Air Induction, OEM Air-intake system and the Exdos Mod
Background
In the MC’s normally aspirated engine, when the engine rotates, the pistons in each of the 6 cylinders suck air and petrol into the cylinders on the inlet stroke, which pulls air from outside the car though the two apertures into the air-intake system, where the air is filtered and then measured by the Mass Air Flow (MAF) sensor so that the appropriate quantity of fuel can be injected into each cylinder for the combustion phase of the cycle for optimum engine performance. For best engine performance, the greater quantity of air that can enter each cylinder on the inlet stroke, the more fuel that can be added. Since the cylinder walls and pistons are made of metal, the maximum internal volume of each cylinder is fixed to a volume of 541 c.c.s, (ignoring thermal expansion/contraction changes), but since air density can change, the greater the density of air in the “charge” of each cylinder, the greater the number of oxygen molecules that can be packed into each cylinder. Therefore the greater density of air that enters the engine cylinders the greater amount of fuel injected into the enclosed space of 541 c.c.s, and the more fuel, “the bigger the bang” which means more power produced by the engine.
Since the pistons descending to the bottom of each cylinder cause the driving force for air entering the engine in a normally aspirated engine, then this causes negative air pressure inside each cylinder that in turn sucks air into the air-intake system. If the car is stationary and the pressure of air around the car is at an “ambient” pressure of, say, 1 bar (14.7p.s.i.) then the pressure inside each aspirating cylinder, when the engine is “ticking over”, will be at, say, 0.7 bar and all the air contained within the enclosed air intake system will also be at less than 1 bar right up to the mouth of the two apertures into the air-intake system. If the accelerator is pressed, the engine will increase its rotational speed, which will increase the air consumption and consequently cause a further decrease in air pressure within each cylinder to, say, 0.6 bar, and within the entire air-intake system. Obviously air at a pressure of 0.6 bar contains only 60% of the oxygen content of air at ambient pressure of 1 bar. Likewise, since density of air is proportional to its temperature (as per Boyle’s Law), cold air is denser than warm air, therefore, in a normally aspirated engine, in order to get the greatest charge of air into each cylinder, it is imperative that all air entering the air-intake system is as cold as possible. In the case of the MC the two apertures into the air-intake system are located at the front of the car facing into oncoming air which is not pre-heated by the engine, so all air entering the air-intake system will be as close to ambient temperature as is possible.
If you carefully examine the OEM air intake system on the M Coupe you will see that there is one “snorkel” aperture in the top of the left air duct in the front bumper and the other aperture lies immediately behind the left headlamp unit. The two air–intake apertures are made in black ABS plastic mouldings and once the air has passed into the air-intake system, the system is almost completely sealed right through the entire system right to the exhaust silencer tailpipes.
When the M Coupe’s engine is running at 7500rpm, the 3,246cc engine consumes 12.17 cubic metres of air per minute. All this air must first pass through the 2 apertures into the air-intake system, which fuse at a “Y” junction before the air enters the air-filter box through a single oval pipe. Obviously, when the car is standing still the engine must suck all this air into the air-intake system by the rotation of the engine alone. However, when the car is moving in a forward direction, the movement of the car, relative to still air, causes an increase in static pressure at the front of the car and this can easily be demonstrated by sticking your hand out of the car window. The faster the car travels the greater the pressure of air against the front of the car (and your hand). The increase in pressure is proportional to the square of the forward velocity therefore the increase in pressure at, say,100mph is 4 times greater than at 50mph. (100x100 = 10,000, 50x50= 2,500. 10,000/2,500=4). The pressure of air against a moving object, or alternatively, the pressure of moving air against a stationary object is known as “ram pressure”. Since the MC air-intake apertures are positioned at the front of the car, then as the car moves forwards they are able to collect air from a position where ram pressure increases the pressure (and consequently the density) of air entering the sealed air-intake system.
Posted Image
This illustration is from the book “Competition Car Aerodynamics “ by Simon McBeath, and it shows a CFD (Computerised Fluid Dynamics) illustration of the differences in pressure over a moving car with front flippers. Since the front bumper of the MC has a pair of integral flippers, one of which is positioned directly in front of the left air duct, then it is clear that the snorkel into the air-intake system is “feeding” from a zone of air (shown as red) at the highest pressure around the car when moving. In other words, the zone of air above the front flippers is at greatest ram pressure. Therefore, as the car increases in speed the air entering the air-intake will be at a pressure which is relatively greater than ambient due ram pressure build up, and as such, the charge of air entering the engine cylinders will be greater at any given rotational speed than when the car is standing still.
