PDA

View Full Version : How Air "Works"


Whizbang
07-17-2008, 07:57 AM
From my experiences with the internet, it's apparent to me that there is a general misnomer about how air "works". Many websites, internet forums, etc tend to haze over the truth of how air acts in both the form of exhaust and induction. It is my intent to correct this and give you, the reader of this long thread, a dumbed down, yet explicit explanation of why air does what it does.

INTRODUCTION

First off, air is a fluid. The textbook definition (literally out of a text book) of a fluid is "a substance that is unable to resist the application of a shear force without undergoing a continuing deformation". (pg.2 of Introduction to Fluid Mechanics). So what does this mean? Its mean that in order for something to be labeled a fluid, it must not resist a force parallel to it. Think of a can of Pepsi sitting on your counter. Without the solid, tin surrounding the liquid contents, the cola would just be a puddle on the counter. Gases behave the exact same way.

So because air is a fluid, it has to follow some rules, the biggest being that air is a compressible fluid. The second most important is the "Conservation of Mass". The "Conservation of Mass" basically means that if you put mass into a system, you should get the equivalent amount of mass out. Granted there are some exceptions and a few of which stem into math that i frankly, don't understand. Now keeping this ideas in mind, I'm going to explain how air works in two of the most important parts of our engines operation, induction and exhaust.


AIR AND INDUCTION

Induction is obviously the intake track leading to the combustion chamber that gives the engine the fresh air it needs to operate. If you didn't already know this, it saddens me. And frankly, i have no idea why the hell your reading this. But getting back on track, induction can take on two methods. These are as follows:

1. Forced Induction - the absolute pressure is limited only by that of the device forcing the air.

2. Natural Induction - the absolute pressure is limited to that of the atmospheric pressure generally (14.7 psi at sea level)

Lets look at natural induction first, even if it comes 2nd on my list.

Natural Induction

Natural induction is probably the most widespread form of induction you will find in the automotive world, regardless of my want to see turbochargers and superchargers (or both) on everything, that is the sad reality. What happens is as your engine draws in air, a vacuum is created due to the pressure change resulting from the air entering the combustion chamber. This is where the name "induction" comes from. Air (or anything under pressure for that matter) will move from the area of higher pressure to lower pressure, meaning it will be drawn in to file up the "void". At wide Open Throttle (WOT) conditions, this vacuum is generally not able to be read on a gauge because the air entering the combustion chamber is easily replaced by air coming in from the outside, which at sea level is at an absolute pressure of about 14.7 psi or 101 kPa for those rocking the metric system. Why your standard everyday gauge will not show "14.7", but instead "0" is because it is reading something conveniently called "gauge pressure"; which is the reading of the pressure about that of the atmosphere. When the throttle plates are closed, the mass volume of air entering the engine becomes significantly limited and the vacuum will build up inside the intake manifold. This is become the mass volume entering the engine is greater than that passing through the throttle body so the vacuum is created.

Forced Induction

Start shoving air into the intake track at pressures above that of the atmosphere, you'll have yourself forced induction. In the WOT condition, the air entering is forced passed the throttle plates. When the air enters the combustion chamber, the pressure instead the chamber equalizes to that of the entering air. In a sense, the air being drawn into the chambers during natural induction is actually being forced in at a set pressure of 14.7 psi. Forced induction is merely the result of these pressure going above the atmospheric pressure. This can be achieved by several methods, most commonly the turbocharger or supercharger. Hell, "ram air" induction will generate a pressure above atmospheric, but the vehicle would have to being moving at a few hundred miles per hour to see any significant benefit. Other methods exist for creating small increases in pressure, but they are generally to complicated to explain here.

So why does forced induction created more power? That is for two reasons, one is that the cylinder pressure increases, and more importantly, more burnable oxygen is forced into the chamber. And now is where I'm going to bring in some math:

Density = (pressure) / (Temp x Gas Constant) [ D = P / RT ] and

Mass Flow = (Density x Velocity x Area) [ M = DVA ]

In order for more oxygen to be in the charge, you need more density. This is why cooler air yields more power, as cool air is denser than heated air. Why? Well look at the formula above for density. If the temperature increases, then the density will decrease. Whereas, if the pressure increases, the density will increase. So if we run about 15psi from a turbocharger, that is roughly double the amount of oxygen coming in IF THE TEMP REMAINED THE SAME. Compressing air will create heat. Turbochargers generally create more heat due to the proximity of the exhaust. This is why intercooling works so well, since it has a minimal loss in pressure but reduces the temperature thus giving us a improved density and the ability to run more pressure without melting holes in pisons, which is a whole other topic.

So what about the "mass flow" formula? Well that comes into play in our next sections: Induction (charge) piping and exhaust.

Whizbang
07-17-2008, 07:58 AM
INDUCTION (charge) PIPING

Welcome to my next section where i attempt to get you to understand what makes a "good" design and a "bad" design in regards to piping. Before we continue, i have to explain "mass flow" that was shown above. You need to understand what is happening in the formula to see why air behaves like it does in certain situations.

recall that: Mass Flow = (Density x Velocity x Area) [ M = DVA ]

So that means Mass Flow (M) is proportional to Density (D), Velocity (V) and the Area (A). Plainly, if you increase either D,V, or A, you will increase M. If one decreases, M will decrease. If they all change randomly, then it just depends on it multiplies out. Now also take into account the "Conservation of Mass" meaning that M1 going in will EQUAL M2 going out unless something is destroying the mass or its being diverted somewhere else. If you get this, then feel free to move on. If not, google.

There are some key ideas that you should follow when designing your charge pipes and for that matter, your entire induction track (head, charge pipes, throttle body, etc). These are as follows:

1.) Use the least number of bends at possible.

Why is this important? Air is a gas. Gases are kinetic, IE move around a lot on an atomic scale. This means they create more friction that say, water would. So when the air is moving through a bend, more friction is created than the straight section. This will slow the velocity of the air. So if the velocity does down and the density and area are the the same, then the mass flow would also have to go down. But the mass flow can not go down. So if the area can't change, then only the density of the air can.

D = P / RT

If you look again at the density formula you'll see that the pressure would have to decrease or the temperature increase, in order for the density to decrease. R is again a constant. Usually it is a combination of Pressure loss and Temp gain.



2.) Limit the change in diameter of the pipe as best as possible

The air will only flow as well as the smallest area in the pathway. So if you have a small throttle body and large charge pipes, it is no better and sometimes even worse than having smaller charge pipes. Even if you increase the throttle body size, the port runners of the intake manifold can present a problem. The reason why this happens is because of a few factors.

A.The Mass Flow rate is reduced. As stated, if the pressure and density are the same, but the area decreases, then the mass flow will decrease.

B. The areas where the diameter increases and decreases generates more friction and was i have previously noted, that isn't good. This will ALSO cause a decrease in mass flow rate due to the pressure loss resulting from the reduced velocity.


There are other factors that make up a "good" design such as overall length of the piping, but that is more for reasons of lag time, etc and is not the focus of this article.

Dr-Rotor
08-06-2008, 02:32 PM
Is that it? I thought you were getting to a point and some exampels. None the less very good info for all. Perhaps this would explain my rediculouse power figures. Take a look at my setup that includes water injection to drop the charge temp and increase density. Power was recorded on a dyno dynamics at a ramp rate of 110. GT3582R rated at 700hp for pistons in the largest 1.06 AR. At 20psi mine recorded 588rwhp....thats just insane compared to what others have achived. What do you think?

http://rotarycarclub.com/attachment.php?attachmentid=2110&d=1217880215

Whizbang
08-07-2008, 08:56 AM
this article is just about the theory of way, a lot of your setup im sure is thanks to tuning.

Dr-Rotor
08-07-2008, 02:00 PM
Yea probly. Thanks for the info anyway.

Phoenix7
08-07-2008, 02:04 PM
well, let's discuss it.

Dr-Rotor
08-07-2008, 04:42 PM
Id love to. What part of it all would you like to start off with?

Dr-Rotor
08-07-2008, 04:56 PM
Haa I know something about air flow we can discuss.

Ever wonder if there is a difference between having the IC right up front or further back in the engine bay with ducting channeling air into it. We have been discussing it at the work shop for the past few weeks.

We had a car on the hoist with a stock bumper and FMIC....we put compressed air through it & you can feel the air that dose not make it through is wasted & spills over the sides....to me this dose not seem all that efficiant. I think an IC set further back with ducting panels to stop the air from escaping is better. My reason for this is that the air is forced into a space that will contain it. Air can be compressed so this would mean that the air that dose not flow through will hit the side of the ducting then be turned & forced back through again by the continuouse pressure of on coming air as the car picks up speed. Therefor minimising the loss of pressure over the face of the IC & increasing its efficiancy to cool & also helping the radiator to recive more air.

Now many companys have turned to V-mount but one tuner (Fujita) still use FMIC. Must be a reason they favor FMIC setup. Could be as simple as its easier to setup & duct perhaps. I could be wrong, who knows.

What do you think?

Dr-Rotor
08-07-2008, 05:17 PM
Some examples

Phoenix7
08-07-2008, 08:31 PM
ok, so, based on the info and your setup I cannot imagine a more efficient hot side piping. Your cold side looks odd to me but then again I never liked the FD TB elbow but it still doesn't have too many bends. The FC one seems better to me but no matter what there is still a 90* bed to the UIM. You're still decreasing the the amount of density loss which just means a few more HP...

Then take into account that IC and Radiator seem to be well insulated from the engine bay heat. Plus your turbo blanket, plus the water injection, along with the vented hood. You're decreasing the amount of heat that the IC pipes would be exposed to and keeping the air cooler and denser at the manifold. Little by little you make more HP.

Ducting just rocks. I think it allows for the air to be forced through in an easier manner than without it. If you duct it right you can force cold air in then isolate the hot air from the IC, radiator and oil cooler right to the vents on the hood without exposing the IC pipes to it.

I've been contemplating heat-wrapping the IC pipes to keep the air cooler in the pipes.

Phoenix7
08-07-2008, 08:37 PM
I like YOUR hot side IC piping with this cold side piping:
http://www.rotarycarclub.com/attachment.php?attachmentid=2135&d=1218147402

The venting I'm talking about is like this; Mind you this is an NA 3 rotor but he's doing the same thing. I solating the hottest air from the intake.

This car is bad-ass:
http://i254.photobucket.com/albums/hh115/fd20na/DSC00598.jpg (http://www.rotarycarclub.com/showthread.php?t=3198)

Click to check out the rest of the car.

josh18_2k
08-07-2008, 11:11 PM
on the topic of less bends.. why are turbos never mounted vertically?
the downpipe would point down (where it has to go anyway) and the charge pipe would come directly off the turbo into the intercooler. 2 90 degree bends eliminated.
either mount the filter strait on top, like so many do already, or have a bend there. its on the suction side, so its probably of slightly less importance.

Whizbang
08-08-2008, 02:59 PM
probably has to do with the gravity powered nature of the oil return

Dr-Rotor
08-09-2008, 03:00 AM
probably has to do with the gravity powered nature of the oil return

Yea so true....there will always be something that hinders ideas like vertical mounting.

Speeking off the DP im not sure what you guys have but mine dose not seem to be 90deg off the back....its quite a stuttle bend.

Dr-Rotor
08-09-2008, 03:04 AM
Phoneix...I hear ya pal....Im going to change my setup to that top pic....keep my hotside as is & make the cold side staighter. Also I am going to half the nozel size of my WI on the elbow & add another nozel to the hot side before the IC. This will cool the air even more & also provide the Rad with cooler air.

Phoenix7
08-09-2008, 04:32 AM
are you going to put AIT sensors before and after the IC? Heat-wrapping the pipes?

Dr-Rotor
08-10-2008, 02:07 AM
Not sure If I will go to those extremes.

Phoenix7
08-11-2008, 12:46 AM
well, how else would you monitor IC efficiency?

I'd monitor temps after the IC to see if the IC is damaged, clogged, or whatever.

If the temps go up before the IC then you can isolate problems before the IC, after the IC and before the manifold. Seems like a little overkill but it'd help diagnose possible issues? :dunno:

I don't know which is more efficient though. FMIC or VMIC. VMIC wins based on number of bends but how well can you create vacuum at the hood area to draw the air out of the IC? Adding a fan would help but more draw on the electrical system.

With FMIC you know air will go though no matter the speed and, when mated to the dual oil coolers and vents, you can create ducting to the underside of the car, which you know for sure will create enough vacuum to pull the hot air out.

http://img.photobucket.com/albums/v164/madrx7racer/DSC00618.jpg

I just don't trust aftermarket hoods. All my conversations over the years end up with people claiming that the D-MAX hood (the hood I think seems to be the most efficient) is pure BS marketing because of low pressure areas of the hood and so forth.

They throw in some weird looking diagram that's supposed to "clearly show" the low pressure areas and call it a day. I don't see it but that's because I hit a knowledge wall.

I'd like to know of a way to test the aerodynamics of the rx7's with different hoods and front ends to figure out the best way to use the pressure around the car to draw hot air out.

My favorite image is this one because I see potential for great heat expulsion:
http://img.photobucket.com/albums/v164/madrx7racer/DSC00626.jpg



All this to keep the IC pipes as cool as possible to have the most dense air possible.

Dr-Rotor
08-11-2008, 02:50 AM
The reason why I said that I might not put temp sensors in is because the WI will drop the temp no matter what.

Besides that you have given me an idea...you spoke of a vent under the car that would suck the hot air out as the under side of the car has low pressure. This gives me the idea to put a reverse vent in the undertray just below the back of the radiator. This would draw the hot air out as it would create a sort of vacum from the high speed low pressure air underneath.

Im not sure of the hood your talking about but mine sure as hell blows air right out. I blocked the front scoop, attached some streemers to each vent & went for a blast around the track. Every streemer was near vertical from the force of the air passing through them....if it had been a low pressure area they would have been sucked in. Right?

Anyway...I will be redesiging my IC setup & I will be doing a few tests with a reverse scoop in the undertray in the next year or so.

As I said before with mounting the IC ferther inside to create an air ram effect, I recon the reverse scoop will aide in the removal & suction of more air through both IC & Rad.

memski
08-12-2008, 02:34 PM
Here's an old pic of my setup, have been thinking more recently of trying to vmount the ic, more to reduce pipe volume than aid rad temps. Greddy elbow was modified and the piping is 3" all round.

http://www.podiatrist.plus.com/mini-postmodfd.jpg

Dr-Rotor
08-12-2008, 02:53 PM
That looks fairly straight on both sides & the first pipe off the turbo is not as harsh a bend as some that I have seen. Nice setup.

Whats the reason for trying to reduce the pipe volume? Is that so the air wont have to go as far to get to the throttle? I dont think it matters all that much because the air travels so fast anyway....Your setup would have less pipe as it is now than mine dose & I dont have any issues with spool up or response.

Try just moving the IC into the bay right in front of the Rad, seperate the 2 with some firm rubber or foam all the way round to seal them off & keep the pipeing as is. It will be slightly shorter, less brakets than a V-mount & the pipe length would be pretty much the same as some V-mount setups. If you realy like V-mounts by all mean change it to what you want.

FC3S Murray
05-02-2009, 12:55 PM
After reading this thread, my IC set-up makes me very sad. So many bends!

I have been brain storming lately on how to fabricate some sort of ducting to force air through my IC, oil cooler and radiator.

My Air temps are already cool with the amount of heat insulation and expulsion I have fabricated/added to the exhaust and LIM and ducting to my TID. I get a reading in the TB elbow of 30 degrees C when the ambient air temp is 65-70 degrees F AFTER WOT tuning. Never really budges after that.

My water temps don't exceed 80 degrees C BUT my Oil temps always stabalize at 170 ish F cruising and I would love to get them at about 160.


My boost response is very good granted the lenth of piping and my 62-1 set-up BUT I could imagine it would be even better with some smart refinements.

Any ideas on ducting??

RETed
05-03-2009, 10:24 AM
Interesting thread...

I dunno why the original poster gave up after the initial flurry, but one of the big things that wasn't touched upon is resonance or Helmhotz[sp?] tuning?
Mazda did mess around with it a lot on the stock induction systems, and having only two combustion chambers, this effect does get magnafied.
This type of tuning is actually counter to the statement about air being a liquid; in this case, the air(flow) is a wave.

As for the replies about pipes, layout, etc...
Placement of the heat exchanger (the IC) is primary.
The placement and paths of the IC pipes are secondary - i.e. just make them fit.
FMIC are popular due to the fact that you're placing the heat exchanger in the most efficient position - where the ambient airflow is highest and most pressurize, i.e. - the front bumper facing forward.
I, personally, don't like all that V-mount crap, cause it's a compromise design - yes, you get short and more direct pipes, but ambient airflow is compromised, usually.
Not many of us measure this efficiency over the entier IC system - pressure in versus pressure out - and dialing most boost is the easy answer to overcome "poor" efficiency numbers...
Running more boost is usually easier than rerouting IC piping...in most cases...


-Ted