Moved/Merged thread. -RG

.

i just talked to mr amemiya him self and hes say hes getting 10whp from his header and he also thinks that racing beat shouldnt really have that hard of time to get the 7 -8whp they said they seen

the pulse cycling in a piston engine is quite different than a rotary

not untrue, not the complete story either. I'm far from an expert at the deep technical levels, but...

The pulse cycling in a 2-stroke piston engine and a 4-stroke piston engine is quite different. I'm pretty sure that I've read that the 2-stroke and rotary have some similarities (not the same). Probably moreso with peripherial ported rotaries. In 2-stroke piston engines a "chamber" is used to aid in scavenging.

Two-stroke header design is very different than rotary applications, especially for the Renesis since it has zero overlap.

^^ where does it say same in my post? I am aware that rotary headers don't employ a chamber for scavaging. I didn't say that the header designs were the same. I said that the pulse cycling would seem to be more similar between 2-strokes and rotaries than 4-strokes and rotaries. Doesn't it specifically say above that it would apply more to PP rotaries that the Renesis? Not trying to bite your head off, but you kinda missed my point...

Two-stroke header design differs from rotaries in that:

-the pulse spacing per degree of crank rotation is different
-the manifold pressure signature (pressure at a location per degree of rotation) is different

But you are right that two-stroke theory is closer than piston 4 stroke theory when talking about a full perhipheral port motor. So yeah, I guess I did miss your point

not untrue, not the complete story either. I'm far from an expert at the deep technical levels, but...

The pulse cycling in a 2-stroke piston engine and a 4-stroke piston engine is quite different. I'm pretty sure that I've read that the 2-stroke and rotary have some similarities (not the same). Probably moreso with peripherial ported rotaries. In 2-stroke piston engines a "chamber" is used to aid in scavenging.


2-stroke or 4-stroke, you're still talking reciprocating piston engine, like I said the rotary is different :dunno: nor did I claim it was the whole story :confused:

^^no, but as tuj has pointed out now, two-stroke theory is closer to rotary theory than 4-strokes. Not the same. I brought up two-strokes to point out that just piston vs rotary dangerously oversimplifies things as we can see in the differences between two-strokes and four-strokes. Side-ports in the Renesis change things, but let's not "throw out the baby with the bath water".

Since another thread started getting into this topic, I decided to bring the discussion to a new thread. Any header design comments, questions, theory, etc can be debated and discussed here.

Thanks - this is my favorite readings while on the forum :)

So, the only thing holding us back, is the Siamese center port, correct? Is the Renesis the first rotary to have this setup? I most likely will only be asking questions.

First rotary to used siamesed exhaust ports? Yes. First engine to? No. Just to add some fuel to the fire, it is also not the first engine to successfully use zero overlap. This too has been tried and while not in any production cars, it is known to work. OK everyone, get studying and come up with some neat ideas! I've told mine a long time ago.

RG...what is port timing of the center ports....when do they open...and close compared to the side ports?? Also..what is the best theory for pulse tuning with the exaust?? I know this is a bit vague...but in general are we looking to propagate even pulses...or have pulses that vary with intake cycle to allow for better filling during intake?

OK everyone, get studying and come up with some neat ideas! I've told mine a long time ago.


not exactly, please explain you longer siamese pipe theory :angel:


BTW, I have an Re Anemiya header in hand, they seemed to have dropped the center siamese port divider plate shown in their website photo

why not copy over the discussion from the other thread, no sense repeating ourselves, I gave one theory there ...

I have a photo of an RX-8 header that appears to be made from Titanium :SHOCKED: I've been meaning to scan it in and see if somebody can do a japanese translation on it

As far as header flow, the stock header is a bus-type of flow, whereas the RE or OBX ones are hubbed + equal lengths that arranges the flow to not overlap. I think this is beneficial for every type of engine, especially the high revving ones. I still don't know what kind of gains we're looking at since no one has formally installed and dynoed one.

no, neither is equal length,, the closer pipes are swung around some to increase their length, but they aren't equal

why not copy over the discussion from the other thread, no sense repeating ourselves, I gave one theory there ...


OK it's been added here. Unfortunately the way I did it, it's added at the beginning.

The rotary is a 4 stroke otto cycle engine. That is all that matters. It makes no difference if it has valves or not. It makes no difference if it has pistons or not. It is absolutely not similar to a 2 stroke nor is it closer to a 2 stroke than a 4 stroke in any way. If you use a 2 stroke exhaust design on a rotary, you will lose power. I'll let that one be debated a while before I chime in with the answers as to why this is so.

wow,
a tread that i tried to start to move an argument turned into a productive header thread....

cool!

beers

The rotary is a 4 stroke otto cycle engine. That is all that matters. It makes no difference if it has valves or not. It makes no difference if it has pistons or not. It is absolutely not similar to a 2 stroke nor is it closer to a 2 stroke than a 4 stroke in any way. If you use a 2 stroke exhaust design on a rotary, you will lose power. I'll let that one be debated a while before I chime in with the answers as to why this is so.

There is no debating the fact that the combustion cycle is a 4 stroke otto cycle. I don't think that was what any of us were really debating. We were looking at things from what the header sees... exhaust pulses. One rotor making one complete crankshaft (E-shaft) revolution vs One cylinder/one piston making one complete crankshaft revolution. No denying even if they are both otto cycle, the piston and the rotary aren't producing the same thing. Maybe I was way off target by bringing in the two-stroke, however, I didn't say to use a two stroke exhaust design on the rotary. I did say that negating your header idea based upon the it coming from piston engines and this is being a rotary is flawed as two-stroke and four-stroke piston engines are both piston engines, but very different and use very different header designs.

I see where you are coming from now. It has been surmised in the past that 2 stroke exhaust theory would work on a rotary but after studying it, it can't any better than on a traditional 4 stroke.

My way of thinking is based purely on acoustics and what is going on inside the pipes. The idea is inspired by exhaust technology of an older siamesed 4 cylinder engine but it should apply as both are 4 cycle engines. That's really all that matters. The effects of rotary vs piston will only affect total length in relation to each engine.

Again I'm guessing because you're still being coy with sharing the details of your idea, but IMO this will only serve to disperse the total pulse energy of a given rotor, resulting in a net reduction in scavenging ability. Attempting to phase in the center port pulses with a longer pipe (I assume 50% longer for inter-phasing purposes) only negates the total rotor exhaust pulse energy potential. The siamese port in in the Renesis is not at all similar to the model you are attempting to use as the basis for your theory. In that model a rotor/cylinder pulses entirely through the siamese port, it does not pulse through both an individual and shared siamese port simultaneously, thus splitting up the energy. IMO the center siamese port energy will be substantially weaker than the individual port energy for a given rotor pulse and won't be capable of the desired effect.

Is there anything that can be learned from the people that hotrod the flathead Ford V8? I believe that had a siamese center port.

in those situations the siamese port is larger to handle flow from both cylinders, the usual performance fix is weld in a center divider and split them apart from each other

the Renesis is different as described in my previous post, not to mention that the siamese port is not larger yet uses a divider plate to keep from blasting hot gases into the cojoined port of the other rotor, it can get away with this because it most likely flows considerably less of what it's non-siamesed port does, hence my previous argument

if I could get my hands on the port cycle timing and flow rates a flow model analysis and header design could be readily generated; the info is easily obtainable with the right equipment and parts, but it's not likely that anyone would just hand it over ...

am I the only one that thinks exhaust resonance tuning on a rotary is a bit of a waste of time?

It seems to me that with how the exhaust port gradually opens and closes you do not get nicely defined exhaust pulses like in a piston engine where the valve pops off its seat and the cylinder discharges rather uniformly and then gets abruntly cut off with the valve closing.

port timing is in a post I made on here quite awhile ago.... here you go... http://www.rx8club.com/showpost.php?p=461803&postcount=17 Merry Christmas... I didn't tell you to search. :)

That thread in general was good... read if you have some time... http://www.rx8club.com/showthread.php?t=4372&page=2

Since the Renesis rotors each have 2 exhaust ports, 1 shared and 1 not, it is the center one that is already throwing the chamber out of balance. The key to retuning it's length differently is to bring balance back to the system. There is more to tuning than just length. You need to consider how much is going through that pipe and what effective resonance it will have as a result. My idea does not necessarily make the center port twice the length although it could be up to that. The center runner, even siamesed, is still smaller in area than the outers. Diameter of the pipe also plays an important role in tuning in relation to length and what flows through it at what speed. As to what I mean by what flows through it, I am really referring to the effects of having air in the pipe twice as often which would result in a higher velocity of gasses through the center runner even though the total flow volume may be lower. Yes this is possible. To get an idea of how this effects tuning, you need to know someone that plays the flute. Blow through it gently making one note. Now only change how hard you blow through it making it much harder. You will get a much higher note and nothing else changed. You don't necessarily blow twice as much air but you did increase velocity through the flute to change the resonance point. For this simple reason alone you need to retuen the center runner to bring it back all into harmony. I know how it all works, but putting it into words isn't as easy.

Exhaust tuning is absolutely not a waste of time unless you don't care about getting the most out of the engine. Then it would be a waste of time. The Renesis exhaust ports open and close more gradually than a peripheral exhaust port which more closely resembles the characteristics of a piston engnies valves. Valves do not just pop open. The ramp speed of the camshaft profile affects this and it is relatively mild compared to a peripheral rotary port.

Wouldn't unequal length headers sound awful?

Exhaust tuning is absolutely not a waste of time unless you don't care about getting the most out of the engine. Then it would be a waste of time. The Renesis exhaust ports open and close more gradually than a peripheral exhaust port which more closely resembles the characteristics of a piston engnies valves. Valves do not just pop open. The ramp speed of the camshaft profile affects this and it is relatively mild compared to a peripheral rotary port.

but once the valve opens, you have the exhaust gases being pushed through a port of fixed sized by a piston of fixed area that starts from zero velocity, reaches a velocity peak half way, and then slows back to zero velocity. The result is a rather well defined pulse of exhaust gas - which produces a well defined frequency (and natural frequency) and a good shot a getting a good clean resonance at 1x, 2x, 3x, ect times the natural frequency.

on the rotary, you have the rotor sweeping by the port - meaning you have gases going through a continually varying port area from continually varying rotor area for the amount of time the port in open. Off the top of my head it looks like instead of getting a good clean pulse, you get a distorted and noisy looking pulse - which IMO limits the potential for resonance.


I'd sooner spend time on a making a header with equal effective length runners (effective length meaning equal resistance, meaning you have to take into consideration pipe diameter, bends, amount of flow, and physical length)... but hey, this is just my little theory

Two-stroke header design differs from rotaries in that:

-the pulse spacing per degree of crank rotation is different
-the manifold pressure signature (pressure at a location per degree of rotation) is different

But you are right that two-stroke theory is closer than piston 4 stroke theory when talking about a full perhipheral port motor. So yeah, I guess I did miss your point

Actually there are more crucial differences:

- A rotary engine and a 4 stroke piston engine both do have a solid rotor or piston expelling exhaust exhaust gases. A 2 cycle engine does not.

