Saturday, April 6, 2013

Understanding Engines - Turbo vs NA... power vs response?! How does it all work.

This article has been spawn due to a comment on one of my youtube videos.
The video in question features a mighty modified Nissan Skyline GTR R34 VSpecII tuned to perfection by Mines. It's a true beast...probably the main reason why I don't like today's RenaultNissan GTR.
You seen the Nissan Nissan GTR...and called Skyline, was beautiful in all of its essence. That is why it managed the nickname GodZilla. Today GTR is no more than an vulgar Engineer wetdream built to impress Renault and trying to avoid Renault turning Nissan into "the factory that builds cheat wannabe suvs".
But this article is not about the GTR but rather the comment it self. After watching the video, the user added a comment in which he stated that a turbo car can be responsive. However he probably got so overwhelmed by the immense power of the car and it's extremely fast revving engine that he concluded that it was responsive.
So after a couple of minutes thinking about that I decided to explain the physics behind the induction system of an engine and the resulting consequences to power or response.
So why did the viewer got confused? You see, the car in question produces over 600bhp from 2600cc. One may say soooo, Place a big turbo in an ordinary engine and you have that same result... true..ish. The engine in question is the RB26DETT. And that stands for Racing Bred 2.6cc Dual over head camshaft Electronically manages fuel injection TwinTurbo....in plain English - The racing engine Nissan developed for the GT class cars, de-tuned for road use.
As a result, the engine is very high revving (because it is supra-square 86mm of bore against a short 73.7mm stroke). Now add 600bhp and extreme profile cams and electronics to such an engine, and you'll get huge power levels delivering in an explosive way from engine midrange up. In the video it's clear that the analogue needle in the car revmeter can't cope with such a fast revving engine.  
This however is not RESPONSE...it is power and torque(loads of it on a beautifully engineered, light engine). Torque is the work ability of an engine... for instance an engine with 1000nm of torque will pull a 3 ton car as easy as it pulls a 1 ton car. Power is the torque at high engine speed.
A diesel engine, is very good on Torque and very bad on Power so a VW PD130 1.9 will deliver 310nm of torque against 130bhp. A gasoline engine with VTEC would be very good on Power nd very bad on Torque... the F20C from the honda s2000 will produce almost 200nm of torque for 240bhp.

Analogy time:
Take for example 3 man. One with small muscles, one with normally developed muscles and one big guy with big muscles but also some fat. The muscle size will allow each to pick-up a certain amount of load (and this will vary in weight as bigger muscle pulls more weight), but bigger muscle will also consume more energy and have bigger movement inertia so the bigger they are, the slower they will move and more energy they will burn.
If you try a weight lifting contest, the bigger guy will win as the smaller guys will not even be able to pick up the weight. However, by breaking same the weight into 1kg packages and making the contest to move the entire volume from place 1 to place 2 10 meters apart, things will enter perspective.
In this second scenario, the less muscular guy will run fast back and forth without getting tired and will do this quickly, however lack of muscle will only allow for 1 kg package at a time.
The Big guy will probably pick up 4 packages at once, but will take longer to travel and will get tired in time.
The normal guy will be able to transport 2 packages at once, will move at an average speed and will experience some stress but will not get as tired as the bigger muscled one.

What I've just shown here in analogy is:
1 - A bike engine with very low torque and big horsepower, as the little guy with small muscles. Like a bike engine, it can't cope with big weight for lack of torque, but it will work very fast and transport the loads faster with ease (high RPM and as a consequence high horsepower).
2 - A normal gasoline engine with decent torque and horsepower figures, as the normal guy.
3 - A diesel engine with high torque but no horsepower, as the big guy without agility. It can pull a lot of weight with the same ease as a small weight, but the slow movement means it will take longer to work (lower RPM and as a consequence lower horsepower)

You might ask... ok, what about an analogy of a normal engine with a turbo. I would say that in that case, the normal guy injects quality steroids and testosterone, partially getting his muscles bigger, without growing bone structure and fat.

