You and me both. I can't wait to show you the results of my project (especially the headers). I learned a lot from your post over the years about NA set-ups. I'm not going to say anything but you did an experiment a few yrs back involving sleeve inserts. That should ring a bell. With my exhaust, I fabricated everything from the exhaust flange back. I even hand made my own merge collector w/megaphone.

 

once you figure out how all that can be adapted to account for the differences between a rotary and reciprocating engine exhaust dynamics then you'll be on to something, then there are the dynamic differences between rotary engines to account for; peripheral w/overlap vs side exhaust port w/o overlap

 

That is basically a true dual setup as each pipe effectively ends at the box. You'd just have a very short true dual. This may or may not be good for power. About the only time you ever see any cars with non collected header pipes (that is basically what this is) is on top fuel dragsters that are using large roots blowers. Keep in mind true duals does work on rotaries just fine but typically they are much longer. Try it. Maybe it will work. There's one way to find out!

 

I've said it before, they are both 4 stroke engines. That's all that matters. The only difference in dynamics comes in the main header pipes themselves but even then the principles are the same. A rotary engine with lots of overlap or no overlap is no different than a piston engine with lots of overlap or no overlap. A 2 stroke engine is different though but these aren't 2 stroke engines and have absolutely nothing functionally in common with them. No the fact that a rotary has ports instead of valves doesn't matter either!

 

Rg aren't you forgetting about the frequency of events between both the individual exhaust ports of each engine? Yes both engines are 4 stroke but the rotary will have two exhaust events in the same exhaust runner while the piston is only having one. Shouldn't these differences make a difference?

 

Two-stroke engines are fundamentally different since for best power they require a timed pressure wave to be reflected back to the exhaust port to counteract the compression in the chamber.

2 cycle exhaust theory makes my head hurt. Rotaries aren't as bad, at least the exhaust port doesn't close halfway into the COMPRESSION stroke!

 

lol, I ran meth powered RC cars for a long time.... 2 stroke sucks.

 

You know I'm rethinking my post at the beginning of the thread, about people not disagreeing with this exhaust theory.

I'm kinda surprised no-one has chided in with the "straight through is the only way to make power!!!!" theory. It always seems that every time a new member asks a question about exhaust and they consider anything other than a straight through muffler they get pummeled with "straight through only" comments. Where are all these guys? Even after this thread started I am still seeing "straight-through all the way back" type posts. I'd like to hear their side of things.

Could it be they just haven't lurked here in the NA performance section lately??

 

I guess I wasn't completely clear in my last post. I realize that a system that rotarygod described could certainly be done using all straight through mufflers, given that one of the straight through mufflers had a big enough core to make the exhaust "see" the end of the system to tune it appropriately.

I guess what I mean is that a system as described by rotarygod could be built using one or more chambered mufflers, and perform just as well as a system using straight through mufflers. And most of the straight through only guys say that good power can't be made with a chambered muffler on a rotary.

I'm curious what people think about that idea. Any mentions of using a chambered muffler in the 1st gen section (where I usually hang out) is met with flames galore, and this info suggests that you could use them without performance loss.

 

The problem with chambered mufflers is that they typically do not last long at all. The internal walls break down quickly, or the case balloons or just plain explodes when the exhaust pops. I had one blow out in impressive style and my car never backfired, it just opened up.

 

The only thing against a chambered muffler being used is as peejay says which basically has to do with how well the muffler is constructed. A rotary is much harder on an exhaust system than others. I've seen chambered mufflers blow apart and I've seen straight throughs blow apart too. Edelbrock being a very good example. Typically an absorptive type of muffler (not chambered) will give the best sounding exhaust note.

I'm not saying to use a Flowmaster muffler under a car as an exhaust box. For that device I'd say build it or have it built. No packing. No chambers. Just a stainless steel box. That's it. It doesn't get much more reliable than that.

Since we have pipes and resonance at work with a 4 stroke engine feeding them, the theory will always still apply. The number of pulses through each pipe per time is a bit different though as was asked above but that only changes our calculations when using a math formula and even then only on the header primary pipes. We'd need to use a factor of 1.5 in the equations but we don't care about all that. Lengthening and shortening a pipe still has the same effect regardless of number of pulses or type of engine feeding it. Since we aren't concerned with math at this point, there is no need to worry about correction factors. It's irrelevant. The same principles still apply. Forget about what it is on paper and worry about what it does in the real world. That means experiment and test. You can debate numbers on paper all day but you can't debate real results. Years ago I tried to show how the math applied. Now I'm telling everyone to forget all that and not worry about it. Understand the fundamentals and then tweak from there. You'll be smarter for it and happier.

None of this is unique to only a piston engine. None of this is any less applicable to a rotary engine. I don't have to compensate for anything rotary wise to "be on to something". None of this is opinion. There is also no one right way to go about a design.

Many people want to be able to plug in some simple numbers in a formula, spit out a result, then go buy some pipe to build it exactly as the answer specified and have the highest performing exhaust possible. Not going to happen!