Posted Image
Since the MC has five forward gears, then for any given engine rotational speed of, say, 7500rpm, the car can be travelling at any of the five gear-dependent speeds of:
1st gear =41.6 mph
2nd gear = 70.3 mph
3rd gear = 105.5 mph
4th gear =141.2 mph
5th gear = 175.1 mph
Bearing in mind that the ram pressure is proportional to the square of the forward velocity, then the air pressure will obviously be greater the faster the car is travelling, therefore, the charge density at 175.1 mph in 5th gear will be greater than at 41.6 mph in 1st gear. In other words, the faster the car moves forward, the greater the pressure of air inside the cylinders, and as such the engine will produce more power at higher speeds. This is known as “ram effect”. It is clear from the design and layout of the MC’s air-intake system that the designers intentionally created the system to benefit from ram effect. I have no idea of the actual air pressures inside the MC’s engine cylinders when at 175.1 mph in 5th gear and at 41.6 mph in 1st gear, all I know is that it will be greatest at 175.1 mph. Some people will argue that ram effect only occurs when the charge of air inside the engine cylinders is at a pressure greater than ambient pressure, however, in my opinion, this is an unnecessary requirement, because what is really important is the actual increase in pressure of the charge of air entering the cylinders when the car is moving at any given velocity RELATIVE to the charge of air that would otherwise occur when the engine is at the same rotational speed whilst the car is standing still. In my book, even if the gain is as little as 1% at 175.1 mph it is better than a 1% loss. The clever use of the knowledge of physics by the MC’s designers must represent a “free” gain in power.
Because ram effect can only occur when the car is actually moving it is impossible to demonstrate this phenomenon on a dynamometer unless there is a fan capable of producing air at speeds up to the car’s maximum velocity. Therefore, when MC owners are comparing the MC’s OE air-intake with an aftermarket induction system by dyno-testing, the results are fatally flawed, because this situation is totally artificial.
The EXDOS MOD
If you remove the air-filter from the air filter-box you will see that there is an area of discolouration at the right hand side of the air filter only, as represented by the red dashed line in the photograph below:
Posted Image
This discolouration clearly shows that approximately only 30% of the air-filter surface is used and, furthermore, if you look inside the air-filter box, you will see that the mouth of the outlet pipe for filtered air lies to the left of the area of the air-filter through which all the air passes. As such, it is evident that the air does not take the most direct route through the air–filter box, instead the air must take a sinusoidal (“S”-shaped) route. Obviously this represents a restriction in the air-intake system, which must make it somewhat harder for the engine to “breathe”. If you look again inside the air-filter box you will see that the oval inlet pipe enters from the front of the box and that the opening of this pipe lies behind the opening of the outlet pipe: clearly, this is the reason why the air passing through the air-filter box must take a sinusoidal path and therefore this is the cause of the restriction within the system. I can only speculate as to the reason why the air inlet pipe has been designed with this restriction, and assuming that this design was deliberate, then there are 2 obvious possibilities: 1. so that the better selling M3 would produce better performance figures so that the MC wouldn’t steal the limelight; or 2. to keep noise levels down.
In order to remove the restriction, the front face of the inlet pipe can quite easily be removed, in situ, with a hacksaw blade so that the air entering into the air-filter box can pass straight through the air filter in front of the mouth of the outlet pipe as soon as it enters the air-filter box. With this simple modification of the inlet pipe, the air is able to pass through the air-intake system much more freely and this is very noticeable as an increased throttle response and more torque at low engine revs.
Posted Image
This photo shows the air-inlet pipe after the front face has been cut away.
Posted Image
This photo shows how you can see right through the inlet pipe into the front brake duct when the Exdos Mod has been performed, showing how air can now directly enter the air-intake system with little restriction.
Since the OE air-filter is quite deep, then replacing it with a K&N filter, which is approximately only half as deep, gives the air entering into the air-filter box more room to take the optimum path towards the mouth of the outlet pipe within the air-filter box. Also K&N claim that their filters are more efficient than paper filters, and if true, then this should further allow the engine to breathe with least restriction.
Posted Image
Since performing the Exdos Mod I have also fixed a simple shroud made from aluminium sheet which is fastened with the screw which retains the air-filter box to the inlet pipe, as shown in the photo below.
Posted Image
The purpose of this shroud is to direct as much air as possible in the most direct path towards the mouth of the outlet pipe for the filtered air, to assist engine breathing.
It is a known fact that for every 5 degree Celsius increase in temperature of air entering the engine, there is an approximate 1% loss in power output of the engine, this is because the density of air is less when the temperature increases. Therefore the cooler the air entering the engine, the greater the power output of the engine. Since the MAF sensor lies in the outlet pipe from the air-filter box, then the cooler the air-intake pipes and fittings can be kept without “heat-soak”, the greater the density/volume of air will be registered by the MAF sensor and thus a greater amount of fuel will be injected into the cylinders. To this end, I have also insulated my air-intake system as shown in the photo below.
