I was thinking of upgrading the headers of the RX8 but keep the stock exhaust. The only reason I would do that is because I've read sites (Racing Beat) say that the Mazda didn't leave much room for improviement on the cat-back. So, just figured this combination would yield some good results on the HP/torque side while keeping most of the OEM parts.

Any ideas of this is wise?

--alex

ps: are the tubular headers for the RX8 being sold?

Originally posted by alex
I was thinking of upgrading the headers of the RX8 but keep the stock exhaust. The only reason I would do that is because I've read sites (Racing Beat) say that the Mazda didn't leave much room for improviement on the cat-back. So, just figured this combination would yield some good results on the HP/torque side while keeping most of the OEM parts.

Any ideas of this is wise?

--alex

ps: are the tubular headers for the RX8 being sold?

yeah, there are equal length headers being made for the RX-8... i think Amemiya-san might have his on for sale, but rotarygod has veeerrrry interesting thoughts about what'll actually work on this engine ('cause the intermediate port is siamesed).

anyways, Racing Beat, and other tuner companies say that kind of thing to cover their ass, and talk up the product. don't think that you can't make more power out of the back end, 'cause there're always factory comprimises which can be swayed one way or the other to gain more performance.

the biggest problem with just replacing the header is fitment: you'd have to find a direct replacement header, and given the design of the current log (which is really REALLY just a log with almost no runner length) it'd be hard to squeeze all the tubing necessary and have the collector in the stock mounting location.

anyhoo, could you?? sure, if the part exists.

BY THE WAY
be a convert of the "Torque Numbers are Marketing BS" crowd, and realise that they're useful only for creating a suggestion of the torque curve of the engine, but given that they really only give you a reading at one speed of the engine are completely useless.

when most people say the word "torque", they're talking abotu "off the line power", which is exactly that: low rpm power.

'cause Americans have always made everythign big (especially their engines), low rpm power is something they're accustomed to in abundance, but for some weird reason keep calling it torque. it isn't, it's still HP, just in the lower specturm of RPM. :)

Thanks for the information, I am usually not too worried about the HP/Torque, its just people in the forum has been bringing up, indirectly, the desire to "boost" the performance. Seemed like lots of credible "post-ers" were making this topic something that was feasible to do and from this, placedin my head the thought of a header upgrade for more "go".

Seeing that you're on of the most knowledgable person on the FD's, I will just leave the of the RX8 performance be (for now .. hehe); I mean, I feel the RX8 is pretty darn fast, even w/ my 19" wheel upgrade.

So what is rotorygod's viewpoint on the Renesis performance potential?

--alex

Originally posted by alex

So what is rotorygod's viewpoint on the Renesis performance potential?

--alex

oh, i wasn't talking abotu the potential: this is one helluva strong motor.

the issue i was talking about was designing the resonance of the header to optimise breathing out of all three ports: an equal length approach apparently isn't the best way to do it, because the intermediate port (middle one) is siamesed between the two rotors. this could be fixed by making the one port into two, then having a 4-to-1 header, or by designing a header with a resonance appropriate to the siamesing.

Originally posted by wakeech
oh, i wasn't talking abotu the potential: this is one helluva strong motor.

the issue i was talking about was designing the resonance of the header to optimise breathing out of all three ports: an equal length approach apparently isn't the best way to do it, because the intermediate port (middle one) is siamesed between the two rotors. this could be fixed by making the one port into two, then having a 4-to-1 header, or by designing a header with a resonance appropriate to the siamesing.

Yeah, I was looking at the exhaust ports from the pictures on Yaw's website, they're very interesting. It certainly looks like a fair amount of testing/tuning will have to be done before a header will be produced that offers significant power gains. Another poster on this board mentioned that the header out right now doesn't offer any power gains!

wakeech I posted an error in the distance ratio.

Here's how the headers I want to build are designed. I looked at the mid 20th century 4 cylinder engines and saw that many had the 2 middle cylinder exhausts siamesed. This baffles me and obviously we have evolved away from this. On race cars however they found a way to make it work almost as good as a true seperate system. There are 2 ways to do it both similar. One way involves figuring out the perfect header primary pipe length and then making the 2 outer pipes collect at this distance. The center pipe does not collect here. Take the length of the primary runners and multiply it by 3. This is how far down after the first collector that the 3rd runner will collect at. That is option one. Option 2 is very similar. The 2 outer runners collect at a certain distance while the 3rd collects farther down. The difference is that the 3rd pipe collects at a ratio of 3/4. Take the first runners lengths divide by 3. This number gives you the distance from collector number 1 to where the 3rd runner collects. This isn't the end though. We want the middle runner to think that it is being tuned the same as the outers. 3 equal length runners won't necessarily do this due to the very different pressures inside the siamesed port. take the original outside runner lengths and make a dead end pipe. Join it into the center runner right outside the engine and make it extend the same length as the outer pipes then just cap the end off. This gives the center port an acoustical resonance at the same spot as the outer ports even though there is twice as many exhaust pulses in that pipe. Acoustics play a huge role in tuning. It isn't only a gas pressure thing, it is also an acoustic pressure thing as well. Unforunately I don't know the lengths that the pipes should be since I don't know the port timing. I could probably guess fairly close although maybe I just think that and it will be way off. The key would be to make the best guess, dyno it, and see where the peak is. If there is a peak I'll know the idea is sound. All that would need to be done then is to work on adjusting the length and then finally proper collector sizes. If I had an RX-8 I'd already be working on it.

I ran the idea by Paul Yaw and he said that he has always wanted to try the dead leg on a rotary but never has. He is more inclined to try it on a bridgeport or peripheral port engine but he'd love to know as well how it works on the Renesis. If Paul is at least curious then it must have some merit. I wish I could build it.

:cool: and there you have it alex.

Wow, I feel extremely humbled .. :eek:

[Peter Griffin - The Family Guy]: Wow, I think I learned something there ..

So, I guess I will save $$$ and NOT buy this enhancement yet.

Thanks guys [Wakeech, rotarygod,SA22C]!

--alex

Paul Yaw had said that the Renesis has proven to be unresponsive to headers so far. If the middle port needs a different length to work right then it will mess up the tuning no matter how long the headers are. Therefore it shouldn't really matter if you have long headers, short headers, or the stock manifold. That middle port is always going to cancel out any benefit as long as it is the same length. This is why I feel that the center port needs all the attention. We can figure out how the outer ports want to breathe. The key is fooling the middle port into thinking that it works the same way. Someone is going to stumble on this and make it happen. When it does, remember where you heard it first!