coldfire

iTrader: (1)

I know the general rule of thumb is that having a relatively large change of volume (or cross-sectional area) in the exhuast system AFTER the header seconary length (i.e. after the collector) acts basically as a termination point for pressure waves (regardless of flow). i will assume this is the case for the rotary.

now what i am trying to figure out is what effect does the length between the collector and the pressure wave termination point (i.e. a resonator box/chamber) have on the power curve, and also what is the determination to classify a pressure wave termination point? it is generally accepted that a large enough empty chamber acts as a termination point, but a "straight-through" muffler such as a Magnaflow actually acts as an extension of the pipe length unless a termination point is placed before it. this would mean, for example, if you only had straight pipe or straight-through mufflers AFTER the collector then you would have no termination of pressure waves until they exit the rear of the exhaust system.
however i don't quite believe this because a straight-through muffler can't simply act exactly the same as straight pipe, and if it doesn't then it must have some effect on pressure waves.

i'm not quite sure if this is the right place to post this, but any relavant discussion is appreciated.

EDIT: i should have actually entitled this "secondary tuned length exhaust theory", that is the part of the exhaust AFTER the primary header lengths collect.
if there is any reference material for studies done on this (specific to the rotary engine), please point me to it.

coldfire

iTrader: (1)

or rather here is an easier question:
after the header collects, how much straight pipe is optimum before putting something that may terminate/disrupt the pressure wave like a muffler or resonator or cat? i know that is a blanket statement though the requires specifying what part of the power band is to be optimized, but still i can't find any information on secondary tuned length, only the primary tuned length...maybe i will post in the race section if there is no discussion here.

(BTW, this is all NA specific of course).

Dick Elliott

Look at the picture of the 79/85 exhaust system on the ICS web site. LOL as I've seen every known pipe layout (2 pipes/1 pipe/short pipe-pipes/long pipe-pipes/collected header/un-collected header) all go just as fast on the track, and in most cases, on the same day. Again I say LOL. I've seen Spec-7 cars (collected header) out run IT-A cars (un-collected header). Dyno HP will do you no good un-less you race on the salt. What ever works on one engine may kill the next 5. Hope you figure it out, but please tell me first.

coldfire

iTrader: (1)

can you give me the link to what you are referring to?

i've limited my discussion to a collected header system, and specifically the part of the exhaust system AFTER the collector.
all i want to know firstly is the effect of the length of pipe after the collector, if any on the rotary engine.

i've made good power with my current setup, but i am sure there is a way to make a quieter system that makes just as much power if not more. there seems to be a lot of literature on this for piston bangers but not too much for the rotary.

i'm redoing the exhaust system on my car for the 4th time now. each time as been an improvement but mainly excessive noise has always been an issue. i figure if i do a bit more research this time i won't have to redo the system again.
i've already got the calculations for resonance noise cancelling, which is a noise issue. but i figure i might as well do some calculations from a performance standpoint. the length of pipe after the collector is one possible point of optimization (and not just peak power either, i would gladly sacrifice a bit of peak power for gains across the rest of the power band), the other is the nature of the first muffler or resonator or whatever you place after the collector. generally everything after that point has no affect on exhaust tuning, just as long as it flows adequately.

but yeah, i am seeing this is not quite a defined science with straight answers...and i'm interested to see how much theory and calculation translates over on the actual car.

PercentSevenC

I've been told that 5.5" is a good length after the collector for piston cars, but I don't know if the same applies to rotaries. I imagine it would, though. But I don't know the physics behind it.

coldfire

iTrader: (1)

was that 5.5 inches for a particular engine?
because i have not heard a single length that works for all engines. i have heard 8 inches for a V8 and sometimes longer. it seems one rule is that the longer the length you have less gains on the high-end but even out the rest of the power band. however that was testing with a V8, so who knows...

i forget exactly the length i'm running after the collector right now, i think it's about 6 inches, maybe a bit longer. that seems to have worked ok so far so i will stick with that if i can't figure out something better.

Dick Elliott

http://www.iscracing.net/

coldfire

iTrader: (1)

thanks for that link. it's hard to tell from the pic on their website any exhaust lengths. it seems like for their 1st gen system they have a length of about 10 inches, but the 2nd gen system seems much shorter, maybe about 5 inches. their "expansion chamber concept" though is exactly the method i am using to design an exhaust system. the 3 main parts of the exhaust are:

- primary tuned length, which consists of the header itself

- secondary tuned length. between the collector and the "expansion chamber". the expansion chamber is just defined as a significant increase in volume/cross sectional area. too bad there are no specific numbers to determine how big the chamber actually needs to be

- everything after the expansion chamber. from that point on you are assuming that the pressure wave has been cancelled out to the point that you can't do any more tuning of it, you are simply worried now about sufficient flow (CFM) and noise reduction.


the average person can't do much about the first part of the exhaust system, you are basically stuck buying headers from a known company (i.e. Racing Beat), which i think is fine. trying to fabricate my own header is not feasible and that is a whole other can of worms...
so the points of optimization is increasing performance by proper tuning of the secondary length, and decreasing sound levels after the expansion chamber.
that is exhaust theory according to me so far, heh...

rotarygod

iTrader: (1)

Keep it simple. Too many people try to get overly complex and never get anywhere. All you need to worry about are whether you are collecting it or not, where you are collecting it (if you are) and what pipe diameter you are going to use. Don't get too involved in trying to figure out anything else as anything else is minor by comparison and more trouble than it's worth unless you are a huge unlimited budget race team with the facilities to test it all.

3 types of systems to consider. Long primary with pipes collected at the back of the car. Short primary with the pipes collected at a more conventional length and uncollected true dual.

Typically the larger the porting, ie more overlap you have, the less effective a true dual system will be. These systems generally work better with less agressive porting styles such as street engines. Long primary and short primary systems both have their advantages and disadvantages. A long primary will typically make more average power than a short primary BUT how you design the collector and bends can change this depending on how you compare it. There are so many things that come into play that it's not even funny. As a result any system can be made to perform well or badly on any setup engine. I personally feel the most important thing to be the collector design. It needs to be smooth.

You seem to be concerned with that happens with pipe length after a collector and my suggestion for this is to first play with some different setups along the line of the more conventional designs to see what works best. Then after you find one you like, adjust pipe lengths and even diamters after the collector to see what works. Any effect you get after the collector is a secondary benefit and as such should take a secondary priority to the rest of the system design. I'm not saying it isn't important. You should just make sure the rest of the system is optimized before you worry about this. Building a header isn't that hard to do. You can buy flanges and pipe. Just get a metal cutting saw from a hardware store and take it all to a welder. It's not that bad. I designed and built new exhaust sleeves for my engine and these determined what the header would have to be like.

I used to know about how large an expansion chamber needed to be. I'll go look for it when I get home tonight. After the chamber, the pipe length is completely irrelevant. One thing you need to be open to is the fact that the engine could theoretically want a longer system that you can get under the car after the collector. Look at the Mazdatrix road race car as an example of this. They looped the pipe back under and then back out again. Weird but it worked.

coldfire

iTrader: (1)

good post, i agree with your points. i was interested in seeing what others knew about this subject.
the basic things that you mentioned i have already figured out (all the important stuff really), so i was seeing what else is left to do in terms of exhaust.
although i think some people think about this TOO simply and just end up wasting money.

there must be something different about the rotary. i have read some websites (piston engines mind you) claiming that the portion after the collector is the most important. of course this is not true for the rotary. i definitely agree that the primary length of the exhaust system is the important.

one thing i am skeptical about is if a true dual really makes more power. even without large overlap, pressure wave tuning seems to be quite important and you don't really get any of that with a true dual system. there is no chance for pressure wave scavenging.
the stockport 13B NA, for example, that i am doing this bit of experimenting on, i have managed to get 154whp on a Mustang dyno, using a RB collected header (short primary), and a Magnaflow oval muffler about 5 inches after the collector (acting more or less as an expansion chamber it seems).
i doubt a true dual would really give much more than that, although i can't comment on the difference in power delivery.

it would be great to see that info on expansion chamber sizing.
also, if you have any info on how a packed muffler (such as a Magnaflow or Borla straight through) or a catalytic converter behaves compared to a true expansion chamber that would be great.

jgrewe

iTrader: (1)

In the book Practical Gas Flow the author says the expansion chamber should be the volume of the engine displacement if you can fit it under your car.

PDF

I'm interested in your calcs for resonant noise cancellation if you'd like to share?

Personally I think you could spend a life-time on n/a rotary exhausts and still not learn it all. Look at the design and development of 2 stroke mx bike expansion chambers and mufflers...

j200pruf

iTrader: (4)

So would we use one chamber, two chambers, or all 6 to calculate displacement?

kontakt

Since that was a statement about piston engines (I assume) I would think it best to use 2.6L for a 13b.

coldfire

iTrader: (1)

there we go, that does make sense. thanks for that reference.

the main reference for my calculations comes from this document:
http://myfwc.com/boating/airboat/Section3.pdf

it is a very good read.

but basically there are two variables that determine resonant noise cancellation of a muffler.
1) the length of the muffler determines what frequency of sound a muffler is optimized to reduce; the thing to realize is that a static muffler can not cancel out all frequencies of sound, it can only be optimized to cancel out a certain frequency and it's mutliples.

2) the effective volume of the muffler (for a round muffler this is determined by the radius) then determines the ammount of noise reduction. basically the larger the volume, the larger the amount of sound reduction.

so if you know some basic physics, resonant noise cancellation works through the destructive interference of sound waves. however a muffler has a fixed physical size, and therefore there is a specific frequency that the muffler does the best job of cancelling out. once you start straying away from this frequency less cancellation occurs.

So it's the length of the muffler that we are really concerned about. The way to determine the relationship between length and frequency is through "quarter-wave theory". You can try to look this up on the internet, but basically:

frequency = (n*c)/(4*length), where 'n' is an odd positive integer and 'c' is the speed of sound constant.
so let's say for a 14 inch muffler, for n = 1, we get:
c = 344 m/s
L = 14 in = 0.3556 m x4 = 1.4224 m
frequency = (1*344)/(4*1.4224) ~= 241 Hz

now an interesting thing to note here is that the muffler will also do a good job cancelling out 723 Hz, 1205 Hz, and so on. but for the typical combustion engine you will not be seeing too many frequencies above 600 Hz.
ANOTHER thing to note, is that for the even multiple, so let's say n = 2 which gives 482 Hz, the muffler will not be doing ANY resonance cancellation at all!

anyways, to put this into practice, you can try recording the sound of a car that is too loud right now, analyze the dominant frequency on a computer, then plug that into the equation to give you the optimum length of muffler you need.
unfortunately though one thing to consider is these calculations do not take into account the internal construction of a muffler. i would assume these calculations would be most accurate for a straight-through type muffler.

for a 13B i would say 2.6L of volume would be a safe assumption for sizing.

kontakt

So, have you determined yet what you feel would be a good muffler length for the rotary (I know it will be variable based on port work and other factors, however it should be more similar than for a piston engine)? At that length, if it exists, do you have any opinions on the other dimensions.

The exhaust system that I've always been interested in is basically a short megaphone collected system going into a decent sized straight through muffler as a "pre-silencer" or "resonator." After that the exhaust shouldn't matter. Like you have said the key factors are how far after the engine do you collect, and after collection how far to the resonator. I've always figured on something like a Borla XR-1 for that.

coldfire

iTrader: (1)

well it really depends how much sound level is "acceptable" to you, and determining the loud frequencies of a particular car is usually hard unless you actually record it and analyze it yourself.

my suggestion, and this is the route i am taking, is firstly make an exhaust system with performance in mind, but leave room AFTER the expansion chamber to add more components. you can put a good straight through muffler, such as a Borla, at the end as a starting point (which may be good enough).
now if you are finding that too loud, record the sound of the car when it is at it's most annoying (i.e crusing down the highway at 80 mph).
with some simple audio software on a computer you can get a good idea of what frequency of sound is primarily making the noise. then plug into calculations i gave in my previous post and try to find a muffler with that length.

as for other dimensions, generally the bigger the muffler the better. ways to reduce noise that won't really affect performance:
- larger muffler volume
- thicker walled piping
- not exceeding the max required pipe diameter (bigger is not better! 2.25 to 2.5 inches is plenty for a stock to mildly ported NA rotary)
- for straight through type mufflers, an "offset" where the pipe goes in on one side and comes out the other side helps the packing material absorb more sound

kontakt

Thanks.

I'm sorta thinking about making an adjustable system for the fabrication process now. I'd essentially make an expansion chamber in 2 halves that slide into each other, and either do likewise with the pipe coming off of it, or weld up some pipes to go after it in various lengths. Then I could run it at different lengths and see first hand what kind of sound difference it makes.

I know calculating it sounds relatively nice and easy compared to this. I love math (yeah I know I'm a freak) but I know that especially with exhaust systems the math is never quite on. It could be that the annoying frequency isn't even the loudest one. I know this isn't the best example as an expansion chamber won't stop a rattle, but rattles are often not the loudest noise your car is making, but often the most annoying.

If I end up taking on such a venture, I will be sure to post up my results.

iceblue

The only thing about expansion chambers is how do you figure out where to place it optimally and how is it going to effect with both the amount of pipe after and the silencer at the end?

As far as the header well God basically said everything but you want to tune length according to pulse width to catch each pressure zone of the wave. The collector will make or brake the system both size, length and placement will matter greatly. You can experiment with an adjustable collector sleeve.

coldfire

iTrader: (1)

yeah an adjustable collector like that would be nice, but isn't really necessary for a street car. it would be nice to see some tests results if someone has done that.

i agree placement of the expansion chamber is important, but everything after the expansion chamber is not too important, just as long as it flow properly.

now with that said, i will add one more thing:
it's generally better to reduce the diameter of the pipe (or total cross sectional area i believe) as you approach the end of the exhaust system. this is because as the exhaust gas cools traveling to the end of the piping it will actually loose energy and so will slow down. reducing the diameter of the pipe will force an increase of velocity. this is more of a factor at lower engine speeds since the pressure waves are more heavily "damped" if the exhaust gases are moving slowly in the system.

i have had the chance to try this out, i got a Borla muffler installed a few days ago with the piping going into it reducing by 0.25 inch (my exhaust is 2.5 inch until the back where it reduces to 2.25 inch). i don't have any dyno results but i can say that below 4k power delivery seems smoother.

Jeff20B

iTrader: (13)

coldifre, the stock Cosmo and RX-4 exhaust contracts slightly right before the rearend. It goes from about 2" OD to around 1 7/8" OD. Interesting stuff.

coldfire

iTrader: (1)

well there you go ^
see i'm not talking out of my ***, haha...