When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
I have been reading this for a while now and I have to say it's good. I mean I don't have the experience that you guys have (not even close). But wouldn't straight pipe with no muffler of any kind be enough to say there is NO back pressure. We're talking NA of course.
somebody is running a 3.5” exhaust now claiming Speedsource said it’d help NA, but Speedsource ran a header with smallish 1.75” primary tubes and a discharge reducing nozzle, which is classic header theory and the opposite of me because they did it just like everyone else using standard theory. So I’m not likely to go there. I feel like my current setup of 2” 18 Ga primaries x 3” 20 Ga main pipe, and dual 2.5” 20 Ga tailpipes are more than adequate NA. Which I previously had an HJS race cat converter running in an emissions class, but it was almost 7” OD with long tapered inlet and a true 100 cpi ultrahigh flow core and shouldn’t have been much restriction at NA flow rates since it was designed for a 5.0L - 6.0L race engine
My original header was essentially a 3” dump collector with ultra short 2” primaries; essentially a lighter, higher flowing version of OE, which the latest theorists can’t recognize as being as totally opposite as possible to their idea. I know some pretty knowledgeable people that had the resources to try it all and the most porting ever got was around 10 hp at the flywheel on an engine dynoand then side seal life was questionable. Even the Mazda Renesis Pro Formula race car header was large dia, short, uneven length primaries into a large dia short main pipe. They could have done all the snake bundling and porting in the world and didn’t bother. The issue with porting is there isn’t any meat where it’s really needed. So what you have is the merchandising of unicorn chasers for where they can get it. Yet we never see any real world results making it to the top stock port numbers.
Well they trolled me enough. Do whatever you want. It’s your dime.
.
For those who watch this thread for educational reasons, found this that I think was excellent, it also might explain why Mazda used log style, as Renesis engine cover a wide RPM, and tuning length makes torque less linear:
I follow Team at there is a difference due to no overlap, the pulse does not affect intake directly if you have no overlap. However, I cannot see that there is absolutely no gain in having correct length runners to outer ports, if you match length so that negative pulse is hitting exhaust port when port is open from next cycle..?
But I do not know if that ends up in excessive length, as I do not know if they typically shoot for 1/2 or 1/4 of wavelength(Team or Brettus?). Say it is 1/2, and we shoot for 7200RPM:
One exhaust pulse pr RPM gives: 7200/60sec = 120Hz.
At 900°C, speed of pressure is 680m/s, not the regular 340m/s we use at room temp.(https://www.weather.gov/epz/wxcalc_speedofsound) Pressure is a factor as well, but lets keep it simplified.
680/120 = 5.67m is full wave length. Half is approx 2.8m(9.3ft). That might not be too practical for the gains achieved?
And I think I also get what Team mean regarding your porting Brettus, if port opens when there's still is pressure from combustion going on, its almost like shooting pressurized air into turbine. Then you have more heat energy as well, on top of that. I do not know the details, so no opinions though.
Sorry, that argument has been made and it fails because the force to pull a vacuum on a closed chamber; which is what zero overlap timing results in, is much greater than the force available by pulse tuning. You pull on it and then it wants to pull back. It’s not just keeping momentum going with overlap creating an open back door to draw from.
In that scenario all you’re doing is adding restriction without gaining enough to overcome the negative aspect of it. It’s all about give and take and the combination of what yields the best end result. There’s also the tapered end of the exhaust port working against this, which imo is intentional EGR carryover designed in by the engineers. So you don’t get a sharp closure, but a gradual one instead.
Let me add that this is just my hypothesis. Although I do get a bit riled up with rehashing old arguments or with people who are just missing the boat entirely imo and can’t progress past it, I still welcome people to try and add to the discussion. Which again, people have tried all that stuff and including some pretty elaborate snake bundles that have come and gone long ago. People seem to grasp bits and pieces of why it works with overlap and not work without, but often fall a little short of really understanding it all in totality, again imo.
Sorry, that argument has been made and it fails because the force to pull a vacuum on a closed chamber; which is what zero overlap timing results in, is much greater than the force available by pulse tuning. You pull on it and then it wants to pull back. It’s not just keeping momentum going with overlap creating an open back door to draw from.
In that scenario all you’re doing is adding restriction without gaining enough to overcome the negative aspect of it. It’s all about give and take and the combination of what yields the best end result. There’s also the tapered end of the exhaust port working against this, which imo is intentional EGR carryover designed in by the engineers. So you don’t get a sharp closure, but a gradual one instead.
Well, on intake side, Mazda is kind of doing the same with success, just opposite way, forcing air into a filled chamber since we are above 100% VE. If you time the low pressure side of shock wave so that it hit exhaust port when engine is going to open, it means less work to pump out, but most likely not measurable. But if you time it to just before the 3° BTDC, then the negative pulse will help to suck out slightly more exhaust, and leaving more room to fill when you start to fill again over at intake side. How much? I guess only a test setup will tell for sure. It might be that you are entirely correct, that downsides is bigger than upsides.