So how does this work? 
A normal engine is, in very basic terms, conceived by multiplying a simple assembly: a cylinder with a piston on it and some valves, the valves open and close allowing air through or not, the piston moves up and down changing the compression or reacting to change in compression inside the cylinder. This assembly creates a translation movement.
This assembly is connected to another one called the crankshaft. The connection is made from the moving part of the first assembly (the piston that travels up and down) to the periferic rotating end of the crankshaft, by something called "the connecting rod".
From this point onward, it's just a matter of how many of these are placed next to each other... a 4 cylinder engine will have 4 of these, etc etc etc.

In this first assembly (the cylinder and piston), the valves will open to allow air with gasoline in, while the piston travels down (this piston travel is called the stroke). Then the valves will close and the piston will travel up, compressing the air/fuel mixture (just how much of the volume of the cylinder gets compressed into the final result is what determines the compression ratio...11:1 means that the volume of air in the cylinder will be compressed into 1/11th of the original space), A sparkplug will create a high voltage spark, igniting the fuel and air under pressure. This result into a burst of energy liberation in the form of an explosion and its expanding gases...and that's what will force the piston down and produce power. After that, the exhaust valves will open, the piston will travel back up forcing the exhaust gases out...and the cycle starts all over again.

So if you think of it, the faster an engine can renew this cycle, the better it will work. So a good line of thought would be: ok... if the pistons and connecting rods, and the crankshaft are lighter, and the amount of length of movement the piston has to travel is minimal, that it will perform this cycle faster...and would be right. Bike engines, F1 engines, the F20C engine, the RB26DETT engine, and a lot of supra-square engines out there will rev very high because they have low stroke values and as a result of revving faster more air will be coming in or out of it. However, the shorter the stroke is, the less mechanical leverage exists to produce torque.
This is due to something called Linear Piston Speed. In basic terms, there are 2 things against the speed an engine works:
 1- The constant change of piston movement direction, creating huge internal structural stress in the piston itself. If not well engineered, it would develop cracks and disintegrate.
 2- Friction. The fact that a piston must generate compression and keep compression, means it will have to stress against the cylinder walls generating heat.
The best engines in the world, using special coatings and the best materials, are able to run at around 25m/s stable Linear Piston Speed and normally they can't top 28m/s, and even 28m/s is sustainable for some seconds only.
This Linear Piston Speed is created by the fast movement of the piston inside the cylinder...so every engine rotation, the piston will move 4x it's stroke distance.

So, for better understanding, if you increase stroke, you also increase the leverage on the crankshaft and generate more torque, BUT you will also be running higher LPS and, as a consequence, lower rotations.
This is the basic math about engines. Big torque means less rotation...less rotation means less high-end power (as power is a consequence of torque with rotation). Torque is the measure of rotating force, while horsepower is a measure of work per time unit...and since time unit, on something that keeps cycling is converted to rotations per time unit, we get RotationsPerMinute, or RPM into the figure of horsepower.

Check my next article about diesel engines vs gas engines for a better insight on the TorqueVsFuelVsLPS

Turbos compressors and stuff:
It should be clear by now, how an engine produces power and how it's volumetric capacity influences it.
Since an engine intakes air and fuel and then blows it into power, the more air and gas it burns, the more power it produces.
There are 2 ways you can increase fuel/air burn:
 1 - engine speed or rotation
 2 - engine size
This second one is the easiest one, but bigger engine means bigger everything and that includes weight and moving parts, that in turn add to inertia and frictions...not what you want in a moving parts machine, or a car. So the old comment "there is no replacement for displacement" is just as refined as pining nails to a wall with a sledge-hammer.

