Originally Posted by rotarygod

Here's all you ever wanted to know about the RX-8 intake.

Aftermarket intakes only do away with VFAD. No there is no ecu retuning to account for this. There are gains (very small) though but they are based on total airflow vs resonant tuning. Somone needs to make a resonant tuned intake that flows good too. This would yield bigger gains that ANYONE else's intake system out there. I've said it here for almost a year now and still no one has done it.

I am not sure but I should at least hope that each valve and VFAD open at a set rpm regardless of load. Tuning is rpm based not load based. This would make the most sense.

If you look at the diagram very carefully you can see how everything works. The wavy lines are the air filter in the airbox. This set of diagrams is the 6 port engine. There is also one of the 4 port engine but not pictured here. It is very similar but lacks 2 intake ports and the VFAD duct on the intake.

Below 3750 rpm, only the primary (2 center ports) are receiving air. This insures that all of the airflow going into the engine is at a high velocity unlike the older rotaries where all of the ports were open at the same time. Mazda tried to overcome this problem with a 3 plate throttlebody and only opening the primary plate alone unit 20% throttle. It is a little complicated and I don't want to explain it all here. Let's just say this (Renesis) is superior.

At 3750 rpm when airflow is at its greatest and is starting to hurt power (intake airspeed around .6 mach), the secondary ports open up. Obviously this is why the fuel injectors are fired sequentially so there is no fuel in a nonmoving runner. Now we have a total of 4 ports being used until 6250 rpm. This again becomes a flow restriction and the tertiary ports open up.

At 7250 rpm the VDI (Variable Dynamic Intake as known on the 2nd gen RX-7) valve opens up which retunes the intake for higher rpm use. I'll explain it in a minute. Also at 7250 rpm the VFAD duct on the intake box opens up. This does a couple of things. It does allow a shorter more direct route for air to enter but that is really a moot point. More importantly it retunes the intake box. The stock airbox size along with the intake duct's length and width combine to acoustically tune the airbox to resonate at a certain frequency. This frequency is set to help the engine gain power at lower rpms. When the VFAD duct opens, it is retuning the intake box and adding its own area to the available area for air to flow through. It does a pretty good job considering that it is only a few horsepower less than the aftermarket cone filter systems but has the capability for way less ariflow. The total area for air to enter the stock box is less than the area of either the throttlebody or the MAF sensor. If Mazda would have given it a little more intake area with the same tuning, the aftermarket probably would have never outdone it. If you want to figure out the perfect size for the intake tube, figure out the diameter necessary for air entering the box at 8500 rpms to not exceed 122.74 mph. That will be the optimum size. Length will be determined based on tuning rpm and intake box size. There is a fun exercise for someone.

Back to the subject of VDI. The reason this valve opens is to retune the intake manifold so the engine thinks the runners are shorter. I've posted this before but here it is again. The intake runner lengths are determined by what rpm the designers want a gain to be at. The continuously variable intake runner length as on the LeMans 787B car is not feasible in a streetcar but we can partially simulate the effects. When VDI opens, the pulses from one rotor do not have to go back to a central point before heading back down the opposite runner to either help or hurt it. In essence when the VDI valve is open, the engine only "sees" the runner up to the valve and not after (before depending on your frame of reference) it. Here's the effective lengths.

6 port "effective" intake runner lengths.

Primary VDI closed 19.8"
Secondary VDI closed 20.4"

Primary VDI open 13.5"
Secondary VDI open 14.2"

Tertiary always 17.3" VDI does not effect these runners.

4 port "effective" intake runner lengths.

Primary VDI closed 27.1"
Secondary VDI closed 29.1"

Primary VDI open 14.8"
Secondary VDI open 16.5"

As you can see when the valve opens it is like installing a shorter intake manifold which retunes for better high end power. The real black art is designing an entire system that works together almost seemlessly. Mazda did a very good job on this intake manifold. It is much better than anything they have ever done and most likely will be very tough to improve upon without some big sacrifices elsewhere in the powerband.


The dips in power as seen on the dyno are the valves opening. The problem isn't so much that there is a pressure drop causing the dips but rather than the valves don't open quick enough. They are not electrically opened as other here have stated. They are opened pneumatically. The electric confusion comes from the fact that there is a solenoid that electrically opens at a set rpm that allows air to flow to the actuators which causes them to open. If a reliable servo could be found, it would be a better system. The faster the better.

I forgot to mention another length difference between the 4 and 6 port engines. The intake duct going into the airbox is not the same between both engines. The 4 port engine does not just use the long pipe from the 6 port engine. It uses a longer pipe. Obviously it does not have the VFAD duct either. Different engine require different tunings.

I wonder if I forgot anything else?