In order for thar to happen it has to snap shut at just the correct moment. You can theorize pulling a perfect continuous vacuum easily. The reality of what we have to work with doesn’t support it and is eluding you though.
Yes, obviously it has to snap shut at the right moment, thats the whole idea. And if that had a measureable effect, it would only work in a limited frequency/RPM. And make other worse probably, as mentioned in video I linked.
Well like many other people before, you completely ignored me explaining why that isn’t possible. You’re grasp of the particulars is very lacking. To hear you tell it, this is a new idea that nobody ever tried before; it’s just the same standard header theory that everyone else is pushing. You’re failing before even starting, but good luck.
Yet again, all that aside, name even one engine of substance, other than the Renesis, that has zero overlap timing between intake and exhaust events. This is the one point that people either gloss over or ignore in insisting on the same theory as yourself. This is the only reason why I’m even the slightest bit forgiving or understanding of the general inability to accept or understand the situation.
At each combustion event a pulse of high-velocity exhaust gas makes its way down the primary tube, and upon reaching the collector at the end of the primary tube, a low-pressure wave is reflected back up the pipe. Now if that low-pressure wave reaches the exhaust valve while it is open during the overlap period, the pressure differential will actually scavenge the cylinder of exhaust gases. Since the intake valve is also open during overlap, the effect is to actually draw on the induction side. This scavenging effect actually aids in filling the cylinders with a greater charge for higher volumetric efficiency and, of course, more power. The cam plays a vital role here, since the overlap provides the window for the scavenging effect to occur. Obviously, a very short duration cam with virtually no overlap will represent a closed door to the exhaust header scavenging. Fortunately, most hot rodded engines are also equipped with bigger cams, which go hand in hand with headers.
but to be entirely honest, if I was going to sell an RX8 header then I’d make it as long as possible because people aren’t interested in what they don’t want to hear. They want to buy into what they believe in and sounds cool & s**t bragging to their pals, or on blogs, Facepalmbook, etc.. Most of the arguments here stand as testimony to it. As a racer doing his own thing (or armchair theorist, take your pick) I don’t have to care about any of that uninformed malarkey.
It’s just a reality of human behavior; if your goal is to sell something then you have to offer people what’s popular or what they want to buy. It doesn’t matter if it’s a header or foreign cuisine. I know chefs from other countries who won’t make it selling the authentic version from where they’re from. They have to make it taste like people here will like & eat or go bankrupt.
I decided I'd like to copy your header design from the first post Team .Hope you are ok with that. It will be part of my N/A project .
Do you have any suggestions for improving on your original design ? One thing I thought of was to remove all the connections between the port flanges to help with the cracking problem you mentioned.
Initially plan to dyno my existing N/A engine as is, then fit the log style header only, to see if any gains come from just the header.
Thinking I would use the existing stock midpipe with gutted cat but give it a 3" entry to match the header. Wanted to leave the rest of the exhaust (R Majic catback) as it is for now. May look at a full 3" system later but a bit concerned about that being too loud for a street setup.
I think it’s still on my shelf. Would you like me to ship it to you? Has a few cracks, but nothing that prevents running it.
my solution for the next one is individual pipes/flanges and slip fit connections more inline with a common header, but this was done for heat expansion flexure and minimal weight, not for scavenging etc. If you build it thick enough from proper material it should withstand the stress. The one I made is only 18 & 20 Ga. thickness and the heat cycling finally took it’s toll despite being T321 material. Also the guy who repaired it the last several times wasn’t gas purging the interior, which weakens it too.
I think it’s still on my shelf. Would you like me to ship it to you? Has a few cracks, but nothing that prevents running it.
my solution for the next one is individual pipes/flanges and slip fit connections more inline with a common header, but this was done for heat expansion flexure and minimal weight, not for scavenging etc. If you build it thick enough from proper material it should withstand the stress. The one I made is only 18 & 20 Ga. thickness and the heat cycling finally took it’s toll despite being T321 material. Also the guy who repaired it the last several times wasn’t gas purging the interior, which weakens it too.
Thanks for the generous offer . I've already bought the materials and got my fabrication guy onto it though. He has made plenty of rotary manifolds so hopefully we can make something decent that will last between us. Also thanks for all the detailed pics. Makes it a real easy manifold to make and fits in well with the other plans I have.
We’ll note that I don’t just bring the ports in the middle like the OE manifold, but angled the cone downward slightly to bring the middle and rear port in approximately flush to the top side of it. I never really mentioned that in detail before, it was a matter of whether anyone was astute enough to have noticed.
there’s maybe more to why it was done that way than some people might realize ...
You mentioned it in the first post and I was intending to copy that feature . I can see that may set up a swirl motion (as well as providing a smooth curve to exit onto)... do you think the swirling has a benefit ?
I have all the components for my manifold and have dropped them off....now just waiting on my fabricator.
The cone cost $100 to get made up and the rest of the bits added up to another $200 odd. Using 2mm mild steel.
08-27-2019, 05:56 PM