- Not only does the 2 cycle engine have more overlap - the exhaust port actually closes AFTER the intake port closes. This is why the expansion chamber design is so crucial. http://www.liebold.com/rumipages/espansioni/twostroke.html (There's no use for an expansion chamber in a 4 cycle engine - be it rotary or not.)

- 2 cycle gasoline engines have lower compression ratios, do have a high volumetric efficiency and have therefore a significantly higher initial exhaust pressure.

The problem is that exhaust ports/valves don't work like that except on a peripheral port. Even then velocity is not constant and is not at it's peak when the valve/port is open at it's peak. It is possible to have higher velocity at port open and closing but less than this at full open. It is all dependent on design. For the most part, the strongest pulse is when the port/valve first opens and it generally decreases from there. On a peripheral exhaust port this is very sudden and the pulse is strong as the port reaches it's full open timing very quickly. On a piston engine or side exhaust port, this is a much more gradual opening as it happens over a larger degree of timing. This makes the inital pressure pulse weaker than what a peripheral port would be but by no means insignificant. It it weren't of any significance, headers wouldn't matter on any engine.

Wouldn't unequal length headers sound awful?

Almost every car from the factory with the exception of a few have unequal length exhaust runners.

IMO it won't work the way you seem to think it will, but that's easier to say than prove. WRT the Renesis engine, the exhaust energy of each rotor has two paths to follow. If both paths are not equal in all respects the energy will bias itself toward the path of least resistance until balance is achieved. You're attempting to take a single port per chamber theory and apply it to a dual port per chamber situation. What you propose will exacerbate the imbalance, not reduce it.

The exhaust ports of a dual port chamber are cojoined at the hip. You can't treat each port as a separate chamber entity. It's not going to happen that way.


Since the Renesis rotors each have 2 exhaust ports, 1 shared and 1 not, it is the center one that is already throwing the chamber out of balance. The key to retuning it's length differently is to bring balance back to the system. There is more to tuning than just length. You need to consider how much is going through that pipe and what effective resonance it will have as a result. My idea does not necessarily make the center port twice the length although it could be up to that. The center runner, even siamesed, is still smaller in area than the outers. Diameter of the pipe also plays an important role in tuning in relation to length and what flows through it at what speed. As to what I mean by what flows through it, I am really referring to the effects of having air in the pipe twice as often which would result in a higher velocity of gasses through the center runner even though the total flow volume may be lower. Yes this is possible. To get an idea of how this effects tuning, you need to know someone that plays the flute. Blow through it gently making one note. Now only change how hard you blow through it making it much harder. You will get a much higher note and nothing else changed. You don't necessarily blow twice as much air but you did increase velocity through the flute to change the resonance point. For this simple reason alone you need to retuen the center runner to bring it back all into harmony. I know how it all works, but putting it into words isn't as easy.

Exhaust tuning is absolutely not a waste of time unless you don't care about getting the most out of the engine. Then it would be a waste of time. The Renesis exhaust ports open and close more gradually than a peripheral exhaust port which more closely resembles the characteristics of a piston engnies valves. Valves do not just pop open. The ramp speed of the camshaft profile affects this and it is relatively mild compared to a peripheral rotary port.

'not sure whether this has been discussed - but the fact that there's no overlap appears to be a good explanation why headers don't seem to lead to significant gains on the Renesis.

The major reason why header increase power on a piston engine is because it improves scavenging. Scavenging increases the amount of burnable gas in the combustion chamber, which generates more power.
It doesn't surprise me that the 6 can get 20HP with a header and the RX-8 doesn't. http://www.cp-e.com/imgs/maz6s_hdr_hp.jpg

If I was to the design a header for the Renesis I'd mainly try to make it as light and simple as possible.

So does less power necessary to expel exhaust gas mean better engine response? would the rotor be easier to spin faster? These all all very new to me!

Well you slightly increase power. (Which translates to faster engine response).

This is how these 4 cycles work:

1. - power to suck the A/F mixture
2. - power to do the compression
3. + power from burning the A/F mixture
4. - power to expel the exhaust gases

Major power increases are accomplished by increasing number 3 which means getting more combustionable gas in your engine. Obviously you can increase power by reducing 1 and 4, but they are already neclectable.
Here's a picture of the 4 cycles: http://www.grc.nasa.gov/WWW/K-12/airplane/ottoa.html (compression and power stroke is what is relevant: See 2, 3, 4, 5 in that picture).

Back to the header design: I guess my question would be, why not simply have 2 main pipes leading to the engine. Why not get rid of the 3rd pipe and lead the exhaust gases from each side of the siamese port back to one of the main pipes immediately after the engine (after all they come from the same exhaust chamber anyway).
In other words: 4 very short pipes leading into 2 mainpipes and then into one. If nothing else it would at least be lighter than the 3 pipe design. Besides it would be easier to model - just need to calculate the length of the 2 mainpipes.

The 3 pipe header is essentially the same as if you would add 3 exhaust pipes on each cylinder bank of a STi.

I don't want to get into this since I have no rotary experiance. I'll leave it to the experts and watch. I have some math that will show that the reduced pressure will help the fill. Just replce the piston events with the proper rotor cycles.

I also don't know if the way I installed the shortcut here will work. If not go to page 5 of the Axial flow thread and get it here.






http://www.rx8club.com/attachment.php?attachmentid=19461

I started down the path of fabricating my own header about a year ago. I can't lay my hands on my original calcs right now since I'm not at home, but needless to say, I decided that it wasn't going to be practical to design for scavenging. I even found some real test data where we had measured the speed of sound in hot exhaust gases! As I recall, the length of pipe needed to tune for sacavenging was going to be greater than 10ft because the exhaust pulses come relatively slowly - basically one exhaust pulse from each rotor per eShaft revolution. You'd have to design the header so that the exhaust pusles were tuned to a SUBharmonic of the header. I remember thinking that piston engines had the advantage here because they have more exhaust pulses per second than rotories do.

Of course, I could be full of crap. What the hell do I know? :angel:

I don't want to get into this since I have no rotary experiance. I'll leave it to the experts and watch. I have some math that will show that the reduced pressure will help the fill. Just replce the piston events with the proper rotor cycles.
If inlet port and exhaust port are open at the same time, reduced pressure in the exhaust can pull fresh gases into the combustion chamber.
In the case of the Renesis inlet port and exhaust port are apparently never open at the same time. (Which means no matter how much you pull on the exhaust side, it won't help pulling fresh gases into the intake).

Nevertheless reduced pressure in the exhaust will reduce the residual gas (from the last combustion) in the intake and therefore will help the fill. So a good header design will increase power even if inlet and exhaust port are never open at the same time. However it won't help the fill to the extent it could if there was overlap - where an appropriate header design can pull all the residual gas out of the combustion chamber.

hey richie you are a genious, im not mechanically able but sort of understand and appreciate your info. question for you tho does it help just replacing a k&n airfilter without doing anything with my exhaust

Since there is no overlap, the only thing to do is flow improvement, by designing the headers to avoid interference during positive pressure. Its possible to tune for improvements at one particular rpm and see significant gains there, but usually no where else. Many do not realize that the pressure in the headers actually goes negative for part of the timing. Yes, the flow in the exhaust is actually going backwards sometimes. By tuning the header lengths to work together, you can get the out-going pressure pulses to line up and increase the pressure, or you can use interference to cancel out some of the negative pressure.

I'm not sure there is much to be had in headers here. The scavenging effect is the most crucial in terms of adding power, and it doesn't come into play here. This is a complex subject, because the pressure in the header varies by 3 variables:

-location of the pressure measurement in the header
-time during the combustion cycle
-rpm the engine is operating at

hey japan 8 where did you get yours m symbol its cool
much appreciated

No

There are two waves, pressure wave and partical wave. The partical wave does not reverse. The pressure wave will reverse when it hits any sort of change. Not just the end of pipe. If there is a sharp increase in pipe dia it will act like it met the end.

I know what idea you have!

we should set up a rx8 header fund for charles hehe :)

If you'd increased the size of the intake and port areas such that you'd end up with overlap, a header would be more effective and especially top-end horsepower could be increased.
Unfortunately fuel efficiency and idle stability would suffer and emissions would go down the toilet.

(Note that overlap is not engine-type specific. Either piston or rotary engines can be designed with or without overlap. Also, the Renesis is possibly the only rotary engine without overlap).

Of note, every modern piston engine has overlap. Overlap doesn't necessarily ruin emissions or fuel efficiency. In the early days of engine development, it was thought that overlap would fail spectacularly.

I was wondering if you could get the tuning right on the pipe length / diameter if you could actually create a slight vaccuum inside the chamber after the rotor closes up the exhaust ports. While it probably wouldn't be as efficient as actual overlap, if there was a partial vaccuum inside the combustion chamber when the intake ports were uncovered, you would get more air into the engine....

I doubt this would even work, since somebody else said that proper tuning length is 10+ft...

Of note, every modern piston engine has overlap. Overlap doesn't necessarily ruin emissions or fuel efficiency. In the early days of engine development, it was thought that overlap would fail spectacularly.
True. The Renesis is possibly the only engine in a passenger car without overlap.
Overlap doesn't necessarily ruin emissions or fuel efficiency and it might actually improve fuel efficiency at higher rpms. However, I believe it's fair to say that if engines wouldn't operate above 2500 rpm, they probably wouldn't have overlap. Overlap is mainly a compromise in order to gain horsepower at higher rpms.
Unless fuel is injected after the exhaust valve or port is closed, a small amount of the air/fuel mixture can always leave through the exhaust port (rotary or piston).

In the case of the Renesis Mazda obviously wanted that the unburnt air fuel mixture would get 'recycled' in order to improve fuel efficiency and emissions. If there was overlap the fresh gases could push the unburnt exhaust gases out of the combustion chamber. The piston engine doesn't have a moving combustion chamber and therefore doesn't deal with the same extent of unburnt fuel. I used to have an RX-3 and it measured about 3000ppm of unburnt fuel at idle, a piston engine from the same time measured about 300ppm. Actually the fact that the RX-8 still does have an air pump indicates that even the Renesis still deals with that issue to some extent despite side exhaust ports and zero overlap.

This is a stupid idea, but would it be possible to pressurize the exhaust manifold at lower rpms where a large overlap would adversely affect mileage? That way it would have a higher pressure at lower rpms, and get better power at higher engine speeds?

If engines do have large overlap, they also have variable valve timing.

I was wondering if you could get the tuning right on the pipe length / diameter if you could actually create a slight vaccuum inside the chamber after the rotor closes up the exhaust ports.
Yes

So would headers improve this cars hp?

Yes but instead of 10% more hp you might only get 3%.

If you have overlap and if you only have a slight pressure drop below atmospheric pressure (not vacuum) in the exhaust chamber, the intake gases can push the residual gases out of the chamber. Or in other words with a compression ratio of 10, this means 10% more massflow = 10% more power.

If you don't have overlap, the pressure drop will only reduce the amount of the residual gas but won't get rid of it. (Unless of course the header can generate a perfect vacuum, which is hardly feasible.)

If engines do have large overlap, they also have variable valve timing.

BS. Tell that to all the V8 guys with their wild cams.

However, I believe it's fair to say that if engines wouldn't operate above 2500 rpm, they probably wouldn't have overlap. Overlap is mainly a compromise in order to gain horsepower at higher rpms.

That's completely wrong. Overlap was found to be beneficial at only 200 rpms in the early days of engines. Overlap is a critical component of any N/A engine.

http://www.amazon.com/gp/product/0837603099/qid=1136381704/sr=8-1/ref=sr_8_xs_ap_i1_xgl14/103-4917101-5195801?n=507846&s=books&v=glance

BS. Tell that to all the V8 guys with their wild cams.
'Ever noticed how stable and emission friendly these engines are at idle?

Overlap is a critical component of any N/A engine.
I guess you should tell this the Mazda engineers that designed a N/A engine without overlap.

Yes but instead of 10% more hp you might only get 3%.

If you have overlap and if you only have a slight pressure drop below atmospheric pressure (not vacuum) in the exhaust chamber, the intake gases can push the residual gases out of the chamber. Or in other words with a compression ratio of 10, this means 10% more massflow = 10% more power.

If you don't have overlap, the pressure drop will only reduce the amount of the residual gas but won't get rid of it. (Unless of course the header can generate a perfect vacuum, which is hardly feasible.)



With all do respect, and I'm not being my usual smart ass self. But are you serious?
Could you show me the math on this? Or are these numbers from out of the air?

With all do respect, and I'm not being my usual smart ass self. But are you serious? Could you show me the math on this? Or are these numbers from out of the air?

Ok here's the math:
(Displacement + Displacement/(Compression ratio)) is the total volume of the combustion chamber.
Example: an engine with 2l displacement and a compression ratio of 10 can ideally burn 2.2l of air fuel mixture. 10% more massflow is approximately 10% more power - agreed? It's not perfect and there are more factors playing a role, but it's a reasonable approximation.
Now, an engine without overlap cannot easily get rid of the residual gas which in this example takes 0.2l (since there is simply no draught). So, the power gain is obviously significantly less than 10%. 2%, 3% or even 4% - I don't know. But to reach 10% you have to come up with some header design that generates an absolute vacuum.
What is the lowest absolute pressure ever measured in an exhaust pipe? If you can give me that number I can do the entire math.

Here's a easy project for home owners to understand what with or without overlap means:
On a windy day open one window on the east side of your house. Do you notice a strong wind in the house? (Probably not). Now open one window on the west side of your house and make sure all doors inside the house are wide open and simply watch what happens.

I'm kinda an AH here, I set you up. I already gave you the math in my paper on mass flow. But note it says "Theory" on it.

Refer to post 37. this thread.

The purpose of that paper was supercharging and filling the unswept volume. But note I didn't even claim the whole gain when supercharged to Pr2. I was skeptical even then. I think you might be carefull to claim that much gain just from blowdown.

Yet we know that a F1 can get 125%+ massflow. Albet in a very limited RPM range. So where am I on this? Somewhere on your side but can't give you that much gain on a road car no matter how much overlap you have.

I like seeing you thinking though. Just be carefull not to use such quick math.
If you read my paper you will see I break off and relate to the supercharged swept plus unswept volumes.

I know it's geek math but highly nessasary with all the variables involved. Reading it is all you need to do, forget doing the math. Leave that to the collage students who need the mental work out. Like Shellys or however he spells it.


One more thing, your example would yield 11 to 1 compression.

Hey, just a thought. Could you drill a hole diagonally through the face of the rotor and out through the side of the rotor so that you effectively produce overlap? Please comment on the idea, itself. I realize that this might cause more problems than worth.

I'm just going to comment that even at 100% volumetric efficiency on an engine you still have 10% exhaust gas dilution. Even with no overlap, there is still a small amount of exhaust gas returning to the intake side. Now lets say we had an exhaust that really scavenged good. In other words what if we had an exhaust that kept velocity high and actually pulled more air out of the engine and created less than ambient pressure inside the chamber just as the exhaust port is closing? It is possible and very feasible to do. You just can't get it at every rpm. You need to pick a spot but this is just tuning as with everything else. What if we got -1 psi in the chamber. What will happen as the intake ports open to a -1 psi pressure inside the engine? The air will accelerate into the engine and you will effectively raise the efficiency.

Overlap does not need to talk place to get intake scavenging. The commonly accepted definition of overlap states that it is when the exhaust has good velocity so that it's velocity pulls some air into the engine through the intake ports during overlap. I have even used this statement before. That is how it works on an engine with overlap though. These engines during overlap will also pull some of the intake charge out through the exhaust and lose some of what it was gaining. If overlap was important to making high rpm power, the Renesis wouldn't make more power up top than the 13B. It most certainly does and it even does it with a higher and wider usable powerband so it's obvious that conventional thinking into how overlap is beneficial is misunderstood in some way. It's not saying that overlap can't be beneficial. It most certainly can. No overlap is just better than what everyone thinks.

The purpose of that paper was supercharging and filling the unswept volume. But note I didn't even claim the whole gain when supercharged to Pr2. I was skeptical even then. I think you might be carefull to claim that much gain just from blowdown.
I agree it's an ideal view, but especially with a supercharger the gains with blowdown due to overlap should be noticable.
I mainly tried to make a point why headers don't lead to the gains people might wish for and I believe this is a reasonable explanation. If there was exhaust pressure data available (with or without header) it would be possible to make a more accurate guess.

Hey, just a thought. Could you drill a hole diagonally through the face of the rotor and out through the side of the rotor so that you effectively produce overlap? Please comment on the idea, itself. I realize that this might cause more problems than worth.
In that case the combustionchamber wouldn't be be sealed anymore (look where the side seals are).
It might be more feasible to add peripheral ports to the exhaust (drill 3 holes and fill them with tubes).

The pressure in the pipe is below atmospheric at times, and it is important that the optimized location of this point be determined by the length from the port. There are two considerations in this variable: column inertia, and scavenging.

Since most people are familiar with the latter, I'll comment on the following; the exhaust gas coming off of the rotor face has some velocity that be utilized effectively in some engines. This velocity can be considerably more important than the resonance vacuum, but is not always. When it is, the resonance vacuum point should be relocated farther down the pipe. If the vacuum is maximized at the port, even a preferrable inertial effect can be overcome by the vacuum.

Some people might dismiss this on the Renesis because of the 90 turns in the ports, but poppet values essentially present the same problem, while scavenging is reduced by the complete absense of overlap. The problem with overlap is that the effect results in blowdown at some rpms, which is wasteful of the mixture, and is the reason why power and fuel efficiency can generally be considered mutually exclusive (and why the curve between the two isn't linear).

The other issue with overlap is partial throttle operation (90% of what most people operate their cars at): At partial throttle or at idle, exhaust gas is pulled into the combustion chamber as long as the exhaust valve is open, which is especially noticed at lower rpms and usually not a very controlled process and can cause misfirings. (The piston/rotor generates a vacuum and the gases take the path of the least resistance - if the throttle is closed and the exhaust pipe is open...)
Race engines are usually not operated at partial throttle and don't need to worry about controlled partial throttle operation and can therefore deal with much larger overlaps.
Also, to some extent poppet valve engines also need to have overlap just to give the cam enough time to actually open the valve (cams with vertical walls won't work). This is an issue the rotary engine doesn't really have to deal with.

Last but not least: If the Renesis had openable peripheral ports in addition to the side ports it could have the best of both worlds. Keep in mind the 26B had the same compression ratio as the Renesis and delivered 700HP at a relatively low 9000 rpm - which would translate to 350HP with 2 rotors...
(This is actually a pretty high number compared to the F1 engines and their HP numbers at 9000 rpm and a compression ratio of 14.)

Sorry, but the part about the cams isn't right. I can easily have a DOHC engine with zero overlap. Ramp angle isn't the same thing. And the rotary does have the equivalent of a cam lobe's ramp in the port shape's leading edge. I certainly agree that race profiles run much more overlap than street cams.

Second, you seem to suggest that overlap is only beneficial at WOT, which is not true. Air is flowing significantly through the engine at many throttle openings. Even during crusing there is benefit to some overlap in conventional engines.

Finally, the 787 was a full race motor. Typically, all-out motors can see roughly double their production hp levels. An N/A 13b in that era making 175 hp wouldn't be unreasonable. I'd like a peripheral port motor that could pass emissions too :ylsuper: , but I doubt that will ever happen.

you seem to suggest that overlap is only beneficial at WOT, which is not true. Air is flowing significantly through the engine at many throttle openings. Even during crusing there is benefit to some overlap in conventional engines.
When you already throttle the airflow why would you need overlap for increased airflow? First you throttle it and then you open it up - why not close the throttle less in the first place?

Finally, the 787 was a full race motor. Typically, all-out motors can see roughly double their production hp levels. An N/A 13b in that era making 175 hp wouldn't be unreasonable. I'd like a peripheral port motor that could pass emissions too , but I doubt that will ever happen.
Actually if you limit rpm (and compression ratio) on an engine, power of a race engine is not very impressive. A 3.0l F1 engine makes 900HP at 19000 rpm, which translates to 280HP at 6000 rpm. If the 3.0l engine in the 6 had a compression ratio of 14 it probably wouldn't be far off from 280HP.
If the peripheral ports would only open at high loads and rpms I believe emissions could be passed, but I also doubt that this will ever happen simply because I don't believe that there's much R&D in the rotary department and it would add a lot of complexity for a relatively small power gain (compared to FI).

[QUOTE=globi]


Actually if you limit rpm (and compression ratio) on an engine, power of a race engine is not very impressive. A 3.0l F1 engine makes 900HP at 19000 rpm, which translates to 280HP at 6000 rpm. If the 3.0l engine in the 6 had a compression ratio of 14 it probably wouldn't be far off from 280HP.
QUOTE]


Globi, Globi, Globi, Pleeeeezzz you cannot extropolate like that. There is no comparing these things. An F1 engine idlles at 6000 and doesn't make near 280 hp. It can't even pull the car into motion with that RPM. It would stall.

Compession ratio does not act linear. From 9 to 1 going to 10 will yeild more then going 10 up to 11 and so forth. Notice that NASCAR lowered their engines from 14 down to 12 and did not slow down a drop.

Static compression ratio is the number that we know as 10.0:1 but the effective compression ratio is what determines power output. The effective compression ratio changes with engine efficiency. When you have 100% volumetric efficiency, you have the same static compression ratio as you do effective copression ratio. If your volumetric efficiency is less than 100%, your effective compression ratio will be less than 10.0:1.

Static compression ratio is the number that we know as 10.0:1 but the effective compression ratio is what determines power output. The effective compression ratio changes with engine efficiency. When you have 100% volumetric efficiency, you have the same static compression ratio as you do effective copression ratio. If your volumetric efficiency is less than 100%, your effective compression ratio will be less than 10.0:1.



Excellent RG, This is why you can run hi ratios on the likes of Buick nail heads and May combustion chamber heads. If nothing gets in you can compress it all you want.

the whole thing is much more complicated than some people here have made it out to be, the wankel concept adds another level of complication, and the Renesis ups it another notch still. This application is not easily modeled using the the typical engine analysis techniques available

Based on my research you're likely looking for a total pipe length to the collector in the 18" - 24" range, pipe sizing from low end to high end power in the 1-5/8" to 1-7/8" range (probably larger still for super high boost applications), you can step up to the next larger size pipe about half-way to the collector to get a wider powerband between the two, and if you want to split the siamese port flow and keep the gases separate between the two rotors in that port then half the area of the center port works out to about a 1" ID pipe. Really high end power will likely require porting and the other usual mods, not very well documented territory for the Renesis yet.

well this might sound stupid but what about a reverse turbo, something that literally sucked the exhaust out. The only thing is that the energy needed to make it work would probably be more than the gains it could acheive...

yep, you're talking about perpetual energy, not going to happen, the losses will exceed the gains in the real world

Globi, Globi, Globi, Pleeeeezzz you cannot extropolate like that. There is no comparing these things. An F1 engine idlles at 6000 and doesn't make near 280 hp. It can't even pull the car into motion with that RPM. It would stall.

I know that F1 engines don't run at 6000 rpm. However power output is more or less proportional to the engine speed. I just wanted to show just because a race engine is a race engine doesn't mean that it can break the laws of physics. The torque number of an F1 engine is not much higher than of a 3.0 l production engine if both engines have the same static compression ratio.

You can always simplify to some level. If you limit displacement, rpm and static compression ratio engines will be around the same power level.
You won't find a new 2.0l engine making the same power at 4000 rpm as another new 4.0l engine if both engines have the same static compression ratio.

One of the main reason why race engines are powerful is because they can spin faster and have a higher static compression ratio.

well this might sound stupid but what about a reverse turbo, something that literally sucked the exhaust out.

It would work if you'd run off the battery, but again it would need much more power without overlap than with overlap.
Besides you'd gain more power by running a compressor to push more fresh gases in than running a pump to suck more exhaust gases out.

A header is essentially a 'reverse turbo' just without moving parts that essentially uses the energy already in the exhaust gases to generate a reduced pressure.

The above posters are right about the compression ratio. It comes into play even more when dealing with the effective compression ratio with FI. Lower static and more boost is better than high static and low boost.

When you already throttle the airflow why would you need overlap for increased airflow? First you throttle it and then you open it up - why not close the throttle less in the first place?


Because scavenging increases the thermal efficiency of the engine.


Actually if you limit rpm (and compression ratio) on an engine, power of a race engine is not very impressive.

Yes, the but virtue of a race engine is that it can rev safely and flow at those high rpms.

Because scavenging increases the thermal efficiency of the engine.
Have you thought about why scavenging at partial throttle should increase efficiency? If there's overlap some of the fuel can leave through the exhaust port and this would obviously reduce efficiency (which btw is not that unlikely considering the fact how close inlet and outlet valves are).
I could see efficiency improvement if it reduces pumping losses (but most engines have EGR - so this is already taken care of) or if it somehow increases turbulence in the combustion chamber and therefore lead to a more complete combustion. Both are big IFs though.

well this might sound stupid but what about a reverse turbo, something that literally sucked the exhaust out. The only thing is that the energy needed to make it work would probably be more than the gains it could acheive...
While this is an extreme example, it has alot of very relevant information as they are using zero overlap. read through all of the choices on the right to get an idea of how the theory works. There are several things that we wouldn't want to do but some of it could apply to the Renesis.

http://www.impulsengine.com/how/briefhow.shtml

Interesting read.....seems to be almost exactly opposite to a lot of conventional wisdom.

Do you think that this will help make the rotary's unique combustion chamber shape more efficient...and speed up the flame front...

While this is an extreme example, it has alot of very relevant information as they are using zero overlap. read through all of the choices on the right to get an idea of how the theory works. There are several things that we wouldn't want to do but some of it could apply to the Renesis.

http://www.impulsengine.com/how/briefhow.shtml


kind of funny that they don't show dyno sheets etc. to back up their claims :dunno:

:bsflag: :bsmeter:


Sounds like that thing was written for a research grant.
It'll start but that's about it.

at a minimum it would greatly sacrifice top end for low end, maybe on a low rpm poppet valve engine but I can't see it working on a wankel

http://www.impulsengine.com/how/briefhow.shtml

If you could design a thing that creates incredibly low pressures, power could be increased on a 4 cycle engine (rotary and piston) but I don't see how one can exceed far more than 10% (they're mentioning double), since the dead volume as discussed before is not more than about 10% anyway.

It actually states here that it doesn't work:
However, the high pressure gas in the small pipe of the header will remain compressed for only a short distance before it builds up back pressure and restricts the gas flow. Therefore, the small pipe of the header is very short and a megaphone pipe is used after the header to allow the compressed gas in the small pipe to gradually expand into the larger diameter of the megaphone pipe. This allows the exhaust gas to flow only away from the exhaust port at a very high velocity and with no back pressure to restrict the gas flow.
Obviously when it builds up back pressure directly after the combustion chamber how can it create a vacuum in the combustion chamber? Yes, it can create a vacuum after the restriction in the exhaust pipe but how would that benefit the engine?
Btw this is called Laval nozzle (or converging diverging nozzle) and was invented about 120 years ago when they designed the first steam turbines.

As the piston moves down the cylinder during the intake stroke, it further reduces the pressure in the cylinder. This pulls an additional volume of hot air into the engine faster than the speed of sound. At the end of the intake cycle the intake valve closes early after BDC to trap the much larger volume of hot air in the cylinder and prevent the air from being forced back into the intake manifold during the compression stroke.
Wasn't there a guy named 'Smokey' that came up with that hot air concept in the 80's? As I can imagine that it would increase mileage due to reduced pumping losses and improved combustion, I don't see how a less dense charge can increase power.

Obviously as I stated not all of that would be usable but some of what they discuss is quite relevent and feasible. I'm not going to go so far as to say that running a smaller exhaust pipe with no overlap and a hot air intake will give you supercharged power. The relevant issue is that you could use exhaust velocity with no overlap to get less than ambient pressures in the engine before the intake ports open. That would be very possible. It would have some effect on power.

Non of this is going to do anything for the rotary because even if you had a negative pressure in there you have no volume of it. example, what say you took the vacuume reserve for your power steering and reduced it to the size of a pill bottle, you'd have not much braking power. Because you ran out of vacuume, you used it up.

Now Smokey Yunick's engine is something I know about. Smokeys idea was homoginazation, you all know that I worked on that too. Mine was doing it with my "blender" as Smokey called it. We talked on the phone alot about this. In his application he used a turbo to homg the mixture but it didn't work well as it tosed the fuel out of mixture. He had to superheat all the ducting after the turbo to keep it mixed. He didn't know about my blower when he worked on this project.

His idea was getting it well vaporized under low load then using the turbo for high power. Problem was transition whereby shutting off the heat to all the tubing took to long. But he told me that once the fuel was mixed it stayed that way for long time without falling out. This is what I found also.

He then was quick to tell me not to bother with Detroit, I'd be beating my head against the wall. He said they don't want to hear anything not even from him.
Many years ago they would fund promising projrcts but not since the '60's have they shown any interest in outside development.
All they want now is to buy a "black box" from someone. Meaning do your own development work on your own money.

Last time I saw Smokey was at the Indy 500 in '96 or '97. At that time I showed him my then new 5 inch blower. (The new one is 4.4) He offered it to his wife for use as a goood garbage disposal. I'm not sure if that was a compliment or not.

Anyway this thread is using up to much of my time. I'm not going to comment any further on any of above. Fred if you want to discuse it call me sometime and we can go over it.

Richard, thanks for elaborating about Smokey's engine. I have never read the entire story and always thought how would he have gained power when he reduced the air density by heating the intake charge, but now it's clear.

Regarding the 'negative pressure' (which is just less than atmospheric pressure):
If you reduce the pressure in the exhaust chamber more exhaust will leave through the exhaust which would then allow more combustionable mixture to enter the intake even if there's no overlap (I don't see why this concept would be not valid for all 4 cycle engines).
I definitely see the benefit of reduced pressure in the exhaust chamber, I just don't see how it could be achieved with a some 'Megaphon' pipe.

Anyway back to the header:
I'd be curious to know what the power increase with headers was on the 10A, 12A and 13B engine. This could tell us whether rotary engines actually achieved higher power with overlap (in percentage to the max power).

Regarding the 'negative pressure' (which is just less than atmospheric pressure):
If you reduce the pressure in the exhaust chamber more exhaust will leave through the exhaust which would then allow more combustionable mixture to enter the intake even if there's no overlap (I don't see why this concept would be not valid for all 4 cycle engines).

).
If there is exaust/intake opening overlap there will be no residual negative pressure...it will just pull intake flow into the exaust. The negative pressure " supercharging" results from the sealed -ve pressure left after the port closes...sucking in intake charge when the intake valve opens...

If there is exaust/intake opening overlap there will be no residual negative pressure...it will just pull intake flow into the exaust.
Exactly that's why an engine with overlap is way more effective in improving filling. Only a small 'negative pressure' (this is a bad word) is needed to get better filling.

True...I think that overlap traditionally allows for dilution of the residual exaust gases with intake charge. I think what they are trying to say is they are getting a lot of "neg pressure" that results from "exaust flow suction" and a closed intake valve.....this would result in better scavenging and less residual exaust gas in the chamber...and better filling of the intake charge.

I can't believe though that it would work....the exaust flow wouldn't continue when there is nowhere for the flow to come from....expansion of the hot gases would only account for so much........I would really need to see this working to believe that it is this simple???

I think the Renesis zero overlap is more to do with lowering HC emissions than anything else.

I do think though that it gives credence though to stiffle the " bigger is better" header designs that result in low exuast gas flow...

What the hell do I know though...there are a lot more qualified people on here than me :D:

Any pressure below atmospheric is a negative pressure or vacuum. Scavenging is the result of moving gases pulling other gases, either exhaust pulling the intake charge into the engine (when there is overlap) or the exhaust gas already in the header pulling the exhaust gases out of the engine.

One reason overlap is beneficial even at partial throttle, is that the intake creates a bigger vacuum at partial throttle openings, which can actually draw the residual gases back into the intake. Residuals are very harmful to torque production. Without overlap, you can't hope to get all of the residuals out of the combustion chamber. When using overlap, its better to have the exhaust closer later to increase overlap than to open the intake earlier.

The problem is that there is a distinct point at which the scavenging effect of overlap results in less residuals than no overlap. Unfortunately, that point is higher than idle, which is where most emissions are measured. Also, uncombusted fuel is worse for emissions and the cat. (ie. blowing down with overlap) than burned exhaust gases.

I was cleaning up today and found a German book about rotary engines that I bought about 15 years ago and thought I lost. It has the title "Rotationskolben-Verbrennungsmotoren" and was written by W. D. Bensinger - the guy that lead the rotary engine development at Daimler-Benz (I guess this guy should have some credibility - after all he lived well before the stock option and severance package era, possibly during a time when managers actually cared about what they were working on :) ).
Anyway on page 75 he writes extensively about the difference between peripheral ports and side ports. He notes that due to the large overlap the incoming almost continous gas stream at full throttle, 'pushes' the exhaust gases back and therefore allows a much better filling than with side ports and less or no overlap (which is obvious since with no overlap the exhaust gases, even if pushed back, have no-where to go since the exhaust port is obviously closed). However he also notes that peripheral ports have a significant disadvantage at partial throttle; a significant amount of exhaust gases can re-enter the intake and this can lead to a misfire. If this occurs the unburnt gases get recycled and eventually ignited again which can lead to a ignition every 3rd, 6th or even 9th revolution and can lead to an unpleasant jerking during cruising (a 'phenomena' piston engines don't have to deal with).

Bridgeport and peripheral ports have always made more low end and midrange power than standard side port 13B engines. The key to that however is at what load. At higher loads this is true. At low loads the increased overlap kills low end power. Gas mileage suffers and the engine sputters and bucks around. Load is everything. The Renesis with it's lack of overlap makes more power everywhere than the last 13B (that has overlap). It takes some decent sized porting and overlap to make a 13B generate the power the Renesis does. Lack of overlap is apparently either misunderstood or underestimated because it seems to do very well.

yep, again it's a whole different game folks, the old rules aren't always going to apply

Bridgeport and peripheral ports have always made more low end and midrange power than standard side port 13B engines. The key to that however is at what load. At higher loads this is true. At low loads the increased overlap kills low end power. Gas mileage suffers and the engine sputters and bucks around. Load is everything.
Actually the book doesn't state anything else. However it especially points out the difference between partial and full throttle than low and high rpm.

Again what I tried to point out is that there is indeed excellent scavenging with overlap (and more so than with piston engines) and that if you have overlap a header will lead to significant power gains. Bensinger specifically states that peripheral ports with their large overlap, exhaust and intake tuning will lead to significant power gains. And frankly the 26B is a very nice example where this was proven.

And again it's not about comparing the Renesis with the 13B and whether overlap is better or not; it's about whether the Renesis can significantly benefit from a header design. So far no-one was able to come up with a header that leads to significant power gains and here's obviously a very reasonable explanation.

yep, again it's a whole different game folks, the old rules aren't always going to apply
I'm sorry but this is simply ignorant, there is no different game. The Renesis is not a new engine concept. This design has been considered by NSU over 40 years ago already. Unfortunately materials back then weren't strong enough to allow side exhaust ports.
Mechanical design of piston and rotary engines hasn't conceptually changed during the last decades. What mainly improved engines is engine management, better materials and better tools.
There is a reason why internal combustion engine books originally printed in the 60's still sell. The internal combustion engine is about 150 years old and it doesn't deal with any thermodynamical or mechanical laws that weren't discovered over a 100 years ago already.

I know all that, you simply misunderstood my intention, some of your own comments in this thread are less than enlightening if judged on the surface, it goes back to what I said before about the Renesis not being easily modeled using the conventional techniques

I'm not saying the Renesis header is easily modelable. Although it's not as complicated as some suggest as long as you strictly keep both rotors seperate (strictly divide the siamese port).
What I'm saying though is that a Renesis with header has a maximum power-gain, which is lower than what another engine with overlap would gain from a header.

well maybe I'm not so ignorant afterall, this didn't get copied over from the other thread:

http://www.rx8club.com/showpost.php?p=1171409&postcount=178

I agree with your statement regarding the 3-1 and 4-2-1 header.

(Actually I'm mainly questioning what maximum gains with any header can be reached and not necessarily what header concept should be applied and how they should be designed.)

So RE Amemiya's idea of splitting the siamese port is a good one?

http://www.rx8club.com/attachment.php?attachmentid=67581&stc=1

they don't come with that plate, I have one

So whats up with that pic? Weird yours does not have this feature.

So RE Amemiya's idea of splitting the siamese port is a good one?
If you want to keep it relatively simple from a modeling standpoint you should split the middle pipe into 2 pipes and then reconnect these 2 pipes with the corresponding pipe on the left and on the right shortly after the split (similar to what TeamRX8 suggested earlier).
Keep in mind the engine has only 2 exhaust chambers (and not 3, as a 3 pipe header design might suggest).

bump

I'm actually wondering now if the Renesis even responds much, if at all, by tube diameter and length like most 4 cycle engines. Maybe Mazda has it right and the best performance option might just be short pipes into a log collector built for maximum weight reduction, use 22 ga inconel 625 and it wouldn't hardly weigh anything at all, especially if you lightened the main flange up some more

what about separating the centre port and running a dual exhaust?. A 2 into 1 with the individual ports and a separate pipe for the Siamesed ones.

With the center ports not seperated, they will cause interference with each other. The idea is to keep the branches that will be flowing at opposite times seperate initially. Interference is detrimental to flow.

Running a completely seperate center port header would double the weight of the exhaust, offsetting any gains that it wouldn't make anyway.

Separating the center port into 2 ports and making a 4-1 collector has already been tried by Racing Beat. They didn't get any more power out of it than the standard 3 pipe system.

With the center ports not seperated, they will cause interference with each other. The idea is to keep the branches that will be flowing at opposite times seperate initially. Interference is detrimental to flow.

Running a completely seperate center port header would double the weight of the exhaust, offsetting any gains that it wouldn't make anyway.

it dosn't need to weigh twice as much as you have smaller diameter pipes.

I was thinking that the centre ports are a basket case. Why bother tuning them just reduce their effect on the other ports by keeping them seperate and hopfully give them better flow.

Well, basic header theory says to eliminate interference first. Interference is eliminated by joining pipes that flow at the same times first, then join those pipes together. Hence, the typical 4->2->1 header on 4 cyl engines joins 1 and 4 together and 2 and 3 together, before merging into one pipe. If you joined 1 and 2 together and 3 and 4 together, you will create interference, meaning some of the exhaust gases from on cylinder will go into another, instead of out the tailpipe.

Of course, this is a rotary, and we have to deal with that siamesed port. My theory is to simply block off the center ports entirely, and have a 2->1 header. The siamesed port is going to have interference, but if its entirely blocked off, it doesn't really matter. The only issue then would be if the side ports could flow enough. I think the reason we haven't seen gains on the headers that divide the center ports with a little plate, is that the center port isn't flowing much anyway. Just my thoughts.

Holy cow. I was just about to ask what would happen if we plugged the center ports.

yay for a guy who doesn't know what he's talking about :D

On the off chance that the side ports can't flow enough, is it possible to widen them?

What I'm suggesting is that possibly all that theory is out the window on this engine. Making a tube header is no big deal for manufacturers now. In fact, it's quite common.

Why would Mazda not make a true tube header if it had something to offer? Maybe the question is the answer.

What do you mean by a tube header?

As opposed to sand cast.

Edit: I should note that the stock manifold is not cast but is a dual walled "tube" header. I hope I have not mislead.

I remember seeing something about this (center port topic) in another thread.... Basically, the ports won't flow enough exhaust gasses to really give any good gains. The exhaust has to go through two 90* bends before it gets to anything, or something to that effect. There was a guy that ported out the exhaust ports a little bit, and made the transition a lot smoother. I bet that setup would benefit from a good header a lot more than the stock situation.

Basically, the ports won't flow enough exhaust gasses to really give any good gains. The exhaust has to go through two 90* bends before it gets to anything, or something to that effect.
No that's not what it is. Since there's no overlap there's no scavenging and therefore any scavenging supporting measures such as headers won't do as much as they would normally do (just read the whole thread).
If the 90 degree bend was indeed a significant issue than pretty much all conventional engines with poppet valves (which are obviously even more restrictive) wouldn't gain anything from headers.

Flow through those ports definitely has a big effect on what an exhaust does. I don't see how any decent piston engine even with a valve in the way can flow much worse than the stock Renesis exhaust ports. It's that bad.

The stock manifold is actually not to bad.
Some guys tried a few ways here,not exactly sure what,but no increase in power.I think it was still 3 pipes.The place said they will need a lot of time to try all the setups and dyno to see, but I dont think much will be saved / gained. AND

....Thats why I love a peripheral port - one hole in and one hole out, Cant be simpler he

Flow through those ports definitely has a big effect on what an exhaust does. I don't see how any decent piston engine even with a valve in the way can flow much worse than the stock Renesis exhaust ports. It's that bad.
The point is that most piston engines have a 90 degree bend and a valve (with a very high friction coefficient at that) obstructing the flow and many of them still get about 10% or even more with a header.

So far no-one has come up with a header for the RX-8 that delivers 10% power increase.
One explanation for this phenomena is: The Renesis doesn't have overlap.
and another explanation is: All the guys that came up with RX-8 headers simply weren't the brightest bulbs. Even if this was true one has to admit that even a blind man may sometimes hit the mark. And so far many have tried already.

That still doesn't mean that a peripheral port won't flow better, it just means that headers don't lead to the gains that many people would like to see.

RG, have you flow-benched it?

What are you going to do?

RG, have you flow-benched it?Nope. Played with them apart and know someone who has flowed it but I personally haven't.

so what's the deal with the port openings in the gasket being so much larger than the actual ports?

That's ridiculous isn't it! No clue what the logic is there.

I was wondering if it's an intentional temperature buffer for the gasket sealing edges ... :dunno:

That actually makes a lot of sense, to get the gasket edges away from the very hot gases. I'm not sure what it does to flow tho, but I imagine its not so great.

The gaskets are always larger to keep them away from the direct heat. what's strange about this engine is that the hole to the manifold is larger bya substantial amount over the exhaust ports in the engine and the gasked is much larger than that. It's realy odd and doesn't make much sense. If you could see the shape of the manifold and of the ports in the engine all put together, you'd wonder how it flows anything at all.

if anyone has the actual dimensions of housing exhaust ports that would help me out a lot, I don't have direct access to measure them myself at the moment, the end housing ports are wider than the center one

if anyone has the actual dimensions of housing exhaust ports that would help me out a lot, I don't have direct access to measure them myself at the moment, the end housing ports are wider than the center one


:dunno:

I could give you a tracing of the actual port runner outlets on the housings but it wouldn't tell you much. The ports themselves have much less area and the transition to the runners is terrible. I have no doubt it would be possible to use a smaller exhaust runner machinged into a new sleeve and still pick up flow. The center runner is the opposite. While the transition is terrible, the area of the ports is far greater than the area of the runner. It's all a big mess.

I'm only looking for the actual height and width dimensions on the exhaust port outlets if you can be so kind please :)

The outer runners coming out of the side housings are 1 7/8" tall X 1 1/16" wide. The center runner out of the intermediate housing is 1 7/8" tall X 7/8" wide.

The size of the holes in the exhaust manifold are much larger than this and the gasket is far larger still but these are the relevant ones.

Thanks, RG, I owe you one.

I'm just trying to finalize the flanges for the custom manifold so I can get them waterjetted. I'm not allowed to touch the ports so there's nothing I can do about the rest of the situation. I've pretty much finalized the design, wwon't be much more than lightweight knockoff of the OE manifold with some performance improvements, but still emissions compliant.

You do know that I have the flanges drawn up in CAD with the proper sized exhaust ports don't you?

no :banghead: :spank: :rant: :uhh: :bottom: :uh: :wtf: :Eyecrazy:


AutoCAD? :dunno: TeamMazdaRX8@aol.com :)

I did an eMachineShop print just to see what they'd charge, glad it won't cost me that

http://www.rx8club.com/attachment.php?attachmentid=70507&stc=1

I have it drawn up as a 3 piece flange much like you do. The outer flanges are the same just reversed and obviously there is a center one. It's less metal this way which is cheaper and it is less likely to warp over the length of the flange when you weld it.

Don't be coy, what's it going to cost me? :)

Sorry I got offline and then I came 50 miles across town to my girlfriend's house. I'll be home tomorrow.

Change your mind? I figured you might like some free flanges or something. I work for a stainless fabrication manufacturer, we have a waterjet ...

so basically, there are no marginal gains by developing a header for the renesis? So what's the weight differences between the stock one and the one being developed here?

I take it this is just a weight savings mod more then a power adder. Why don't you just yank out the a/c if that's the main objective?

that's on the list too, just got the SpeedSource pulley kit which has no AC drive pulley

:worship: ^^^^Like that way of thinking.

I need to resize the images that I have in picture form so everyone can see them here. As far as the format they are in on CAD, they are a .prt file. See if you can open that. If so, I'll send them to you. If not there isn't much I can do other than going back and doing it all over again. :(

I'm pretty sure I can convert that into an appropriate format :)

What is your e-mail address?

TeamMazdaRX8@aol.com

e-mail sent.

got it, thanks :)

anyone ever think of a 3-2-1 header design?

I'm not quite sure what you mean. That still looks like a 3 into 1 header.

I've had the idea for several years now to collect the outermost runners at one point and then collect the center port farther downstream. I've never tried it even though I have the piping and the flange. I just don't have the car as a template.

Hey Team: How much did emachineshop want for those?

I don't remember exact number, but way too much

don't bother wasting your time, don't even bother doing anything more than putting the round pipe over the rectangular hole

RotaryGod, et al--
I've read through this entire thread and it's very informative, but there's something that's bugging me. This may be a dumb question, but I have to ask: Everyone says the Renesis has no overlap. I understand that this means the exhaust and intake ports for a single rotor are never open at the same time. But with the siamesed/shared center exhaust ports, is it possible that there is in fact an effective overlap from one rotor's exhaust port to the other rotor's intake port? Or does the basic geometry of the engine and its ports preclude this possibility?

Anyone? Bueller?
Thanks,
Bug

This looks like a good place to post this since PM's don't work that well with some people...LOL

This is what im making now...

I should have it dynoed sometime August.. gotta finish the cat-back first.

It will either be stainless steel or regular steel but ceramic coated... I'm leaning towards the ceramic coated to save some cash and engine bay temps.


Racing beat did a 4 to 2 to 1 I think or something simular to that... according to Rotarygod

The exhaust of the front rotor were together and the exhaust of the rear rotor were together.

So it looked like this... YY

I have not seen anyone making it like the way I'm doing it and I dont see anywhere that anyone has tried... or I just could not find it when I searched.

I'm putting the front rotor front exhaust with the rear rotor front exhaust and the front rotor rear exhaust with the rear rotor rear exhaust.. like a ..... W... get it?

Well, hope all goes well.

I'm not quite sure what you mean. That still looks like a 3 into 1 header.

I've had the idea for several years now to collect the outermost runners at one point and then collect the center port farther downstream. I've never tried it even though I have the piping and the flange. I just don't have the car as a template.

Hey Team: How much did emachineshop want for those?


Ours hits the dyno on Monday, but it's not your perfect version of this theory though, so it may not show either way.

bigger is better and all the other usual header theory is out the window on this application ...

She looks a heavy bugger though!

it was 18 & 20 Ga T321 material, the OE flange could be reproduced lighter, but it was expedient WRT to retaining the air injection function. This was my original intention, but I was short on time and had to make do on that one feature.

it weighed 8 lbs completed, the OE manifold is 17 lb; a 9 lb (53%) reduction even with the OE flange

the entire exhaust system with FIA ultra-flow racing cat converter, resonator, and dual tip mufflers out the back weighed only 33 lbs total, the entire OE exhaust weighs 79 lb total

very nice, dyno numbers, throttle response?

it was 18 & 20 Ga T321 material, the OE flange could be reproduced lighter, but it was expedient WRT to retaining the air injection function. This was my original intention, but I was short on time and had to make do on that one feature.

it weighed 8 lbs completed, the OE manifold is 17 lb; a 9 lb (53%) reduction even with the OE flange

the entire exhaust system with FIA ultra-flow racing cat converter, resonator, and dual tip mufflers out the back weighed only 33 lbs total, the entire OE exhaust weighs 79 lb total


Aye, the stock flange weighs a hell of a lot, milled ours out of 316Ti and saved about 2lb just there.
I thought ours was quite heavy until that, 79lb for a stock system? Ouch!

I will say one thing about the unequal length header setup though, it sounds fantastic.

they all sound fantastic

you wrap that Team?
Looks nice
olddragger

they all sound fantastic

Nah, it's got a little off-beat thrum to it, little different to the normal setup.

I wouldn't really know what the normal setup is for an RX-8 ;)

Well, here's one that claims to have a 32whp gain!


http://carview-img02.bmcdn.jp/carlife/images/UserCarPhoto/883426/p1.jpg

http://carview-img02.bmcdn.jp/carlife/images/UserCarPhoto/883426/p2.jpg

http://carview-img02.bmcdn.jp/carlife/images/UserCarPhoto/883426/p3.jpg

http://carview-img02.bmcdn.jp/carlife/images/UserCarPhoto/883426/p4.jpg

http://carview-img02.bmcdn.jp/carlife/images/UserCarPhoto/883426/p5.jpg

http://carview-img02.bmcdn.jp/carlife/images/UserCarPhoto/883426/p6.jpg

http://carview-img02.bmcdn.jp/carlife/images/UserCarPhoto/883426/p7.jpg

http://carview-img02.bmcdn.jp/carlife/images/UserCarPhoto/883426/p8.jpg


32HP gain with that exhaust manifold!

http://img230.imageshack.us/img230/6915/32rx3.jpg

Feel free to comment. :)

Oh, and subscribed so I can read your guys comments. :)

Chop

Chop?

Well, here's one that claims to have a 32whp gain!

Feel free to comment. :)


too funny :lol:

if you didnt have so many pics, i'd totally think they were photochopped. thats crazy looking.

speechless

speechless

+1

insane :eek: but I like it lol

i saw this thread and got the wrong idea and lauphed.... hahaa i suck

Seriously though ... lets say I want to keep the air injection thing.. what headers are available that give a noticible gain ..... at what cost?

It looks more like metal sculpture than something functional. I've always thought a long tube header may be the ticket.

Racing beat did a 4 to 2 to 1 I think or something simular to that... according to Rotarygod

To my knowledge they have never tried that setup. From what I heard all they had tried was a conventional 3-1 system or a divided center port to make it a 4-1 system. Each design had only 1 collector but they did play with pipe lengths. They never got more than about 3-4 hp for any different design.

While I'd like to think my ideas have merit, and I have no proof they do, from everything that I have heard about RB's experiments as well as the experiments from Speedsource, gains are extremely hard to find in a Renesis header and all info points to team being correct when he states it's about flow and not header design. Currently that is accurate.

Now saying all that, the best exhausts on naturally aspirated rotaries of the past (engines with overlap) have been long primary systems that collect at the rear of the car before the muffler. That is what this spaghetti header appears to be doing but the fact that the center is still one pipe and not 2 seems counter intuitive to this working. While I can't cry absolute bs as I can't prove without a shadow of a doubt that their dyno is fake, it does seem interesting to see that their base power number is 228 which is basically what everyone knows it to be close to. Their new top number states nothing about the state of tune. This chart implies that the only difference is the header. I see nothing to confirm that though. We also see no mention anywhere of temperature, or octane or type of fuel used, or correction factors. Speedsource has stated that the max they've seen out of the engine with all the tricks is about 255 and that is also consistent with the new top number in this dyno. While the beginning and end numbers are realistic for an engine measured at the crank, I do have questions in regards to what else was done to achieve the new result.

was that a chassis dyno or a wheel dyno?

hey team wats the verdict on your header?

was that a chassis dyno or a wheel dyno?

I'd be more inclined to believe it if it were a chassis dyno. I'll cry absolute bs if it's claimed rear wheel power as the engine can't hit that without raising redline significantly and doing a host of other things in order to make this possible.

I say you guys shoot ЯX-8 a pm and ask him.

hey team wats the verdict on your header?

http://www.rx8club.com/showpost.php?p=2327891&postcount=414

very impressive :)

the person who now owns my exhaust is fitting it on a turbo system, so you may be able to buy the exhaust manifold from him since it won't be used

oo ic. who is it?

team wat are ur opinions on venturi connectors? or anyone else for that matter?

o ps wat are the specs on ur car team at the time of that 220rwhp dyno?

bump

Here is Rmagic's design that was posted in the Super Autobacs thread. Hoping the gurus could chime in on the design.

ЯX-8 stated they made it for their track car and it produced 32hp on the dyno.

http://carview-img01.bmcdn.jp/carlife/images/UserDiary/8713353/P1.jpg

http://carview-img02.bmcdn.jp/carlife/images/UserCarPhoto/883426/p8.jpg

yeah, they've never made power claims on parts that didn't deliver for anyone else ...

REPOST!
pdxhak, go to page 7

REPOST!
pdxhak, go to page 7

I missed it so my bad. Repost police in the same thread now...sheesh :lol:

I missed it so my bad. Repost police in the same thread now...sheesh :lol:

Don't worry, I still love you. :grouphug:

Don't worry, I still love you. :grouphug:

LOL man that is great :lol: Or is it :squint:

So, has everyone stopped researching this or just said fuck it, its not worth it?

This thread is over 2 years old with no real conclusions... RG, did you get a chance to try any of the 4 different setups you talked about earlier?

Θ █ ∞ █ Θ
1a R1 1b 2b R2 2a
Above is an ascii diagram of how i understand our engine to be. R1 & 2 being the rotors and the 1a, 1b, 2a, 2b, being the side exhaust ports. So when R1 is letting out exhaust gasses is it letting it out 1a and b at the same time? If so is 2b completely close off as to not interfere, or does 1b and 2b overlap?

From what i have read it sounds almost impossible/impracticle to fully split the middle port. If i read correctly the middle port is also smaller? This would lead me to believe the fluid would be traveling faster in the middle port and almost constantly. so has anyone tried the following 3 setups:

Block the middle port and use a little larger piping on the two side ports (maybe even make the ports a little larger)

Bring the 1a and 1b together at equal lenght and run it next to the middle port for a little bit and then merge them. Basically a 3-2-1 setup. To me it would seem like goin from the 2-1 you would have pulses merging at the same time, so maybe you would want to make one that would delay or advace the middle port runner a little to provide a more consistant flow.