 

Ah! I see what you are saying. Thanks for the clarification. I guess I have just had chambered mufflers on the brain since crispin posted his flowmaster setup. It really sounds different than I expected. if you had an expansion chamber, theoretically the pulses after the chamber would be less intense, almost fully smoothed out and a flowmaster (maybe their stainless version) could work as a rear muffler to get a different sound, and it seems like in that way it wouldn't destroy the muffler.

Here is cripin's flowmaster on a 12a with a very large streetport. Sounds cool. Too loud for my taste, but that could be taken care of with another muffler.


http://www.youtube.com/watch?v=AxJvNWfroK8

 

On another note, parts to make an expansion chamber can be had here...

 

i've learned a lot about mufflers lately. for instance ALL of the factory mufflers are chambered. just because you can hear the flowmaster from 3 miles away doesn't mean chambered mufflers can't work...

i think if there is anything "unique" about the rotary is that its more sensitive to backpressure than a reciprocating piston engine. of course this depends on the amount of overlap you have...

 

i agree. I have been trying for a while to come up with a quiet performance setup, so this thread has been great for me, a real revelation. It has made me look at things in a completely different way. So I think rather than using a bunch of straight through resonators back to back and then a straight through muffler, I'm gonna use an exhaust box, (where I can change the distance from the collector, and after that a chambered muffler. As long as the chambered one will flow enough, I'll be fine. In fact it will hopefully be quieter than the multiple straight throughs, and probably make more power, since the mufflers won't be adding length to the collector.

 

I like Dynomax Super Turbo mufflers. They are fairly cheap, easily flow what the engine needs, and are pretty quiet. They aren't shiny chrome or stainless but they work well. The one I like is a round style with an offset 2-1/4" inlet and outlet. Obviously on a 2nd gen, use 2.

 

about them blowing up, here's my borla blown up and it was a straight through:

https://www.rx7club.com/showthread.p...ighlight=borla

 

a car craft issue:


Also some other numbers:


Just tossing this out there for fun I guess ... also kinda show's that CFM doesn't mean more power output.

 

Once you hit the required flow of the engine, any more cfm ability through the mufflers is a non issue and offers no benefit. Any power differences at that point are merely due to design differences that affect tuning in the way the muffler is seen to the system.

I'm surprised no one mentioned not to use fiberglass packed mufflers because the Dynomax that I like are fiberglass. The key is where they are. The farther back they are, the less heat there is. Racing Beat says not to use fiberglass packed mufflers as front mufflers which means not as presilencers. As rear mufflers they are fine. If you have a muffler that fails at the rear, it was just a poorly built muffler. Good material doesn't necessarily mean good construction.

 

Completely contrary to what most top end NASCAR builders profess. Or any other engine builders for that matter. There might be some who spend development $$$ looking for ways around the rules. The great ones spend an inordinate amount of money to maximize the engines contribution through each single cylinder. The common theory and philosophy is to make the perfect "one cylinder engine." Then develop the packaging to put these into an 8 cylinder unit. The Chevy SB2 was done this way, as has every product since.

It appears that you're describing torque loading and delivery peaks. Actually, through the transmission and final drive,and rear wheel inertia, it really becomes almost invisible or inconsequential to power delivery. It then is more of how much torque where. Flatter is usually better.

gd

 

Does anyone here believe in data? I've been in the R & D arena for a large automotive company in powertrain development for over 26 years. I never believed anything until I saw data. No disrespect intended, but I read, Trust me," and I get nervous. I work with the numbers and I live with the numbers, When someone says don't rely on the math, I stop right there.
With that said, Our business behind the collector was not tuning for power, only for resonance and pleasability. We did do some backpressure reduction valves for our sports car but that was about it. The major gains are in the primary and collector.
The power in the gasses are about all spent by the time they get to the collector exit. There isn't much more energy left to do any work. JUst handling the exit with no obstructions.

So let's see some numbers on a dyno sheet, or sheets in this case. Or , go through the motions on the math, I'll happily follow along , if I can. I'm open to new things!!

Please?

gd

 

There's that, and then there's the issue of how I experience better acceleration while in the "wrong gear" lugging the engine at 4k instead of running the engine in its actual powerband.

I know that anecdotes aren't statistics, but I always wondered why that was until I learned about the V-twin thing. Yes, drivetrain inertia plays a role, and yes, it's greatly dependent on being traction limited. In the gear I normally play in, one-half revolution of the engine is a bit over four and a half inches of travel. (2.12 trans gear, 4.09 final drive, 65cm tire) It's interesting to think about, at any rate.

BTW - Here's an interesting read. http://www.theoldone.com/archive/pro...ing-engine.htm

 

I agree gawdodirt. Numbers are king. But I think that there are 2 big reasons some of the people here avoid numbers and dynocharts in some instances:

Rotarygod was saying earlier that doing the math then claiming you've found the "perfect system" is not gonna happen. I think you can design a very good system... Too many variables in exhaust tuning makes it so basic math and then designing and trying it in several ways will eventually end up with a better system than doing the math, designing the system once, then sticking with it for good, especially for guys without access to a dyno.

And the other reason is, I think, that power claims don't always transition from car to car. You can find a legitimate looking test and dyno test for every competing muffler, header, etc., each company showing that their product outdoes another, each with dyno proof.

Also sometimes a member here will do a back to back dyno test with different gear, and make more power with some change, and a lot of people will try and copy that particular setup or change that guy made, because it's backed with a dyno numbers. They make the change on their cars, and it doesn't work as well for them.

I think that most all of us can agree that numbers and testing is the best and only way to make a valid point about something, but all things have to be taken into consideration. (I'm not saying that you, gawdodirt, are not seeing it this way, or that you are wrong, I mean that a lot of us don't) Rotarygod has been preaching that since heaven knows when, and I think that's one of the main points of this thread.

So I don't think that these guys are saying "forget the math, experimentation is best" or are trying to prove that this way is better. I think the point is, "each system is different, dyno sheets for one car unless the car is IDENTICAL to yours won't make YOUR car faster"

No-one said don't dyno test. It is the best way to know if your design worked. But if I make a header that works on my dellorto setup with a long intake manifold, and my buddy sees my dyno proof, and copies it, and trys the header on his car with a modded nikki carb and manifold, it will not be the same.

A lot of us tend to look for a good setup posted here and copy it, we need to be learning the concepts to a point where we can design something for ourselves. That seems to have been rotarygod's motto for years. Again gawdodirt, I'm not directing any criticism at you, I agree with what you are saying I'm just trying to explain the reason this thread came off the way it did.

 

That's pretty much it. When I had posted the math years ago, lots of people went right out and tried to build a system immediately after plugging in a few numbers. Others tried to figure out what they already had and then when the numbers didn't agree they started to question everything. There are many unknowns and the reality was that no one had all of the necessary information.

If I give you a formula to figure out exhaust header length and you use it and then find that your exhaust doesn't perform as you thought it would, would you have a problem with it? Would you consider the math faulty or inaccurate and then abandon it altogether? Most would. However maybe the problem isn't the math but with a lack of understanding as to what it applies to. Every math formula I've ever seen about runner length only takes into account the primary header pipe length and this is what people spend their time trying to play with. Go searching for threads describing who has the best header and you'll quickly see that this is what everyone thinks does all the work. Remember I said the primary header pipes are less than half of the total tuning. Could it be possible that the numbers don't agree with the results because the don't account for the rest of the system? You wouldn't have known that if you didn't understand what happens in the exhaust and only cared about the math. An understanding of what happens and what changes affect the system in what way should be the first thing you learn. Understanding the math should be last. Everyone wants to do it the other way around because they feel the math is infallible. If you get all the variables right, it is. I have yet to see anyone get all the variables right. I can't do it. Lots of it involves assumptions.

The perfect math problem isn't a set formula that can be given out. There are variables that may not be the same between the systems being built. To get a more accurate calculation some of the things that you need to know are the material being used. You need to know what it is and how thick it is so you can calculate it's heat loss over distance. You need to know outside ambient air pressure. This is always changing so even this is an estimate and the reason why we have SAE corrected numbers on dynos. You need to know the exhaust temps entering the pipes. This is always changing unless you specifically tune based solely on exhaust temps. This doesn't work on a rotary btw! This variable turns into an average guess.

You will need to have a formula for the primary header pipes which takes all of this into account and material heat rejection properties have to be a part of it. Then you have to have a formula for the effect of a collector whether it is a diverging cone or straight. It has to account for total collected area, length and have a tuning variable applied from your header output. Then you get into the tailpipe calculation. Again it needs to have a variable based off of the answer given from the collector. You need to know area and again heat rejection properties of the metal.

It's not impossible but with all the effort you may as well write software to simulate it. This has been done and it exists. Many good race teams use software like this. Even then the numbers get them close. It is ever exact and fine tuning still needs to be done at the track.

Now is it easier to learn all of this and sit there plugging it in with the hopes that you get close to what you want, or is it just easier to understand the concepts and what will be affected by those changes? We aren't shooting for the moon or mars here. This is a car exhaust system. It is also confined to a fairly small area. There is only so much that we can do and functionally only so much we need to understand to get great results. Why overthink things?

Everything can be explained with math. Is it really necessary though? In this case, no.

The main problem I see is that most people just want one person to figure everything out so they can copy them having never learned anything on their own. I disagree with this form of instant gratification and for that reason I will never provide a dyno to anything and will never make any power claims. All I will ever do it try to help people understand things and encourage them to do it on their own. I won't show them the answer. You don't have to believe me. That's your right. The more people that learn how and why things work the way they do, the more the community as a whole can benefit from it. After 8 years on the forum I'm still seeing the same threads and topics repeated over and over again that were asked back then. Too many want the answer without searching for it or learning it on their own. I don't want to see threads asking how to do something. I want to see threads from people showing off how they actually did something. Even a failure is something to learn from.

 

How could that be a straight through with the opposite pipe being offset?