Posted Image
I have taken temperatures from the air-filter box when my MC has been very thoroughly warmed up with vigorous driving, and the temperature of the ABS plastic beneath the insulation is always very close to ambient air temperature, whereas in an un-insulated OE air-intake system, the air-filter box will get to around 50 degrees Celsius. Although air entering into the air-intake system will not get heated to the same heat-soaked temperature as the ABS plastic of the system, it will, however, still increase to a greater temperature than if the ABS plastic were maintained at ambient air temperature, thus there will be some diminution of power in an uninsulated air-intake system. Likewise, it is the passage of air over the air intake system, which acts as a coolant and keeps the heat soak to as low a temperature as around 60 degrees Celsius in the air plenum for filtered air, again proving that the charge of air gets heated by passage through the air intake system.
Some MC owners have created additional air intakes into the air-filter box, with the rationale that a more plentiful air supply will be better. Will this actually be better? The answer to this lies in the size of the outlet pipe. The external diameter of the outlet pipe containing the MAF sensor is 8.5 cms and it is made of 3mm thick material, therefore the maximum internal diameter would be 50sq. cms. However, as the diagram below indicates, the opening into the outlet pipe has a thick plastic mesh across it’s face, and I would estimate that the plastic of this grille reduces the full aperture by at least 30%, therefore the true aperture will be of around 35 sq.cms or less.
Posted Image
The OE inlet pipe into the air-filter box has an area of approximately 40 sq. cms., therefore the area of the OE inlet pipe exceeds the OE outlet pipe, therefore no additional inlet pipes are required, especially if the inlet pipe has had the Exdos Mod. Further, since the 2 apertures of the OE air-intakes are situated in the zones of the greatest pressure on the outside of the car, then unless any additional intakes also utilise other zones of greatest pressure, say, an additional snorkel in the right front air duct, then additional air ducts feeding from zones at lower pressures will “dilute” the pressure of air entering into the air-intake system thus defeating the principle of ram effect which the OE air-intake system utilises. It is interesting to note that the Z4 MC also incorporates a very similar air-intake system to the Z3 MC which indicates that the BMW engineers really think that they can harness ram pressure to advantage. The photo below shows the air intake of the Z4 MC and it appears that the air filter lies in a horizontal position which would improve upon the vertical position of the Z3 MC air filter. Interestingly the power output of the same S54 engine in the Z4 MC has a greater torque value.
Posted Image
Aftermarket Air Induction kits
Some MC owners have replaced their OE intake pipes and air-filter boxes with aftermarket systems. Most of these after market systems are essentially conical filters mounted on pipes, which are attached to the OE pipe containing the MAF sensor. From what has been written above about ram effect, it is clear that non-enclosed systems cannot benefit from ram effect. Therefore, in real world driving conditions, most aftermarket air induction systems will rob the MC of power, particularly at higher speeds. Likewise, since many aftermarket systems are not completely heat-shielded, then the air may well be heated by under-bonnet conditions thus air at higher than ambient temperature may enter the air-intake system, again, robbing the engine of power. Personally, I can see no advantage in replacing the OE air intake system, since to my understanding of the principles of physics, the OE system is nearly as good as you can get.
ADDENDUM
Further, to the original Exdos Air-Filter Box Mod I have been experimenting with modifications to the near-side brake duct in order to further harness ram effect. More on this can be found here:
http://www.z3mcoupe....opic.php?t=4703
Both these air intake mods are simple DIY jobs which will cost nothing to do.
Additionally
In order to allow the MC engine to generally "breathe" more easily, I have also devised a relatively simple modifictaion to convert the "reverse flow" design of the exhaust rear silencers to "free flow" configuration, which is described here:
http://www.z3mcoupe....opic.php?t=4703
This mod is a relatively simple task for someone with welding skills and the right equipment. It requires a 115mm angle grinder equipped with a 1mm "plasma" cutting disc, a 230mm angle grinder, an electric arc welder with a cutting rod or a plasma cutter, and a MIG or TIG welder equipped to weld stainless steel. It will take about 3 hours to perform the surgery and reweld the rear silencers, so if you were to pay for this job to be done then it would cost around £200.
Some interesting reading.......
Here's an interesting article exploring the air-intake system and looking for ways to improve air supply http://autospeed.dri...le.html?&A=0629
Also another article which discusses "ram effect" and showing that street-cars can benefit at speeds as low as 40mph and not just at high speeds: http://www.vararam.c...inaroadcar.html[/b]
_________________
Exdos
. I think you can ad real value to the forum and a lot of owners and in turn gain a huge amount of respect what you could contributed to the forum.
)
Liverpool
searched his posts for the answers on the z3mc forum 
and inturn there will be…