Last but not least make a 3-1 with 1a and 2a being equal length and the middle being shorter or longer to help pull the exhaust out of the outside ports faster. Thinking about it now i dont really think this one would work as it would end up pulling harder from a than it would b and you would have uneven flow anyway. So what about enlarging the siamesed port and blockin off the ends? that sounds really bad... but i dunno.

Block the middle port and use a little larger piping on the two side ports (maybe even make the ports a little larger)


there is VERY little that can be practically done to enlarge the ports.

there is VERY little that can be practically done to enlarge the ports.
Yeah thats why i put it in parenthesis, wasn't sure on that one... so what about the half dozen other things i listed?

RG, what do you think? Are any of these suggestions even a half decent idea?

I have the time and patients to make about 3 of them... what are the best bets, or most likey to work?

you should go spend your time on better things to spend your time on

I've never gotten around to trying any of the designs that I mentioned a few years ago. I basically did just say "fuck it". I hope someone does try it at some point though.

The problem with closing off the center ports altogether has nothing to do with the small center runner but rather with the port windows themselves. You lose 50% of yoru exhaust port. Just using larger outer runners won't give you back that port area you lost. Unfortunately there is so little that can be done to the exhaust ports that it's almost not worth touching them. IF the ports could be reworked successfully then there "might" be some potential left in them but again, it's not something I've kept pursuing. I may never get around to it, I may do it next week. You never know with me.

I haven't had chance to get to a dyno with a standard manifold yet, but we got 215 rwhp with our split manifold (outer ports equal length, centre port meets at a different length to keep it out of the way of the pulse tuning)

you should go spend your time on better things to spend your time on
Yeah, i hear what you're saying, but its this or go get drunk... personally i'm over the whole going and getting drunk all the time.

I've never gotten around to trying any of the designs that I mentioned a few years ago. I basically did just say "fuck it". I hope someone does try it at some point though.

The problem with closing off the center ports altogether has nothing to do with the small center runner but rather with the port windows themselves. You lose 50% of yoru exhaust port. Just using larger outer runners won't give you back that port area you lost. Unfortunately there is so little that can be done to the exhaust ports that it's almost not worth touching them. IF the ports could be reworked successfully then there "might" be some potential left in them but again, it's not something I've kept pursuing. I may never get around to it, I may do it next week. You never know with me.
Ok, yeah i was thinkin the same thing as far as the center port goes. So, what are your ideas? Maybe we can split the workload; you make 2 and i make 2!!! I take it the other 2 designs are more worthwile? I just need more info man.

I haven't had chance to get to a dyno with a standard manifold yet, but we got 215 rwhp with our split manifold (outer ports equal length, centre port meets at a different length to keep it out of the way of the pulse tuning)

This is the design I've been talking about for years!

This is the design I've been talking about for years!


Well try it, 'cause I've a suspicion it works!

It also sounds far less 'raspy' than any renesis I've yet heard as a bonus :)

This is the design I've been talking about for years!
ok, so the 3-2-1 design... i picture it as all 3 pipes starting the same dia. when the outter join together double the area, and contiune on with the middle the same size. then when the middle and larger pipe converge into a pipe with triple that of the origional pipe... does that sound about right?

So what pipe sizes and lengths are we talking about? I thought Team had the initial pipe dia. posted on here, but i cant find it now...

RG, would you happen to have a cad drawing of your design I could take a look at? That is if you are willing to share it...

I can come up with something simple for illustrative purposes but don't expect it to fit the car!

I've run 1 7/8" primaries, into 2.25" secondary, collected into 2.5" system just slighty further down.

That's how I'd do it initially. I don't see a reason to go to a 3" pipe when the total port area isn't even close to that.

2.5" is enough flow for anywhere up to about 270bhp or so anyway.

OK I took a few minutes to draw up my design in AutoCAD. I've converted it to PDF. Forgive the rotation. I'm just glad it linked! It gives an idea of what I'm wanting to do and I added a few notes to show my train of thought. The actual pipe length dimensions can and may very well end up outside of what I have listed when all is said and done. These are just my initial thoughts. Someone please feel free to build it, modify it, and either prove it does or doesn't work. I don't care either way. I just want to see it tried rather than someone to blatantly say "it doesn't work". You can't claim that until you have proven it doesn't. It might not but I'd be fine with that.

I have other ideas as well. One is a variation of this one but it would add another pipe that goes nowhere that gets capped off! I'll modify this drawing tomorrow to show what I mean.

The other 2 ideas are very simple. Just imagine separating the center port into 2 runners. Now you'd have a 4 runner header. Collect the front rotor pipes together into their own dedicated front rotor single pipe. Now collect the rear rotor pipes together in the same way. This leave 2 header pipes, 1 for each rotor. Now collect these 2 pipes right before the muffler. This would be a "long primary" system.

The last idea starts out the same way as the long primary but instead of collecting the 2 pipes before the muffler, don't collect them at all. Run them each through their own dedicated muffler. This is called a "true dual" system.

The last 2 ideas would be the simplest to test as they either work or don't work. The first 2 ideas could be modified with any number of pipe lengths and collector sizes and convergence angles. Just because 1 doesn't work right doesn't mean modifying it wouldn't work. It might. Those 2 would take some time to work out and in the end it may either be great or a whole lot of effort to learn what doesn't work!

Have at it world! If any of these work, remember where you heard about them. If none of them work, they were TeamRX8's idea!!! ;) j/k You know I respect you!

Here's idea #1 that I first proposed after SS7.

Really should try and dyno ours on a stock manifold if we get chance, see how much difference there is.

If you left the center ports separate, with basically zero pulse tuning, when a good pulse came through an outer port, couldn't you end up pulling exhaust back in from the center port? I don't see getting negative pressure in the chamber unless both ports are working together to pull the gases out.

I just caught a mistake in my notes on that drawing. Swap the pipe diameter sizes in the notes between colletors #1 and #2. I got them backwards. Collector #1 should be 2-1/2" and collector #2 should be 2-1/4". Oops.

If you left the center ports separate, with basically zero pulse tuning, when a good pulse came through an outer port, couldn't you end up pulling exhaust back in from the center port? I don't see getting negative pressure in the chamber unless both ports are working together to pull the gases out.

Since the ports from each rotor open and close at the same time as each other, that's not an issue. The whole point of collecting them at a different length is to gain an acoustic benefit that you don't get at equal length. There is nearly twice the energy in the center port. There needs to be nearly twice the runner length to acoustically bring it back into balance. It's a bit more complex than that but that's the basic rundown of it.

You know more about this than I do so you're probably right. It's just hard to imagine it working any other way. The gases through the center port may have enough velocity to fight being pulled back in, but it seems like it would be much easier for the pulse to pull from the center port rather than to cause negative pressure in the chamber. I'd like to see this actually put into use to see how it works.

It seems like the center exhaust sleeve can be widened a good amount which may make separating the two ports more effective. Are the sleeve walls solid to allow this?

The pulses through the runners are the same as through any other system. There is no delay between the center or the outer ports as all of the ports open at the same time in relation to rear rotor vs front rotor. All we are doing with these pipe lengths is to change the resonance point within each pipe. An equal length system with 3 pipes has 2 pipes in harmony but one that is out. That's stock. This way we have all 3 in harmony. It may not do anything though as 4 equal length pipes has been tried with little success and this is more or less a way to trick the engine into thinking it is a 4 runner setup.

You can never have a time when one port tries to pull other gasses back into it. In order for that to happen we'd have to have the engine on the suction side through one of the ports and that's not possible. When there is a wave travelling down the front port runner, there's the same wave travelling down the center runner too. When theres a wave travelling down the rear port runner, there's the same wave travelling down the center port too. The center runner sees twice the numbers of waves travelling down it as the other port runners do. Each wave travels down each runner the same distance over the same amount of time. Look at my diagram. From the engine to collector #1 is the same distance no matter what pipe you travel down. There is no opportunity for anything to get sucked backwards and the collector itself prevents this anyways due to it's shape.

The sleeve walls are very thin. In pictures it looks like the sleeve is thick but that's just a flange. In order to widen the runner you'd need to physically grind down the center plate itself. While it may sound like no big deal, you'd hit the water jacket as there isn't much room. It kind of sucks that there so little that can be done.

That makes sense.

With the center port, would it be possible to block that coolant passage similar to what's done on the intake side with big bridges and pports. Obviously you would have to use something capable of withstanding the heat. That would give you more room to play with. Maybe I'm just dreaming. I just don't like to settle with mediocre.

I have already tried unsuccessfully to cut into and then repair the water jacket with no success.

Thanks mazda

So, has everyone stopped researching this or just said fuck it, its not worth it?



Θ █ ∞ █ Θ
1a R1 1b 2b R2 2a
Above is an ascii diagram of how i understand our engine to be. R1 & 2 being the rotors and the 1a, 1b, 2a, 2b, being the side exhaust ports. .


Im still working on my header, 1a and 2b will be together and 1b to 2a will be together. Still messing with the lengths but the dyno will let me know if it works or not. Kind of a W design. It will be 4 - 2 - 1. I prefer to look at actual test results to show me that things do not work.:)

Ben

P.S. I wanted to add that it will b 4-2-1 to go to the cat, but without the cat I will have a midpipe and exhaust to match the header to keep it dual to the back with seperate mufflers simular to the helix look. The header and the cat-back will have adapters if you want to keep the cat. Also it will have the flanges to mate up to the factory cat. But the adaptors are sleeve type and secured together with bolts on a bracket. So if someone wants to keep the cat they can and if they want to swap it out real quick it will be easy. The adapters stay bolted to the cat, and the two mid-pipes go between the dual header and dual cat back. Just trying to keep things simple and universal. I may have the prototype done by seven stock and if it is I will have it dyno tested and on display. "IF" it is done the car will be up on jacks and I will have a crawler so anyone can see how it looks on the car.

My site will be open I hope by this friday, if not next week for sure. I will be adding RX-8 custom products as I make them. Some of the custom RX-7 stuff on my site will give you an idea of what I have comming out for the RX-8.

An ETA on the header back exhaust system, I will start making it next week and I will have It at sevenstock for show with dyno results.

I will also have free T-shirts and pens, so look for me Ben with Rotary Extreme.

Good luck Ben! Any ETAs on any of that?

RG - ok the only thing left is lenght of the pipes. I haven't been under my hood or the car its self a whole lot lately and dont really know how much room we have to pay with... so are you thinkin header that automatically bypasses the cat or making it so you can still have a cat for those emission state guys?

Somewhere I read that the optimum length before combining the tubes, on a 13b, was 118 inches. I bought a exhaust system, for my race car and the tubes came together just in front of rear axle. This was on a 1st gen 7. This was purchased, in the late 80's, from a company called Rotary Engineering in California. I don't think it is the same RE that is in existance today.
I think you are going to end up finding that a long tube header system will make the most gains. Nobody has tried this yet.

Well, I got delayed paper work wise in re-opening the business, but I'm working on getting the rx-7 stuff back into production. I hope to have some RX-8 stuff ready within 2 to 3 months. But I do not think anything will be ready for display at seven stock.

Sounds good ben... does anyoe know the velocity of the exhaust gasses? averge, peak, or what ever?

