Damn .................it's sounding more and more like the manifold is going to do sweet FA on its own .

I wouldn’t be so sure. I always said the main improvement for exhaust mods was weight reduction, but opening the whole thing up for maximum free flow is likely worth some gains at higher rpms. So it’s not just the manifold, it’s the rest of it too. Which I had a real 100 cpi metallic racing cat rated for 500 hp with full 3” from the rear exhaust port to the rear of the car splitting into dual 2.5” straight-thru racing mufflers & tips out the back. The flow capacity could support a 400+ hp rotary turbo engine. Some people are claiming Speedsource was using 3.5”, but that seems a bit OTT to me. The best thing is probably a drag style header megaphone, but that’s going to be a serious sound and heat issue, lol.

https://cimg0.ibsrv.net/gimg/www.rx8...5a4731db2.jpeg

Progress!!!!!
https://cimg4.ibsrv.net/gimg/www.rx8...0206a7895d.png

Think I used a shorter radius elbow to match the OE centerline, likely doesn’t matter though. Looks good.

I thought it looked a little too far out ..but it's deceptive because the flange isn't nearly as deep as the stock one .

Yes, looks good. But wouldn't it be advantageous to angle the runners toward the rear to minimize reversion, and increase flow? Probably minimal additional positive effect, but since the point is optimization.

Maybe, hard to say if that would make any difference.................. too late now.

There’s no reversion because without overlap there’s no place for it to go. You have to get your head wrapped around the whole new way of looking at it outside the box of traditional convention. That’s the whole point of making it as free flowing as possible. It just needs to get out freely. The rest will take care of itself.

I get your point regarding no reversion d/t port overlap. However, I'm not referring to inter-port interaction...but the increased backpressure within the runner from perpendicular entry into the collective "log". Same principal involved in designing bends/elbows with as great a radius as possible. Certainly a 90 degree incidence into the log increases backpressure, I.e. reversion in that sense. Quantifying the effect would require comparative testing with a manifold design fitted with all curved runners similar to that fitted on the rotor 1 primary exhaust shown here.

jcbrx8 ..... I think the increasing diameter plus the offset of the port to the log does exactly that to a large extent. Remember there is limited space here so making radius bends flow into the log isn't possible . To do that you need to make a conventional header . And Team has already shown that this design does better than most conventional headers so................................

no, those two ports are discharging to a larger volume where the pulse pressure expands and is decreased. That’s the purpose of expanding come shape of the log itself; the volume increases as each port adds it’s contribution. The front elbow is also working to push the other two down the pipe in a spiral flow motion. Again, while Mazda may not of extracted every last bit due to other factors, they worked very hard at extracting as much power as possible given those limitations. My contention has always been that they already knew what I’ve been claiming about no overlap and how it voids common header theory. That said, I don’t see anything wrong with three short header pipes with radius bends into a proper long angle merge collector. It just adds complexity and cost without any real benefit imo. The exception is weight vs strength, which I now have a tubular header to use ultra thin tubes that have the thermal expansion flexure capability for long term durability. The advantage there is weight though, not power.

While I believe the idea has merit for "some" benefit in the upper rpm range..., I do see both your points: the expanding diameter log and induced swirling motion on exhaust entry would work to reduce the back pressure / reversion effect. And I will concede that the likely minimal gains on an NA application are insufficient to warrant the additional complexity & cost. :icon_tup:

Following w/ interest...

Car is on the dyno ATM and have done a base power run . Only problem..... never thought I'd say this about a Renesis ....it made too much power ! So will do some more runs and try figure out why that was .

You’re not running it on an all wheel drive dyno with it set to all wheel drive mode like the Motortrend guys did with the C8 corvette they tested??? ;)

No ...it's a Mainline hub dyno.
I have tuned a few other NA Rx8s on this dyno and typically we don't see any more than 195 for a good stock setup and 205 for a race prepped setup (stock engine, header, intake but 3" header back exhaust).

We started the day with the stock header. From the first pull we were up around 220whp which I thought was an anomaly so we did probably another 10 or so pulls........... but it was always up at around 220.
So this was : Stock header , stock midpipe with 1/2 cat still in place (rear half gone) , R majic catback , Brettspeed intake , stock coils , mazdaspeed flywheel. Engine has been rebuilt and is in good condition with good compression , porting ...unknown.

Then we set about fitting the new log header . Gav at HPR made an excellent job of it , couldn't have asked for more. I was able to do most of the removal and replacement in HPR's shop myself which was good of them to let me do that. When we removed the midpipe we were very surprised to find 1/2 the cat material still in place. We removed the rest of it and fitted the new log style header then headed back to the dyno.

We did maybe a dozen pulls. Most of these were around the same numbers as earlier with the stock header but one pull showed more power at 233whp . I'm going to say that pull was an anomaly but it's the one we ended up printing off so that's the one you will see here. I'm also going to do some street pulls today to see if anything has changed from two days ago.
Pics to come.....

Well to be honest when I had an engine professionally rebuilt and installed it I was pretty sure it was way stronger than when the car ran the 220 hp dyno on the original OE engine (only around 20,xxx miles or so on it), but I never put it back on the dyno to find out. Just my butt dyno, but the car was super fast and I was having issues with power oversteer competing on 180TW street tires. That was the engine that did around 270 bhp/190 btq on the builders engine dyno though.

Some pics of the install :

https://cimg2.ibsrv.net/gimg/www.rx8...2bb6a97d83.jpg
https://cimg6.ibsrv.net/gimg/www.rx8...7c50836ede.jpg


https://cimg4.ibsrv.net/gimg/www.rx8...1978cd2e72.jpg

The blue line is with stock header and red with new log style.
I really don't trust that 'after' pull as it seems really odd and the other logs weren't showing that. Possibly something to do with TCS as the TCS button seemed to crap out early in the day. I really couldn't see any definitive gains from the header looking at all the pulls together.

https://cimg6.ibsrv.net/gimg/www.rx8...6dfaa8e3ee.jpg

Hopefully there’s a different dyno around you can use. The rolling motion of the data leads me to think it’s dyno induced. My car doesn’t have DSC, but I suppose there could be something going on with that too.

world record for NA renesis? woudl equate to about 300 FWHP?

Peripheral port incoming?

How much of an impact do you think removing the rest of the catalyst had on the results?

Would be interesting to see what a manifold with separate runners of the same size you're using now routed into a merge collector would produce...something like Team's latest design. I suspect that some portion (perhaps small) of the flow from the back 2 runners tries to initially go forward when it first enters the log. If that does happen, I would expect a merge collected manifold would eliminate that problem.

I know we don't trust VD but these two logs were identical conditions as close as I could possibly have done. It is showing small gains from the manifold (green line) up top.

https://cimg4.ibsrv.net/gimg/www.rx8...dd83d857f1.png

I really wouldn't trust that red line dyno ...it just looks ...odd . And when that happens you know you have a rouge result. I really didn't want to post it , but it's all I had.

:suspect: