Welcome to the North American Subaru Impreza Owners Club Thursday July 24, 2014
Home Forums WikiNASIOC Products Store Modifications Upgrade Garage
NASIOC
Here you can view your subscribed threads, work with private messages and edit your profile and preferences Home Registration is free! Visit the NASIOC Store NASIOC Rules Search Find other members Frequently Asked Questions Calendar Archive NASIOC Upgrade Garage Logout
Go Back   NASIOC > NASIOC Technical > Built Motor Discussion

Welcome to NASIOC - The world's largest online community for Subaru enthusiasts!
Welcome to the NASIOC.com Subaru forum.

You are currently viewing our forum as a guest, which gives you limited access to view most discussions and access our other features. By joining our community, free of charge, you will have access to post topics, communicate privately with other members (PM), respond to polls, upload content and access many other special features. Registration is free, fast and simple, so please join our community today!

If you have any problems with the registration process or your account login, please contact us.
* Registered users of the site do not see these ads.
Reply
 
Thread Tools Display Modes
Old 12-05-2008, 04:41 PM   #1
SaabTuner
Scooby Specialist
 
Member#: 67608
Join Date: Aug 2004
Location: SoCal
Default Header Design Physics: Discuss.

The simple pipe resonance for header function is relatively well-known. However, there are several other aspects which have not been widely published.

Here's the Burns Stainless article on wave action tuning: http://www.burnsstainless.com/TechAr...ry/theory.html

Here are the following topics I've not seen sufficiently addressed:

1: When the exhaust valve on an engine first opens, the flow through the slowly-increasing orifice is choked. The time the flow remains choked depends on the size, and shape, of the entire header (even muffler, as restrictive mufflers do have an effect on wave reflection even upstream), as well as the size of the exhaust valve(s), their aerodynamic properties, and the cam profile. Here's a technical snippet on Mass Flow Choking: http://www.grc.nasa.gov/WWW/K-12/airplane/mflchk.html

2: The flow through a header tube causes the speed of sound to be different in different directions, based on the flow of the medium itself.

3: Due to flow-choking, the expansion of the exhaust gas into a header tube is not an isentropic process: http://www.grc.nasa.gov/WWW/K-12/airplane/compexp.html

4: Exhaust blow-down is neither a constant pressure, nor constant volume, process. Because the change in temperature (and speed of sound) varies between constant volume, and constant pressure, the local speed of sound varies over the distance of the header pipe, and over time as the local density/pressure changes: http://www.grc.nasa.gov/WWW/K-12/airplane/specheat.html

5: Exhaust composition changes as it moves away from the valves, contributing to the variable nature of the gas's specific heat values.

6: The Reynold's Number is not a constant from one end of the flow to the other.

I'm sure I'll think up more later ... but these are what has been niggling me for a while now.

Please add your thoughts on the subject!

-Adrian
* Registered users of the site do not see these ads.

Last edited by SaabTuner; 12-05-2008 at 04:44 PM. Reason: Typooooossss ... BLEAARRGGHH!!
SaabTuner is offline   Reply With Quote
Old 12-05-2008, 05:12 PM   #2
Mr_Snips
Scooby Specialist
 
Member#: 152117
Join Date: Jun 2007
Chapter/Region: SCIC
Location: Hills Have Eyes
Vehicle:
2012 Ford F150 ECO
White

Default

my thoughts are that that will be way over alot of people's heads.
Mr_Snips is offline   Reply With Quote
Old 12-05-2008, 06:44 PM   #3
bluescoobywagon
Scooby Specialist
 
Member#: 14911
Join Date: Feb 2002
Chapter/Region: MWSOC
Location: The land of E85
Vehicle:
Not A Wagon!
2011 WRX - 285s on 18x10s

Default

I was thinking the same thing, and was glad I wasn't one of them. There's probably 6 people on NASIOC that will have useful input. Hopefully I'll be one of them...
bluescoobywagon is offline   Reply With Quote
Old 12-06-2008, 12:11 AM   #4
Homemade WRX
Scooby Guru
 
Member#: 33782
Join Date: Mar 2003
Chapter/Region: MAIC
Location: 3MI Racing LLC
Vehicle:
96 bastard child
search FIRST, then PM!!!

Default

I'm thinking I'm tired and that would be a lot of curves to work out...

I'll read it tomorrow...but choking at the valve still makes me wonder why so many people target the wave pulse to start reflection just prior to the exhaust valve opening...it just makes a greater pressure ratio, which is a pointless target, IF the flow truly is choked at that moment during blowdown.
Homemade WRX is offline   Reply With Quote
Old 12-06-2008, 07:29 AM   #5
bluescoobywagon
Scooby Specialist
 
Member#: 14911
Join Date: Feb 2002
Chapter/Region: MWSOC
Location: The land of E85
Vehicle:
Not A Wagon!
2011 WRX - 285s on 18x10s

Default

After reading all of the articles, I come to the conclusion that the biggest variable most of us are dealing with is the turbo.

Typically, in a n/a application you tune the pressure waves so that you scavenge the cylinder more efficiently and ignore the gas particle choking factor. This leads me to believe that the flow into the cylinder is more important than the flow out of the cylinder provided the flow of the exhaust is sufficient to maintain low backpressure.

However, in a turbo application you also need to consider the spool of the turbo. So my questions are:

Is it possible to time the pressure waves after the turbo by tuning the diameter and length of the exhaust in order to increase turbo spooling? Could you tune the positive pressure waves before the turbo and the negative waves after the turbo to alternate in a fashion that has some of the advantages of a twin scroll?
bluescoobywagon is offline   Reply With Quote
Old 12-06-2008, 12:28 PM   #6
hotrod
Scooby Specialist
 
Member#: 14141
Join Date: Jan 2002
Location: 13.239@102.85 @ 5800 ft on 13T
Vehicle:
2002 Impreza WRX
e85forum.net

Default

Choked flow only occurs at the valve seat and shortly down stream for a very short time during the early stages of blow down as the exhaust valve first leaves the seat. In dry air the pressure ratio across the orifice must exceed 1.8:1 for choke conditions (sonic flow) to occur. Most other gases are near 1.8-1.9. A good exhaust system has much too little restriction to create sonic flow at any other location except at the turbo hotside where it might go sonic due to high pressure ratios across the turbo (ie exhaust pressure 2.5x boost) is easily high enough to create critical flow at the Turbo scroll as the gasses expand through the turbine to the ambient exhaust pressure in the tail pipe of about 6 psi or less in a properly sized muffler, tail pipe setup.

If you have 6 psi back pressure in the tail pipe, and 14.7 atmospheric pressure, if exhaust manifold pressure exceeds 1.8x (6+14.7) = 37.26 psia, then you can have choked flow in the turbo scroll.


Choked flow is only mass flow choked at a constant density, as pressure rises on the upstream side of the orifice, mass flow will continue to increase in spite of the fact the orifice is choked in velocity. Due to high exhaust gas pressures in the exhaust manifold sonic flow at the exhaust valve seat is shortened even further as cylinder pressures must be 1.8 x the exhaust manifold pressure for sonic flow to occur.

Regardless of sonic flow at the valve, there is still a pressure wave that moves away from the cylinder. In tests done in the 1960's on NA engines typical "effective" wave speed was approximately 1800 ft/sec. This led to several wave tuning formulas that all clustered around the same results plus or minus 2 inches or so in header tube length.


Exhaust tuning formula posited by Gordon Blair in the 1964 Hotrod article:


L = (135 x S )/ N

L = length of exhaust pipe in feet from exhaust valve head to tip.
S = number of crank shaft degrees from Exhaust valve opening to intake valve opening. { this formula obviously tunes to enhance early cylinder filling by having the negative exhaust pulse arrive back at the port during the beginning of overlap to assist scavenging and start of air motion in the intake port}

============

Two Stroke Tuners Handbook

L = (Eo x Vs ) / N


L = Exhaust tube length in inches
Eo = Exhaust valve (port on a two cycle) open period in degrees
Vs = velocity of sound wave in ft per second usually assumed to be 1700 ft per second for the exhaust
N = engine rpm

=================
Intake tuning formulas have similar values:

The Chrylser formula used in their patent application as presented in the 3 rd reference I posted ( Scientific Design of Exhaust and Intake systems ) lists it as:

L= ( 72 x C )/N { + / - 3 inches }

L = duct length (inches) from Plenum ( the first reflection point) to the back of the intake valve.
72 = constant ( probably derived from valve timing )
C = velocity of sound in air ( this is usually assumed to be about 1100 ft/sec)
N = Engine RPM for maximum tuning effect


=============
(simplified Chryser formula as specifed by the 1999 Hotrod Article

L = 84000 / N

L = duct length (inches) from Plenum ( the first reflection point) to the back of the intake valve.
N = Engine RPM for maximum tuning effect


=============

Intake tuning formula posited by Gordon Blair in the 1964 Hotrod article:

L = (1100 x S) / N

L = duct length (inches) from Plenum ( the first reflection point) to the back of the intake valve.
1100 = assumed speed of sound in the intake tract
S = Number of crankshaft degrees after BDC until the intake valve closes up to a maximum of 75 ( any greater cam duration after BDC use 75 ----- is probably where the Chrysler formula comes up with the value of 72 )





Some good references on the topic would be:
======================
Hot Rod Magazine -- July 1964 ( not 1962)
Gordon Blair B. Sv, PHD G.I. Mech Eng. Mexico State University.
4 page article with formulas for intake ram tuning. Includes additional technical references.

Hot Rod Magazine -- August 1964 ( not 1962)
Gordon Blair B. Sv, PHD G.I. Mech Eng. Mexico State University.
3 page article with formulas for ram tuning exhaust systems.

Tests and calculations conducted by the above author show that for a 2 psi peak suction at the intake valve at maximum piston velocity would be reflected as a 2 psi positive pressure pulse from a properly formed bell mouth inlet stack.
Formulas derived by the author produce results consistent with the Chrysler formula, but include a more refined approach.

Scientific Design of Exhaust and Intake Systems
Philip H. Smith FI Mech E MSAE
John C. Morrison BSc PhD MI Mech E


274 page book covering in great details the dynamics of the automotive intake and exhaust system with Chapter 5 pp75 - 116 devoted exclusively to pressure phenomena. Source for the above mentioned Chrysler Patent application formula. Many illustrations of actual pressure diagrams of pressures in cylinder and in the intake and exhaust tract of running test engines. Very interesting data.



Two-stroke Tuners Handbook.
Gorden Jennins

156 page book on tuning the two cycle engine with 21 pages devoted to exhaust ram tuning design, including formulas which like all the above sources focus on tuning the arrival time of a reflected pressure pulse and not pressure recovery from the enertia of a fast moving column of gas.
Gives very good illustrations and easy to read discussion. Includes design and effects the megaphone has on ram tuning.

===============

Wave tuning is useful, but the dominant factor for power production is, cylinder filling, which is usually dominated by the peak intake port velocity, with maximum VE achieved when peak port velocity nears .6 Mach. The same effect plays in the exhaust side but there high port velocities improve extraction from the cylinder and induce a higher vacuum signal on the intake port to start filling earlier.
Wave tuning is a compliment to the pressure signals created by the piston motion, and the extraction signal generated as the exhaust blows down. One of the best tools for playing with exhaust tuning numbers is also the cheapest. Check out Maxrace software called Pipemax. The guy who wrote it is an NHRA record holder engine builder and has his act together. There are lots of other interesting things you can calculate with this package like how far down the cylinder the piston is for each degree of crank motion etc.

http://www.maxracesoftware.com/drag_racing_software.htm

Larry

Last edited by hotrod; 12-06-2008 at 12:52 PM.
hotrod is offline   Reply With Quote
Old 12-06-2008, 02:18 PM   #7
Unabomber
Big Ron
Moderator
 
Member#: 18062
Join Date: Apr 2002
Chapter/Region: MAIC
Location: I can save you a ton of cash
Vehicle:
on car parts so PM
me b4 j00 buy

Default

Larry and Adrian in one thread....I'm friggen outta here!
Unabomber is offline   Reply With Quote
Old 12-06-2008, 04:52 PM   #8
SaabTuner
Scooby Specialist
 
Member#: 67608
Join Date: Aug 2004
Location: SoCal
Default

Your input is appreciated as always, Larry. However, there are a couple things I think you might want to look at more closely.

Quote:
Originally Posted by hotrod View Post
Choked flow only occurs at the valve seat and shortly down stream for a very short time during the early stages of blow down as the exhaust valve first leaves the seat.
I think the phrase "short time" is a bit misleading. Automobile exhausts have VERY unique flow properties when it comes to choke conditions. Also, as you know, I own a copy of, "The Scientific Design of Exhaust and Intake Systems". It's one of the reasons I made this post, as it did NOT take into account the unique variable mass-flow rate seen in piston engine exhausts.

The variation in mass-flow causes some unique interactions with the flow-field as it approaches the local speed of sound.

Quote:
Originally Posted by Dr. Genick Bar-Meir
Fanno Flow

Figure: Control volume of the gas flow in a constant cross section

An adiabatic flow with friction is named after Ginno Fanno a Jewish engineer. This model is the second pipe flow model described here. The main restriction for this model is that heat transfer is negligible and can be ignored 9.1. This model is applicable to flow processes which are very fast compared to heat transfer mechanisms with small Eckert number.

This model explains many industrial flow processes which includes emptying of pressured container through a relatively short tube, exhaust system of an internal combustion engine, compressed air systems, etc. As this model raised from need to explain the steam flow in turbines.

...

Why is the Flow Choked?

First, it has to be recognized that the critical point is when M=1. It will be shown that a change in location relative to this point change the trend and it is singular point by itself. For example, dP(@ M=1) = infinity, and mathematically it is a singular point (see equation (9.24)). Observing from equation (9.24) that increase or decrease from subsonic just below one M= (1 - e) to above just above one M= (1+ e) requires a change in a sign pressure direction. However, the pressure has to be a monotonic function which means that flow cannot crosses over the point of M=1 . This constrain means that because the flow cannot ``crossover'' M=1 the gas has to reach to this speed, M=1 at the last point. This situation is called choked flow.
From here: http://www.potto.org/gasDynamics/node138.php
And here: http://www.potto.org/gasDynamics/node143.php

(it's actually quite interesting that this explains much of what Philip H. Smith was unable to explain regarding the wave dynamics of headers in certain situations)

Quote:
Originally Posted by hotrod
In dry air the pressure ratio across the orifice must exceed 1.8:1 for choke conditions (sonic flow) to occur. Most other gases are near 1.8-1.9. A good exhaust system has much too little restriction to create sonic flow at any other location except at the turbo hotside where it might go sonic due to high pressure ratios across the turbo...
That's actually not correct in the case of a car exhaust. That's an ideal case for which a positive-displacement (piston/rotary/etc) engine is a unique exception. That ideal case assumes two infinite volume spaces with a smooth orifice between them, one with xx more pressure than the other. (if the orifice is merely rough, it also doesn't work)

Here's another related article on how the dynamics of a compressible fluid, like air, being displaced by a non-compressible item (a liquid, a piston, a rotor) tends to cause flow choking in instances where the more ideal case would not: http://www.potto.org/gasDynamics/node189.php

It should also be noted that an exhaust valve could easily reach choke conditions several separate times during the exhaust stroke as the piston slows at BDC, then accelerates up towards TDC. If it happens, it obviously depends on valve size, engine geometry, RPM, etc. But it's an interesting idea.

Quote:
Originally Posted by hotrod
If you have 6 psi back pressure in the tail pipe, and 14.7 atmospheric pressure, if exhaust manifold pressure exceeds 1.8x (6+14.7) = 37.26 psia, then you can have choked flow in the turbo scroll.
Actually, that's not entirely accurate either: there's far too great a drop in temperature after the turbine. The value would be significantly different.


Quote:
Originally Posted by hotrod
Choked flow is only mass flow choked at a constant density, as pressure rises on the upstream side of the orifice, mass flow will continue to increase in spite of the fact the orifice is choked in velocity. Due to high exhaust gas pressures in the exhaust manifold sonic flow at the exhaust valve seat is shortened even further as cylinder pressures must be 1.8 x the exhaust manifold pressure for sonic flow to occur.
That might be what intuition would suggest ... but a LOT of math says otherwise.

http://www.potto.org/gasDynamics/node111.php <--- describes multiple iterations of valve-opening shock. (I can't believe I finally found someone who's mathematically modeled this!)





http://www.potto.org/gasDynamics/node112.php <---- describes the effect of piston velocity, and the shock formed at the crown as the piston rises.

Quote:
When a piston is moving, it creates a shock that moves at a speed greater than that of the piston itself. The unknown data are the piston velocity, the temperature, and, other conditions ahead of the shock. Therefore, no Mach number is given but pieces of information on both sides of the shock.
Quote:
Originally Posted by hotrod
Regardless of sonic flow at the valve, there is still a pressure wave that moves away from the cylinder. In tests done in the 1960's on NA engines typical "effective" wave speed was approximately 1800 ft/sec. This led to several wave tuning formulas that all clustered around the same results plus or minus 2 inches or so in header tube length.
Not just ANY pressure wave- a shock wave!

Quote:
Originally Posted by Dr. Bar-Meir
Shock-Choke Phenomenon

Assuming that the gas velocity is supersonic (in stationary coordinates) before the shock moves, what is the maximum velocity that can be reached before this model fails? In other words, is there a point where the moving shock is fast enough to reduce the ``upstream'' relative Mach number below the speed of sound? This is the point where regardless of the pressure difference is, the shock Mach number cannot be increased.


Mmmmm ... math ....

I'm really surprised I found more interesting information there than at NASA's Glenn Research Center. (at least about wave dynamics)

The heat transfer problems, minor vorticity from curved tubing, and individual engine's shock-formation algorithms would be needed to accurately model potential changes to any given engine, even with all this data. I'm sure there is a line of diminishing returns, but I really don't believe we've reached it just yet.

So keep your thoughts coming!!

-Adrian
SaabTuner is offline   Reply With Quote
Old 12-06-2008, 05:06 PM   #9
SaabTuner
Scooby Specialist
 
Member#: 67608
Join Date: Aug 2004
Location: SoCal
Default

GAH! Sorry for the second post. This was the other very important part!

I've had two main complaints about ideas which were not well-addressed in "The Scientific Design of Exhaust and Intake Systems".

1: The apparatus did single exhaust events, one at a time. The effect of multiple, iterative, events is drastic! Hence the graphs in the last post.

2: This one I forgot to post just now, but my other "beef" was in how Smith ignored the obvious formation of a shock wave when the faster-moving gases come out of the cylinder and "hit" the slower moving gases. That sudden influx of mass-flow makes a huge difference to the wave-tuning, and especially to the functionality of collectors and "x-pipes", etc!

Quote:
Partially Open Valve

The previous case is a special case of the moving shock. The general case is when one gas flows into another gas with a given velocity. The only limitation is that the ``downstream' gas velocity is higher than the ``upstream'' gas velocity as shown in Figure (5.17).


Quote:
An additional parameter has be supplied to solve the problem. A common problem is to find the moving shock velocity when the velocity ``downstream'' or the pressure is suddenly increased. It has to be mentioned that the temperature ``downstream'' is unknown (the flow of the gas with the higher velocity). The procedure for the calculations can be done by the following algorithm:


Quote:
Earlier, it was shown that the shock choking phenomenon occurs when the flow is running into a still medium. This phenomenon also occurs in the case where a faster flow is running into a slower fluid. The mathematics is cumbersome but results show that the shock choking phenomenon is still there (the Mach number is limited, not the actual flow). Figure (5.16) exhibits some ``downstream'' Mach numbers for various static Mach numbers, My′, and for various static ``upstream'' Mach numbers, Mx′. The figure demonstrates that the maximum can also occurs in the vicinity of the previous value (see following question/example).
http://www.potto.org/gasDynamics/node114.php

Examples of shock dynamics, with the math done for you: http://www.potto.org/gasDynamics/node116.php

Just about moving shocks in particular: http://www.potto.org/gasDynamics/node107.php

-Adrian
SaabTuner is offline   Reply With Quote
Old 12-07-2008, 12:14 AM   #10
Kosmic
Scooby Specialist
 
Member#: 99140
Join Date: Oct 2005
Chapter/Region: E. Canada
Location: Montreal, Canada
Vehicle:
GD Racing chassis
Slowly in construction

Default

I'm one that like reading those theoretical discussions. Put Micah, Larry and Adrian (and others) mixed in a thread and you're up for a very interesting read. unfortunately, there is never a conclusion to those threads. Same thing happened in the intake manifold, exhaust pulse big turbo threads, lots of calculations/theories (very interesting) but no conclusion. What I mean is, that would be nice to see you all engineers/genius agree on a part to be fab (for a particular application ex: low rpm, high rpm, torque gain, hp gain) and come up with a part (material, ID, OD, length, radius volume etc etc). You guys have the means to calculate, some of us might have the means to fabricate and test it...come back with data and go forward from there until we reach a well design part that work in the real world for a particular application. I hope you see it as a positive criticism.
Kosmic is offline   Reply With Quote
Old 12-07-2008, 12:24 AM   #11
Mycues1982
Scooby Specialist
 
Member#: 86995
Join Date: May 2005
Location: Pittsburgh, PA
Vehicle:
2005 Tacoma

Default

I'm very glad you started this thread. I have just started researching the idea to start prototyping a few different exhaust parts for our car's and am very interested in this discussion. What are everybody's thought's on anti-reversion?

Last edited by Mycues1982; 12-07-2008 at 12:32 AM.
Mycues1982 is offline   Reply With Quote
Old 12-07-2008, 12:32 AM   #12
Mycues1982
Scooby Specialist
 
Member#: 86995
Join Date: May 2005
Location: Pittsburgh, PA
Vehicle:
2005 Tacoma

Default

I havn't read this thread entirely yet but like Kosmic states about there being so many variables that contribute to the best outcome I think being able to produce a product for our car's that can be adjusted for the modifications might be really cool. My ideas for an intake manifold that allow you to adjust the runner lengths given the cam's and turbo you are running is one. I havn't really thought of a way to produce an exhaust manifold yet that can be adjusted. Maybe doing a transition off of each exhaust port to allow the user to port there exhaust ports open more and still be able to give a good anti-reversion effect. Then doing a v-band connected lower section with different diameter runners givin the application. I wish I had a dyno..
Mycues1982 is offline   Reply With Quote
Old 12-07-2008, 12:35 AM   #13
Mycues1982
Scooby Specialist
 
Member#: 86995
Join Date: May 2005
Location: Pittsburgh, PA
Vehicle:
2005 Tacoma

Default

I can fab it, I just need to work on way's of testing everything.
Mycues1982 is offline   Reply With Quote
Old 12-07-2008, 01:20 AM   #14
AlxSti
Scooby Guru
 
Member#: 94615
Join Date: Aug 2005
Chapter/Region: MAIC
Location: Wilmington, DE
Vehicle:
|'06 STi 341/356|
www.AffinitiBrokerage.com

Default

subscribed so ill learn something
AlxSti is offline   Reply With Quote
Old 12-07-2008, 02:14 AM   #15
wadester
Scooby Newbie
 
Member#: 188054
Join Date: Aug 2008
Chapter/Region: MWSOC
Location: Lawrence, KS
Vehicle:
05 STi Cosworth LB
FP Green 375whp

Default

@SaabTuner, Hotrod, and bluescoobywagon.

How do I get to the point that you guys are at? As in, what kind of education do you all have that allows you to understand this stuff?

The amount of good information that you all put out is astonishing. We're lucky to have you on the boards.

Subscribed for later use, when I can make sense of the info.
wadester is offline   Reply With Quote
Old 12-07-2008, 08:33 AM   #16
SaabTuner
Scooby Specialist
 
Member#: 67608
Join Date: Aug 2004
Location: SoCal
Default

Quote:
Originally Posted by Kosmic View Post
I'm one that like reading those theoretical discussions. Put Micah, Larry and Adrian (and others) mixed in a thread and you're up for a very interesting read. unfortunately, there is never a conclusion to those threads. Same thing happened in the intake manifold, exhaust pulse big turbo threads, lots of calculations/theories (very interesting) but no conclusion. What I mean is, that would be nice to see you all engineers/genius agree on a part to be fab (for a particular application ex: low rpm, high rpm, torque gain, hp gain) and come up with a part (material, ID, OD, length, radius volume etc etc). You guys have the means to calculate, some of us might have the means to fabricate and test it...come back with data and go forward from there until we reach a well design part that work in the real world for a particular application. I hope you see it as a positive criticism.
Fair criticism is always positive criticism.

I can't speak for anyone else, but there are two things I'd like you to remember when you see a technical discussion involving me (which appears to have no conclusion):

1: I am always in the pursuit of truth, with as little personal bias as I can muster. I've lost more than one friend by taking the other side in an argument when his/her argument seemed flawed and I couldn't just lie and say I agreed. (I can't claim to be finding "truth", but I'll put logic before irrational bias any time I can)

2: I am never expecting to find the "real answer" ... EVER. How many times, in science, have we thought we were right, only to be proven wrong later? Seriously, what are the odds of me being right, even with a DANG good argument? Almost nothing! It's almost guaranteed that I (or anyone else) will miss at least SOMETHING!

What I WANT is for the information/ideas I bring to make your OWN choice more accurate, as best I can. I AM also sorry that I don't always make things as clear as I could; I'd like your decision to be as clear as possible, based on the data, but I also do NOT want it to be at all misleading!

Quote:
Originally Posted by mycues1982
What are everybody's thought's on anti-reversion?
Exhaust, or intake? For exhaust, the speed of the exhaust valve closing makes backwards exhaust gas flow very difficult: the valve(s) continue pushing it out while closing. Excessive overlap is a problem, but that's a cam timing problem more than an FD problem.

On the intake side, variable intake valve closing time, or duration, is the most common solution. However, the most successful solution I have yet seen has been a VERY large intake port, which can fill most/all of the cylinder by BDC. If the cylinder can be filled by BDC, intake reversion is a non-issue.

Quote:
Originally Posted by wadester
How do I get to the point that you guys are at? As in, what kind of education do you all have that allows you to understand this stuff?

The amount of good information that you all put out is astonishing. We're lucky to have you on the boards.

Subscribed for later use, when I can make sense of the info.
Ok, here goes:

--Shock waves behave VERY differently from "sound" waves. They are relatives of eachother, and share some properties. However, they have some very DISTINCT properties for each.

--Until recently, the role of shockwaves hasn't been well enough understood in car intake and exhaust systems.

I'll try to think of more later. Very tired right now!

-Adrian
SaabTuner is offline   Reply With Quote
Old 12-07-2008, 12:06 PM   #17
hotrod
Scooby Specialist
 
Member#: 14141
Join Date: Jan 2002
Location: 13.239@102.85 @ 5800 ft on 13T
Vehicle:
2002 Impreza WRX
e85forum.net

Default

Quote:
What are everybody's thought's on anti-reversion?
Quote:
Exhaust, or intake? For exhaust, the speed of the exhaust valve closing makes backwards exhaust gas flow very difficult: the valve(s) continue pushing it out while closing. Excessive overlap is a problem, but that's a cam timing problem more than an FD problem.
Most exhaust ports flow better in the reverse direction than they do in the normal intended direction of flow. I would strongly recommend you investigate anti-reversion cones on the header flanges. Since turbo charged engines typically have 2-2.5 x the pressure in the exhaust manifold, than in the intake or cylinder, reversion is a major problem and the reason supercharged engines typically have very low if not 0 overlap on the valve timing.

The speed of closure of the valves is mind numbingly slow compared to what the gases consider fast. Gasses have no problem making 180 G acceleration changes, the exhaust valve head is barely moving as it approaches the valve seat to avoid valve bounce at high rpm.

Same for piston speeds the crown of the piston is almost stationary compared to the air flow velocities and never moves fast enough for significant compressibility effects to be an issue. Air is considered incompressible until you are dealing with speeds that are a large fraction of the speed of sound. In aircraft design you can largely ignore compressibility until the vehicle is going about 400 mph or about 600 ft/second. At 7200 rpm peak piston speed is about 90 ft/second. Average gas flow velocity in the exhaust header is typically about 350 ft/sec, and in a properly designed port peaks near 670 ft/second, in the throat. Gas velocities just behind the exhaust valve as it first cracks off the valve seat jump almost instantaneously to near the speed of sound (1800 ft/sec at exhaust gas temperature).

At 7200 rpm the engine turns over every .008 seconds, exhaust valve duration is about 240 degrees so the exhaust valve opens 10 mm and then closes 10mm in .00555 seconds. The valve accelerations are very high for a mechanical component but the peak and average speeds are not impressive as it only moves 20 mm in .00555 seconds or about 11.8 ft/sec average speed. (about the speed of your average recreational runner)


I left anti-reversion lips on my heads when I ported my exhaust manifolds and I have no complaints.


Quote:
How do I get to the point that you guys are at? As in, what kind of education do you all have that allows you to understand this stuff?
In my case over 40 years of learning everything I can, talking to and listening to others who are also competent in engine design, and performance modifications, and reading everything I could find that seemed even remotely relevant.


Larry

Last edited by hotrod; 12-07-2008 at 12:41 PM.
hotrod is offline   Reply With Quote
Old 12-07-2008, 12:17 PM   #18
whozatguy
Scooby Newbie
 
Member#: 179775
Join Date: May 2008
Chapter/Region: South East
Location: Huntersville, NC
Vehicle:
. 03 WRX/05 FXT
14 BRZ

Default

Wow, I'm glad I found this thread, it almost directly mirrors some of the concepts and things I am learning about in a fluid dynamics class I am taking
whozatguy is offline   Reply With Quote
Old 12-07-2008, 01:08 PM   #19
charliew
Scooby Specialist
 
Member#: 125304
Join Date: Sep 2006
Location: Crawford, TX.
Default

I hate their sound but it really seems like the best future approach of gas motors would be 2 cycle with reed valves. My old kawasaki from the late 60's would compete with most anything 4 cycle.
I hate to dumb this down but if I had a dyno and I put the flow bench on the exhaust to pull about 800 cfm out of the exhaust and could keep from burning up the flow bench what would it do to the motors output. And would it prove how restrictive the exhaust side is?
Would several anti reversion lips be better than one in the exhaust header port?
For a wot moment would a variable dump valve on the exhaust side of the turbo be worthwhile? Sorta like a wastegate. I'm a street motor guy.
I'm sooo glad there are guys that thrive on math because it numbs my brain. My brain likes to make things not compute them.
It just seems that the configuration of the suby port and exhaust to the turbo doesn't allow much in advancements.
More power to you very smart guys, I hope you come up with viable answers.
charliew is offline   Reply With Quote
Old 12-07-2008, 03:17 PM   #20
SaabTuner
Scooby Specialist
 
Member#: 67608
Join Date: Aug 2004
Location: SoCal
Default

Quote:
Originally Posted by hotrod View Post
Same for piston speeds the crown of the piston is almost stationary compared to the air flow velocities and never moves fast enough for significant compressibility effects to be an issue.


Larry
The speed isn't, but the piston accelerates at several thousand g's.

In most aerodynamics courses they teach you that speed is the only concern for compressibility, but aerodynamics is traditionally looking at an object which, itself, is moving through a free stream.

Fast acceleration in a free stream also creates a shock wave, but, in the free stream, you can make the assumption that it will dissipate before it becomes significant.

Airplanes do not make 2,000g accelerations, let alone repeatedly, and in quick succession. Pistons do.

More food for thought.

-Adrian
SaabTuner is offline   Reply With Quote
Old 12-07-2008, 05:52 PM   #21
SaabTuner
Scooby Specialist
 
Member#: 67608
Join Date: Aug 2004
Location: SoCal
Default

Here's a link to a good snippet about "Compression Wave Steepening", which is a form of non-linear acoustic steepening: http://books.google.com/books?id=zIX...sult#PPA190,M1

Any wave motion over 170dB can be subject to non-linear steepening. If the acceleration is strong enough (even if low velocity) the wave will steepen into a shock front. However, the shock only exists until after the piston stops accelerating, and the rarefaction wave catches up to the shock front and dissipates it.

A quote from this link: http://journals.cambridge.org/action...ine&aid=396897
"Dynamic compression and weak shock formation in an inert gas due to fast piston acceleration"
Quote:
Unsteady gasdynamic concepts are used to model the piston-driven compression of a confined gas. Perturbation methods, based on the limit of small piston Mach number, are used to construct solutions. The piston Mach number increases smoothly from zero to a maximum value, Mp = O(10−2) during an acoustic time period ta* = O(10−4 s). A linear a coustic field is generated and is represented in terms of an infinite series of Fourier spatial modes. During the longer piston time period tp* = O(10−2 s) the piston moves at constant speed. A multiple-timescale formulation is used to separate the instantaneous acoustic field from the accumulated bulk response of the gas to piston compression. The latter is found to be identical to the classical quasi-static results from equilibrium thermodynamic calculations. Nonlinear effects become important on the piston timescale. Modal interactions are represented by a system of coupled, nonlinear ordinary differential equations for the time-dependent Fourier coefficients. A numerical solution for this system describes the wavefront steepening to form a weak shock and its propagation back and forth repeatedly inside the cylinder.
Another example: http://books.google.com/books?id=zVf...esult#PPA44,M1

Anyway ...

-Adrian
SaabTuner is offline   Reply With Quote
Old 12-08-2008, 12:28 AM   #22
Homemade WRX
Scooby Guru
 
Member#: 33782
Join Date: Mar 2003
Chapter/Region: MAIC
Location: 3MI Racing LLC
Vehicle:
96 bastard child
search FIRST, then PM!!!

Default

well finally got some time to read it now that its 12:40 at night...don't think too much stuck being I'm this tired. I'll read it tomorrow at work and hopefully have something of use to add
Homemade WRX is offline   Reply With Quote
Old 12-08-2008, 10:08 AM   #23
Homemade WRX
Scooby Guru
 
Member#: 33782
Join Date: Mar 2003
Chapter/Region: MAIC
Location: 3MI Racing LLC
Vehicle:
96 bastard child
search FIRST, then PM!!!

Default

Ok Adrian, so let me try and catch your last statement; with the formation of the shock wave formed via the piston until acceleration stops...how much longer does that shock wave dwell once the piston has started to decel? It would obviously related to rod ratio as that relocates the pistons accelerations and how strong they are...obviously the shorter the rod, the stronger the shock wave as it approaches TDC. I'd assume we don't have a similar wave "chasing" the piston down the hole or do we? Am I just that lost

Now, with this potential for the shock wave to be "seen" in exhaust port and the intake port (?...overlap); how can this effect the sound/resonance wave that is all to frequently used for tuning manifolds? how does it react when pressure waves are approaching and leaving each valve?

For reversion reasons, a stepped exhaust manifold is beneficial in two ways; it helps to reduce reversion of exhaust gases by causing vortices in back flow making somewhat of a pressure "wall" and it also broadens the area around your peak VE by acting much like a progressive spring. On the intake side, a similar, though reversed, step can help in a similar way, though in the case of our motors, by the time you get to a stepped surface (that isn't big enough to disrupt intake flow) and have enough pressure built up in a vortice to stall the reversion into the intake manifold, you'd have a decent amount of exhaust in your intake. So on our forced induction cars where we have higher pressure in the our exhaust manifolds than our intake manifolds and have reasonably long runners in the heads, it is hard to use "old school" techiniques to help stop reversion. However a lot can be found on the flow bench in port design. Ahh, residual gas and all the techniques to reduce it

I'm testing the long rod method which allows me to physically reduce the mass fraction volume, as well as allows more time for valve timing/overlap control, as the piston spends more time at TDC. The piston also has its peak acceleration lower in the cylinder; so that potential for a shock wave (if my understanding is correct) will have decreased effects on tuned wave and reversion.


As for background, I'm just an ME who got sucked into badly doing my motorsports engineering minor...now I'm debating going back for my masters or taking some specific higher level courses for my own "enjoyment" that are applicable to IC engines/my company.


...still wondering if I'm missing the point of this thread

Last edited by Homemade WRX; 12-08-2008 at 10:22 AM.
Homemade WRX is offline   Reply With Quote
Old 12-08-2008, 10:57 AM   #24
SaabTuner
Scooby Specialist
 
Member#: 67608
Join Date: Aug 2004
Location: SoCal
Default

Quote:
Originally Posted by Homemade WRX View Post
Ok Adrian, so let me try and catch your last statement; with the formation of the shock wave formed via the piston until acceleration stops...how much longer does that shock wave dwell once the piston has started to decel? It would obviously related to rod ratio as that relocates the pistons accelerations and how strong they are...obviously the shorter the rod, the stronger the shock wave as it approaches TDC. I'd assume we don't have a similar wave "chasing" the piston down the hole or do we? Am I just that lost
As soon as the piston stops accelerating quickly enough, a "rarefaction" wave travels up from the crown of the piston at the local speed of sound. However, the shock wave should reach the top of the cylinder before that can happen ... so it should reflect off the combustion chamber surface and then be canceled out as it comes back down by the still-rising rarefaction wave.

A rarefaction wave does occur when the piston goes down; but a rarefaction wave travels in the same direction to the density gradient, so the "stacking effect" cannot be set up to create a wave discontinuity (shock front).

Quote:
Now, with this potential for the shock wave to be "seen" in exhaust port and the intake port (?...overlap); how can this effect the sound/resonance wave that is all to frequently used for tuning manifolds? how does it react when pressure waves are approaching and leaving each valve?
When it reaches the valve opening, it will likely disrupt the flow, unless the flow is already in a choke condition. However, very little of that shock is likely to actually exit the valve. In any case, there's no way to really stop the formation of this shock while also using an engine at any useful speed.

Quote:
I'm testing the long rod method which allows me to physically reduce the mass fraction volume, as well as allows more time for valve timing/overlap control, as the piston spends more time at TDC. The piston also has its peak acceleration lower in the cylinder; so that potential for a shock wave (if my understanding is correct) will have decreased effects on tuned wave and reversion.
The shock should dissipate well before TDC. However, the less abrupt acceleration of a long rod setup may slightly decrease the disruption the shock wave causes during exhaust blow-down and evacuation by making the shock itself slightly weaker.

Quote:
...still wondering if I'm missing the point of this thread
Not at all! The point is just to discuss any, and all, of the physics (or science in general) behind exhaust header design and function.
SaabTuner is offline   Reply With Quote
Old 12-08-2008, 11:05 AM   #25
Schralperdotcom
Scooby Newbie
 
Member#: 182613
Join Date: Jun 2008
Default

Schralperdotcom is offline   Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
custom exhaust header ideas and discussion Bad Noodle Factory 2.5L Turbo Powertrain 52 06-15-2010 06:59 AM
Boost Creep and Down Pipe Design - a discussion Dyno Flash Factory 2.5L Turbo Powertrain 33 01-17-2007 09:13 PM
Ideas on top end with this header design ebeck Factory 2.0L Turbo Powertrain 11 08-06-2005 08:29 AM
New STI header design for 2.5L engines? Albert Normally Aspirated Powertrain 9 11-21-2002 03:09 PM
Header design Keiho Normally Aspirated Powertrain 7 01-17-2002 10:32 AM


All times are GMT -4. The time now is 02:16 PM.


Powered by vBulletin® Version 3.7.0
Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
Powered by Searchlight © 2014 Axivo Inc.
Copyright ©1999 - 2014, North American Subaru Impreza Owners Club, Inc.