RG - ok the only thing left is lenght of the pipes. I haven't been under my hood or the car its self a whole lot lately and dont really know how much room we have to pay with... so are you thinkin header that automatically bypasses the cat or making it so you can still have a cat for those emission state guys?
Cat? What's that? I don't even have a cat on my daily driven 2000 Civic! Screw the EPA! My mileage is far more imporant as it and not the air directly affects my wallet. I'm an environmentalists only when it's cheaper than not being one which isn't very often. You won't be able to fit a cat under the car with any of my designs. I'm aiming at power improvement if that's even possible, not emissions compliance. It's for "off road use only"!

Somewhere I read that the optimum length before combining the tubes, on a 13b, was 118 inches. I bought a exhaust system, for my race car and the tubes came together just in front of rear axle. This was on a 1st gen 7. This was purchased, in the late 80's, from a company called Rotary Engineering in California. I don't think it is the same RE that is in existance today.
I think you are going to end up finding that a long tube header system will make the most gains. Nobody has tried this yet.
This is one of the ideas I mentioned in post #205. It's called a "long primary" system and is by far my favorite 13B exhaust system.

Cat? What's that? I don't even have a cat on my daily driven 2000 Civic! Screw the EPA! My mileage is far more imporant as it and not the air directly affects my wallet. I'm an environmentalists only when it's cheaper than not being one which isn't very often. You won't be able to fit a cat under the car with any of my designs. I'm aiming at power improvement if that's even possible, not emissions compliance. It's for "off road use only"!
Hahahaha, fair enough... with the smaller area of the middle port i am thinking the middle runner should be the smaller pipe and be 2.5 times longer... but i still need to try running some numbers... thinking the total volume should be teh equal... Thoughts?

interesting...

Hahahaha, fair enough... with the smaller area of the middle port i am thinking the middle runner should be the smaller pipe and be 2.5 times longer... but i still need to try running some numbers... thinking the total volume should be teh equal... Thoughts?


Don't forget you've got two rotors worth of gas exiting there though, there's some serious velocity in the centre ports, and a fair bit of turbulance sending it into the walls of the header.

Don't forget you've got two rotors worth of gas exiting there though, there's some serious velocity in the centre ports, and a fair bit of turbulance sending it into the walls of the header.
uh, yeah... smaller area = higher velocity... what are you getting at or suggesting?

Although the port area is smaller your might want a slightly larger header pipe anyway to help prevent reversion from turbulance from the shape of the centre ports.

wish someone would try RGs idea - maybe in another 5 yrs or so ......

What exactly do you mean by turbulance? From my understanding of fluid dynamics (which is quite limited), the gas will enter the smaller pipe at the same velocity as its leavin the port causing a more consitant flow... if you go instantly to a larger pipe than port size you will lose some velocity due to the venturi effect.

correct me if i'm wrong, but mazda made the center port smaller for a reason... not sure why, but i would think the higher velocity would have something to do with it...

wish someone would try RGs idea - maybe in another 5 yrs or so ......
yup, got measurements today! hoping to pound out some type of protype in the next 2-3 weeks...

Although the port area is smaller your might want a slightly larger header pipe anyway to help prevent reversion from turbulance from the shape of the centre ports.

If you could see a port in person you'd see that they've pretty much got this feature built into them. There is a huge lip on the bottom of the exhaust port that you can not get rid of without cutting into the water jacket and then repairing. I have yet to be successful at repairing this and I've tried. The damn cast iron like to crack when welding. Even when welding slowly with pre and post heating of the part. It sucks.

correct me if i'm wrong, but mazda made the center port smaller for a reason... not sure why, but i would think the higher velocity would have something to do with it...
They actually did it for a very simple and non scientific reason. Space. There isn't much and they didn't widen the center housing to make more. They just made due with what little area they had. I wish it were for some other good reason but it's not.

speacking of 5yrs or so when the F is RB gonna publish the release of thier header,,, I ve emailed them over the last two years about this and we keep get developement dates, Ithink the last Iheard was the end of July so july has passed and anyone else hear anything...

They actually did it for a very simple and non scientific reason. Space. There isn't much and they didn't widen the center housing to make more. They just made due with what little area they had. I wish it were for some other good reason but it's not.
ah, ok, so all pipes off the header will be the same size... kinda shitty reason tho :(

The center port could be one size smaller but once you take a round pipe and form it into a rectangle, you end up using too large of a pipe. The proper way to do it would be to machine a piece that transitioned the shape for you but maintained the area of the runner. This is what RB is doing with their flange and they did it on their Miata header too.

The center port could be one size smaller but once you take a round pipe and form it into a rectangle, you end up using too large of a pipe. The proper way to do it would be to machine a piece that transitioned the shape for you but maintained the area of the runner. This is what RB is doing with their flange and they did it on their Miata header too.

The Renesis has no relevance to the MX5 engine. With zero overlap it's a wasted effort except for marketing to people who really don't understand what's going on with this engine ...

We live in America --Team.
Headers may help a little with us FI guys--proper headers that is.
Now you can get a little with na headers but it will be too expensive for most. And for those it is not--they already have the means.
OD

If you could see a port in person you'd see that they've pretty much got this feature built into them.

You absolutely must prevent as much reversion from the centre port as you can though - otherwise it compromises any benefit you gain from the pulse tuning on the outer ports, given they are open at the same time - and we're not talking huge gains to start with.
So, a step to interupt any reverse flow is beneficial so and then slowly transition to the correct size pipe to keep gas flow speed up for the same reasons, the step won't harm the centre port flow much as the main disruption is in the port itself, the low-pressure wave it would create would have little effect as it would be too high frequency, and centre pipe gasses still have huge amounts of energy even 4-5ft down the system in the LCB header so plenty of energy to pick the gas speed back up with.

I can't see what you're grumbling about Team - no one sells one, so how can marketing come into it?

Team sounds a little miffed off... so let me ask you guys this... what are they doing for the RX8s that they race?

The Renesis has no relevance to the MX5 engine.

I hate to say it but the rotary is nothing special in relation to any other engine. Each engine works the exact same way using the exact same principles following the exact same airflow rules and the exact same laws of thermodynamics. Whether it has a piston or rotor is irrelevant as air and heat have no idea which is which. They are each 4 cycle internal combustion engines which makes them the EXACT same thing! Saying that, the thing that matters is the lack of port overlap. Maybe nothing will work better than anything so far but don't be so quick to claim nothing will. You're a smart guy which is why I've always been confused about your absolute certainty that nothing can do better even though you haven't tried it. You can not make that claim with ANY degree of certainty until you've tried everything and directly proven them to not work. Until everything is tried and disproven (and it all may very well be disproven), there is always hope and the possibility that something may give a positive result. You shouldn't give up until you've tried everything you can think of. Don't be satisfied with mediocrity. Just remember, innovation isn't usually accomplished by those who reach definitive conclusions based on partial information.

You absolutely must prevent as much reversion from the centre port as you can though - otherwise it compromises any benefit you gain from the pulse tuning on the outer ports, given they are open at the same time - and we're not talking huge gains to start with.
It's already compromised in relation to the outer ports.



So, a step to interupt any reverse flow is beneficial so and then slowly transition to the correct size pipe to keep gas flow speed up for the same reasons, the step won't harm the centre port flow much as the main disruption is in the port itself, the low-pressure wave it would create would have little effect as it would be too high frequency, and centre pipe gasses still have huge amounts of energy even 4-5ft down the system in the LCB header so plenty of energy to pick the gas speed back up with.
I found on 13B's that the most important thing in the exhaust was velocity. Anti-reversion is a good thing but not if it means using a larger pipe than necessary. You can flare out the end of a pipe or get creative in other ways to accomplish the same thing but just using a larger pipe isn't a good idea. It's a compromise. You are reducing one benefit to gain another.

The other large difference between the MX5 engine and this besides the overlap is the exhaust pulse itself... piston pulses or even peripheral port pulses are much more sharply shaped then renesis pulses (due to the relatively slow opening and closing of the port) and dont have a giant hunk of interference in the port leading to more effective resonance tuning

I found on 13B's that the most important thing in the exhaust was velocity. Anti-reversion is a good thing but not if it means using a larger pipe than necessary. You can flare out the end of a pipe or get creative in other ways to accomplish the same thing but just using a larger pipe isn't a good idea. It's a compromise. You are reducing one benefit to gain another.

Hence bringing the pipe size back down afterwards :)

Team sounds a little miffed off... so let me ask you guys this... what are they doing for the RX8s that they race?

I only get miffed when n00bs spout off without doing a proper search, and hence are completely uninformed about what they're spouting off about

WRT the rest of the replies, I rest my case ...

I only get miffed when n00bs spout off without doing a proper search, and hence are completely uninformed about what they're spouting off about

WRT the rest of the replies, I rest my case ...
n00bs... since when is RG a n00b? :icon_no2:

kind of a general statement that doesn't target the issue...

anybody can talk the talk, you must be greener than fresh grass since you're not aware that I'm one of the exclusive few here who has walked the walk with posted the results to back it up ...

anybody can talk the talk, you must be greener than fresh grass since you're not aware that I'm one of the exclusive few here who has walked the walk with posted the results to back it up ...
yeah i pretty much am... maybe you can point me in the directions of your results :beer05:

I would search for it, but i'm not exactly sure what i would be searching for...

Look up his header thread. Neat design. It's a very easy and simple 50+ year old design that's still common on hotrods today but then again my proposal is nearly that old too as siamesed ports were common on engines back then. Everything seems to come back around again at some point. In many ways people were more creative back then.

You only need results to back up things when you make claims. I state ideas complete with disclaimers that it may or may not work, tell people how I'd try it and then hope someone who is interested in potentially learning something actually does try it. If it works, great. If not, that's still good info too as you can definitively say it doesn't work. I am not a person that will discount every single potential design idea before anyone has even tried it. I'm well past the point where I have anything to prove to anyone anyways. Besides, I make no claims. I just share info and my train of thought with the challenge to others to either prove me right or WRONG! I'm cool with it either way.

I want everyone to learn. To do this I'd like to see others get involved in the process which means everyone should have the information to experiment with ideas that they may not have thought of should they choose to. I want multiple brains to think about these things as there may be different perspectives on a topic. I think of rotary knowledge kind of like open source software. At least it should be. It's shareware and everyone should be able to expand on, play with, improve, disprove, alter, aspects of it as they choose and then share the results with the whole world. It should be well understood that ideas are just that and nothing more. There are no promises made with those ideas. Theory is fine. Hunches are fine. Proof is what happens when someone does it. Only then is it a conclusion where a claim can be made. I do think general trends can start to be seen over time though and up to this point we definitely have an exhaust trend to follow.

Well, it certainly doesn't hurt power going by our results, just need to get hold of a stock manifold for comparison.

Well, it certainly doesn't hurt power going by our results, just need to get hold of a stock manifold for comparison.
So i take it you already have one made? Whats it look like... pics plz!

So i take it you already have one made? Whats it look like... pics plz!

It won't fit in an RX-8 engine bay as it's in an off-road buggy, sorry :lol: