospheric engine.

It is often perceived that real power and nice drivability are not compatible in the same automobile. This is frequently true in atmospheric engines but decidedly not true in turbocharged engines.

Consider the facets of an engine that create drivability: conservative camshaft profiles, small intake ports, fuel system flexibility and calibration. A proper turbo engine has a short-duration, low-overlap cam, generally referred to as an 'economy cam.' Port sizes are usually small, to create good cylinder filling at low speeds and to let the turbo pack it in when high pressure is wanted. Fuel system calibration must always be spot on, at least with electronic fuel injection. Obviously, then, the factors creating nice drivability are present in turbocharged cars. The fact that a turbo is available to push more air in when desired has no influence on 'Get in, turn on, drive off smoothly.'

Two factors affecting drivability do come into play when the turbo is in use: boost threshold and lag. These do not significantly degrade atmo engine performance, since the cam, compression, ignition timing, and fuel mixture remain virtually the same. If you stick a rack under the throttle and go for a trip around the block, you just can't tell the difference.

Boost threshold.

Boost threshold, defined in the glossary, is essentially the lowest engine rpm at which the turbo will produce boost pressure when full throttle is applied. Below that rpm, the turbo simply is not supplied with enough exhaust gas energy to spin fast enough to produce above-atmospheric pressures in the intake manifold (see fig, 1-8). Up to the boost threshold, the engine's torque curve remains virtually the same as that of an atmospheric engine. To accelerate through this range at full throttle, the driver would feel a surge in power as the torque curve takes an upward swing at the boost threshold. If full throttle is not used, the turbo makes no contribution to the torque curve, and acceleration continues the same as with a non-turbo engine.

The nonboosted torque curve can sometimes be compromised by an unreasonable reduction in the compression ratio (displacement volume plus clearance volume, divided by clearance volume), causing a soggy feel at low speeds when not under boost. It is here that some of the automotive manufacturers have made a serious engineering (or economic) error, by not fitting suitable intercoolng systems to remove enough heat from the intake charge. This would permit the use of higher compression ratios, retaining that sweet, low-speed response of an engine with an adequate compression ratio. If you are shopping for a turbo car, have some fun and ask the salesperson to tell you the efficiency of the intercooler. That is, of course, after you ask if it has one. It is certainly reasonable to assume that low-speed drivability is superior if the vehicle is fitted with an intercooler and the compression ratio is kept over 8 to 1.

Judging the merit of a turbo system solely on a low boost threshold is a serious error. It would be tough to argue that boost at low rpm is a bad thing, but it is easy to argue that boost at low speeds achieved by small turbos is a potential problem, due to higher exhaust gas back pressure, A well-designed system that has had great attention paid to all its parameters will display good low-speed boost as one of its features.

Small turbochargers frequently produce an annoying response when the throttle is applied in small increments. This distinctly affects drivability, in that a small motion of the throttle will produce a quick and usually unwanted small surge of boost that upsets the smoothness of the car. To some extent this causes a passenger to think the driver inept. This small surge frequently gives the driver the impression the car will really fly when full throttle is finally reached. Instead, he realizes sadly that the small surge was all the surge the little wimp could make. OEMs do this to us hoping we will think the car has instant response and gobs of low-end torque. They have generally overlooked the fact that it was raw power we really were after. This OEM phenomenon has left many journalists, writers, would-be last drivers, and other social outcasts wondering ' Where's the beef?'

** RULE: In general, OEM turbo applications are a long way from what enthusiasts and engineers would pronounce fast, fun, and first class. Let us call OEM turbos conservative.

Fig. 1-9. Comparison graph of the torque-increasing capability of small, medium, and large turbos applied to the same engine

Lag.

Seldom are turbos discussed without the mention of lag. Equally seldom, it seems, are discussion participants really talking about lag. Usually they are talking about boost threshold. Please read the definitions of lag, boost threshold, and throttle response in the glossary In the clay-to-day use of a turbo, sure, lag essentially means how long you have to wait to get boost after you nail the throttle. By definition, then, it is a bad thing. But lag has nothing to do with throttle response. Throttle response remains the same, turbo or no turbo.

Consider that if you did not have a turbo, the brief lag would be followed by no boost at all. Reasonable to say, then, that lag would extend from the point at which you apply throttle all the way to the redline. What fun that would be! The situation boils down to some tolerance for lag with a huge torque increase as opposed to no tolerance for lag accompanied by no torque increase.

Lag decreases as rpm rises. While lag can be as much as a second or more at low rpm, the delay in boost rise virtually disappears at revs of about 4000 or greater, For example, in a properly configured turbo system, boost rise will follow the position of your foot any time the revs are above 4000 rpm. Response here is virtually instantaneous.

** RULE: If you have no lag, you have no turbo. You also have no huge torque increase to look forward to.

Fig. 1-10. Respective lag times of small, medium, and large turbos

The shape of the torque curve of a turbo engine is different enough from that of an atmo engine that drivability of a turbo is only slightly affected. Torque peaks are virtually always at lower rpm on turbo engines. Chart all the published data and no other conclusion is possible. The more performance-oriented the atmo engine, the greater the difference. The net effect on the driver is that he or she need not rev the turbo engine as much to move rapidly. This is quite contrary to popular opinion but is indeed fact,

Hot and cold starting are frequently perceived to be problems of high-performance engines. To some extent this is true in carbureted turbo systems, but these arc few and far between. Fuel injection systems depend solely on various engine-temperature sensors for all cold- and hot-start air/fuel mixtures and are completely automatic. Cold starting is particularly a problem for engines with lower compression ratios. If an engine has a problem in this respect without a turbo, it will likely have the same problem with a turbo, since the turbo does not influence these temperatures or the electronics. Either way, the difficulty is not related to the turbo.

Cruising.

The turbo is out of the picture in all cruise conditions except those that must have boost pressure to achieve a particular speed. Consider that a given vehicle may have a top speed of say, 130 mph, no turbo. Now add a turbo. It is reasonable to say that the vehicle will still reach approximately 130 without the need for additional power; hence, no boost is required. For all practical purposes, even the wildest imaginable cruise speeds are unlikely to require any boost pressure to sustain.

The idea that a superpowerful, maximum-effort turbo car would be fun to drive at full throttle but be a bit of a cantankerous beast at low speeds is not unreasonable on the surface. This idea does not, however, hold up under closer scrutiny. To create an effective high-pressure turbo car, one need only do more of the same required to produce the turbo car in the first place: reject more heat, increase fuel flows, raise the octane, and be certain the structure of the engine is adequate. The factors that are the basis of good low-speed behavior conservative cam profiles, small intake ports, and fuel system calibrationare unchanged by higher boost pressures. All other things remaining equal merely turning the screw on the boost knob does not alter drivability. It is most unreasonable to claim that a 500 bhp street turbo carwhich, given Full throttle in second gear, has the ability to create tire marks in directions perhaps other than those intendedhas a drivability problem.

And Furthermore . . .

How much power can I expect from a turbocharged engine?

With currently available fuels, 7 to 12 psi boost is a practical upper limit for stock engines (at sea-level elevation). Intercooling permits this when elaborately and properly done. Certainly not all turbo kits or systems will perform the same, due to widely varying engineering efforts on the above items. Special preparation of engines specifically for turbo applications can frequently permit boost pressures of 15 to 20 psi. To claim, calculate, or estimate a specific figure for power from a turbo engine can be precarious indeed.

Of known dyno runs on piston engines with a variety of turbo systems, the lowest output we have achieved is .052 bhp/cid psi and the highest is .077 bhp/cid psi. The variance is due to the engines1 basic designs. To guess at the output of your own engine, choose a logical boost level and multiply each of the two values by both displacement in cubic inches and boost pressure plus 14.7.

Example: A 350 cid engine with 10 psi boost Lower value = 0,052 x 350 x (10 + 14.7) = 449 bhp Higher value = 0.077 x 350 x (10+ 14.7)= 666 bhp

Does the rated boost of a kit have any merit?

It does if, and only if, the conditions required to achieve that boost are defined and accurate. For example:

Was the gasoline used commercially available pump gas?

Were octane boosters used?

Was detonation present?

What was intake air temperature?

Is this the same boost-pressure setting the buyer will receive?

Considering the large power increases offered by the turbocharger, what keeps the entire structure of the engine from going south?

A proper answer to this question is a complete analysis of the inertial, power, and thermal loads before and after turbo installation. If this is performed, the conclusion will be two interesting bits of information:

The inertial loads in a modern internal combustion street engine are so large at maximum power that the power component of the total load is of little significance. For example, to induce as much power load into a con-rod bearing as the bearing already sees from inertial loads, the actual power of the engine would need to increase approximately 50%.

The thermal load in an engine not originally designed for a turbocharger will cause an increase in component and cooling-system tempera Lures when operating under boost, The components and cooling system can handle the temperature increase for a limited period. This is true for Buicks, Porsches, Saabs, Volvos, Nissans, etc. It is also true for all aftermarket turbo kits. The time limit is subject to many judgments and conditions. Experience has led me to believe that the time limit at full boost is on the order of 20 to 25 seconds. This is an operational restriction but not one of any consequence. Consider, for example: How fast will you be traveling if you hold full throttle in a 325 bhp Toyota Supra for twenty seconds? The answer is obviously an impractically high rate of speed.

When should the turbo start producing boost?

In most cases, there are trade-offs between a low boost threshold and maximum power. To bias the turbo size toward low-speed boost capability generally means operating the turbo in a very inefficient flow range at the engine's top end. Conversely, if maximum power is to be achieved, the turbo will usually be so large that no boost will be available until the last half of the rev range. Compromise is obviously necessary. I believe the reasonable balance between low-speed response and top-end power is to size the turbo such that it begins producing boost at about 30% of the redline rpm.

How will the turbocharger affect drivability?

Drivability of fuel-injected engines will remain the same. Drivability of blow-through carbureted engines will remain virtually the same. The starting of carbureted engines will be degraded slightly. Please note that draw-through units will virtually always degrade drivability and starting somewhat, with cold weather proving the Achilles' heel of a draw-through system.

Will the turbocharger hurt my mileage?

Yes. The turbo, when installed as an aftermarket item on a spark-ignition engine, is not an economizer and cannot be construed as such. There is no engineering basis for making such claims. If you are led into purchasing a turbo under the premise of improving your fuel mileage, be sure to get a written guarantee. When not operating under boost, a turbocharger is a small system restriction. This restriction causes a small loss in volumetric efficiency. Volumetric efficiency and fuel economy are definitely tied together If your driving habits are about the same as most, your mileage will drop about 10% city and 5% highway No miracles here.

Will the turbocharger affect engine wear and maintenance?

Certainly the turbo will affect engine wear. Do you really expect to add power and not increase wear? No miracles here either. If you drive vigorously but with some respect for the equipment, you can expect about 90% of normal engine life.

Will the transmission and drivetrain be adversely affected?

Very unlikely. Consider that the drivetrain endures more torque in first gear from the stock engine than almost any turbo can produce in second gear, Occasionally a clutch comes along that won't do the extra duty. Most clutch problems are going to crop up when shifting habits are less than acceptable. Not to worry.

What does it feet like to drive a properly set up turbo car?

A turbo can justifiably be called a torque multiplier: the more boost, the more torque. This situation is analogous to gear ratios. For example, a third gear with a tranny ratio of 1.4 will develop 40% more torque at the rear wheels than a fourth-gear ratio of 1.0. A boost pressure of 6 psi will increase torque by about 40% (using an intercooler). Thus you can see that 6 psi boost will produce fourth-gear acceleration virtually equal to a stock automobile's third-gear capability. Imagine what the proper turbo oar will do in second gear! Another reasonable comparison is that a proper turbo car operating at 10 psi boost will do 0-60 in two-thirds the original time; i.e., 6 seconds versus 9 seconds.