This is where forced induction comes to light. If a 2 liter engine displaces 2 liters of air on a rotation at normal atmospheric pressure, if you increase atmospheric pressure, than you force more air into the same 2 liters...for instance a 2 liter engine at 1 bar atmospheric pressure, will squeeze the same amount of air into it, as a 1 liter engine at 2 bar .
With different techniques, forced induction actually raises the atmospheric pressure on the intake, and by doing so it multiplies the engine capacity (it doesn't do so is a linear way, as air being compressed will heat and make the hole thing loose part of the volumetric gains obtained with compression).

Turbos, will use exhaust gas against a turbine that is connected to a compressor, that will compress air.
Compressors are chained to the engine crankshaft and use the engine mechanical force to compress air.
The centrifugal compressor is the compressor part of a turbo, linked to either an electric engine or the crankshaft, making it a kind of hybrid turbo....and some manufacturers are working on the electric turbo (a good idea that would eliminate the turbo lag problem).

So what is response? 
Well, response has to do with the time you engine takes to respond to your right foot solicitation and if it responds in the right amount of solicitation within a short time period. That has to do with a lot of stuff (including the engine design and lightness), but above all with fluid dynamics.
So what are the physics behind the response part?
The intake of an engine is controlled by a throttle assembly. It basic terms it is a tube with a choker that can be controlled from fully closed to fully opened. When you open the throttle, air rushes into the engine cylinders, when you close it it doesn't...simple.
The shorter and more open the path of air into the engine is, the better the response (the lighter the engine is, the better it will be to respond too). That is why the independent throttle body engines have great response. The entire throttle assembly is near the cylinders and since they each control one cylinder, the air flow is stable and  less turbulent.

On turbo engines however things get much more complicated. You see the air is being forced in by a rotating turbine that has inertia... if you just lift off and close the throttle, the entire intake will suffer a pressure spike that will force back on the compressor, trying to stall it, and this will degrade the compressor blades in time. That is why turbos normally use dump valves...its purpose is to remove this pressure and allow the turbine to maintain rotation (and health, for that matter)... turbos also have something called a Waste-gate that partially serves this purpose. The waste gate is used to control the rotation of the turbine (and so controlling the boost produced on the other end of the shaft), by opening or closing and controlling the amount of exhaust allowed to escape directly to the exhaust pipe and bypass the turbo... this does mean that in case of HELPING the turbine to slow down the waste-gate can fully open and facilitate the process.

On turbo engines, since there is no mechanical control of turbine rotation, your engine needs to produce exhaust gas to allow the turbine to spin (depending on the turbo type, size and construction it might have more or less inertia)...this produces a delay between your foot request and the power to become available... it will also come in an elastic way because the turbine inertia means a delayed spin-to-requested speed. Dump valves to minimize this effect by allowing the turbine to continue to spin, however this will not change the fact that on lift off, you produce less exhaust gas and as a result, there will always be a delay when you step back on the gas.
It's pure physics... Even in a PERFECT turbo that spins instantly to boost... the simple fact that the engine will also need to spin up to drive the turbo the even more boost means that the response from a 2.0 turbo running 1bar and full efficiency (acting as a 4.0 engine), will never have the same response as an Naturally Aspirate 4.0 engine...and this is VERY perceptible at low engine speeds, as the 4.0 has ALL the capacity there and the 2.0 turbo has no air to drive the turbo into 1bar of boost and so acts as a 2.0 with a bad exhaust as the turbo is actually working against the flow by obstructing on purpose.
And all this is, obviously, looking at things under a LAB perspective... because in real life, you have to consider that the important thing inside the engine is oxygen molecules... and that means that HOT air will have less oxygen molecules. And since air compressing warms air due to the force of compression and the fact that the turbo is very hot (due to the exhaust driving it), 1 bar of pressure is really NOT EQUAL to + 2 liters of engine.... sure you can add an inter-cooler, but this means MORE piping to fill with air until you have the right pressure, and no inter-cooler will be 100% efficient and cool the air down to the same temperature air as the outside temperature.

So turbo lovers that try to argument against this... just use the physics books you've ignored so far and admit the truth.

Can it be cured?
Sure! Are you rich?
Let me explain why: Most drivers don't have enough skill to steer the car using the accelerator...however the ones that have, prefer NA supra-square engines to turbo engines because of immediate response and correct proportion of response (no elastic effect) to drivers foot.
But what if you drive professionally is a championship where engines size and power are strictly controlled and capped? Like Rally for instance... the driver will steer the car with just about everything, including the throttle, but engines are small and compact, so they are turbo charged.
Well, they simply keep the turbine spinning independently of the throttle being opened or closed. They do this using an anti-turbo-lag system also called "Miss-firing system".
In simple terms, the system injects gas into the exhaust manifold and this will produce and explosion into the manifold.... past the already closed valves, making the turbine continue to spin..or spin even faster...that way, when you open the throttle back-on, no lag will occur as the pipes are already filled with air at the best possible boost pressure (controlled by a dump valve).
You can see this happening when you watch a rally car passing by and approaching a curve, on lift-off a series of "pooping" sound, together with some exhaust fireballs.
Why the "are you rich?" part? well, imagine the stress on the turbo by having explosions happen constantly all over your thing, light and gentle blades. Normally a turbo will not survive a season, being changed and serviced regularly...and unless you run on a racing budget, it will cost allot.

OK, so let's hybrid the turbo
humm not there yet.
A Hybrid turbo is a turbo that has a compressor or/and a turbine of different sizes. This is actually like balancing something for a specific usage.
You see, a turbo is not ONLY about pressure... it is also about flow. A 4 cylinder, 2.0 liter engine will consume 1liter of air every time it rotates (one cylinder pushes air and then compresses, one cylinder powers and exhausts, while other exhaust and pulls air, while other compresses and powers)... but at 6000RPM, the engine is actually consuming 6000 Liters per minute... and if that is supposed to be at 1 bar, than the turbo is actually having to pull 12000liters per minute of air.

If you make the compressor big enough, it will do that without problems, but this implies a bigger inertia to spin, and this will mean that the turbo will require a lot of air pulling the turbine... so this make the engine spiky, as the turbo will not produce boost until high revs and the, all the sudden will boost into it's operating speed.

To counter this, some people reduce the turbine, making it spin faster with less need of air... but this also means that the turbine will over-spin earlier and trigger the waste-gate to open to bypass the high volume... rendering the turbo unable to produce any more boost and while the rotation of the engine climbs, the pressure will drop.

So a hybrid turbo is actually a mild balance between when do you want the turbo to respond and up till what rotation vs engine volume do you want the turbo to "properly" feed the engine... they do allow you to tune the turbo to the engine to the usage, but they will not "solve the issue".

Wouldn't an electric turbo to the job? 
Well sort of!
See, the volumetrics that make a turbo produce big pressure, are worked around the size of the turbo and the speed it can turn VS the size of the engine it is feeding air to. A big engine at Big rpms consume a LOT of air and the turbo may not be able to feed the air at the desired pressure without "overspining"... that is the turbine spinning to a speed that causes a lot of cavitation on the air it is pulling and by so not being able to produce more flow at a specific pressure.
So an electric turbo would have to actually either drive a very big turbine, or spin very fast... ultimately melting the bearings.
There are however some works from VW that imply the next generation of cars will have an electric turbo always spinning to produce the low-end response and then a standard mechanical turbo will kick into gear when the engine is already producing enough exhaust!
That turbo engine I would love... let's see.

So sum up:
So turbo engines produce loads of power from a small package, however by having a turbo, response is sacrificed... that doesn't mean that, when producing power, it doesn't surge in huge numbers and make the revs climb fast...the thing is, this is not response, just power... and since the high rpms of a usable racing engine would drain the turbo out of his response range fast, turbos are more suited for torque figures than RPM.
Except of course the applications of bi-sequential-turbos like the Rx7, or the biturbos on inline 6 as te supra and the skyline... as an attempt to have the "best of both worlds".