KATOOM

Go ahead and remove the turbo. You will be shocked as to the muffling ability of the turbo. So much that DOT says it is a muffler.<br>Well if you have read his replys how can you argue that he is proving me wrong. He's not sure either. (no offence AlpineRAM) Like he said the tests were stopped. But he did say that the tests were conducted on a small motor with a catalytic convertor which holds a ton of heat, that brings us back to the subject of hot exhaust being BAD. And we all know that. ;D <br>I really think this sound wave discussion is getting beatin into the ground. Besides a muffler company has proven the situation wrong already. aka &quot;Flowmaster&quot;.<br>

AlpineRAM

Well I wouldn't be in a discussion and tell you my opinions and assumptions based on the facts I have seen if I wasn't here to learn. I know that the Cummins is a very different animal to the other I6s I worked on, and I am here to get lot's of infos about the specific tips and tricks for this engine. The members of this site have proven to be a great resource of knowledge, thoughts and practical experience. <br><br>AlpineRAM

HOHN

Doug, Alpine, Gary, et all.<br><br>Good posts. Doug, you posted something that got me thinking. In some respects, we can NOT separate the sounds waves from the flow aspect. Why? Well, what CAUSES the sound? It's the rapid increase in pressure as the exhaust tries to force itself into the manifold and pipes.<br><br>In fact, ALL sound is simply our perception of wave energy within the spectrum of human audibility. We can hear 60hz hum (from lights and such) just as surely as we can FEEL 60hz when blasted through a subwoofer.<br><br>So could we not then reason that sound energy is a result of the gas trying to flow? Furthermore, it doesn't flow one constant rate. It pulses, and spikes; it goes from trying to flow 0 cfm almost instantaneously to MAX cfm.<br><br>I imagine the analogy of arriving at proper carburetor size (on a gas hot rod). Simply math tells you that all you have to do it take the displacement* the max rpm and that will get you the CFM of the ideal carburetor, right? WRONG!! This calculated CFM only gives you the AVERAGE flow rate! For example, if you run the math for a 350 chev, you find that (350cid*6500rpm)/3456= 658CFM (if you assume 100% volumetric efficiency). Yet, when you run engine tests on track and dyno, you will find that a 750 cfm carburetor will produce more HP!<br><br>This is because if we don't make the carb large enough to accomodate the &quot;spikes&quot; it creates an intake restriction. Sometimes this is desirable (increases intake velocity, makes car more streetable).<br><br>I once modeled the change in volume of a cylinder as it goes through its strokes. It always follows a sine wave pattern (when graphed against a 360° crank rotation).<br><br>So I reasoned as follows: when you have a sine wave in electricity, you can find the &quot;average&quot; value by multiplying by .707 to get the Root Mean Square (RMS). So, if the &quot;average&quot; value of a sine wave represents only 70% of the peak instantaneaous value, then we can reverse to find the peak demand.<br><br>Now if we go back to our carburetor and look at that as only 70% of the max air demand, we THEN find out that the largest carburetor should be 930CFM!! Anything larger will not increase power, since the carb poses no restriction.<br><br>While this example falls apart for the carburetor (when NEEDS restriction or it will not function), it DOES help us look at the maximum theoretical air demand in the diesel exhaust.<br><br>If the engine is consuming 600CFM (at standard temp and pressure) into the turbo (which, remember is not a pulsing intake, but a constant &quot;whoosh: as mentioned earlier), then we could say that the momentary &quot;spike&quot; in the pressure waves would be substantially higher- on the order of 848CFM.<br><br>Now, we must also account for the fact that we are expelling far more VOLUME of gas than we took in: our diesel fuel has been converted to hot gas. I don't have the calculations handy, but you could easily see that the fuel introduced to the engine might cause our momentary air flow to &quot;spike&quot; well over 1000 cfm.<br><br>The turbo does a great deal to minimize these &quot;spikes&quot; in airflow. Because it is a restriction, it tends to slow down instants of fast flow (max flow), and speed up the moments of very little flow (because of the residual pressure from the max flow moment). It adds &quot;inertia&quot; to the flow-- much like a carburetor that is sized a little small.<br><br>I think, though, that even when you account for the turbo's damping effect on the pulsing, you can still have momentary air flows that are MUCH higher than what a simple displacement*rpm formula would give you, even if you factor in added boost as additional displacement.<br><br>HOHN

AlpineRAM

Well I can throw in some cents here now ;D<br>When we look at an exhaust cycle of a single cylinder we get a certain distance of pipe filled with hot gas travelling at some speed. (Depending on the diameter of the pipe compared to the size of the cylinder, speed of piston movement, pressure at the end of the exhaust pipe and filling of the cylinder) We called this a gaspillar. Due to the inertia of this gaspillar it will leave a zone of low pressure behind when the exhaust valve closes. These pillars can interact with each other. Naturally most of these mutual influences are governed by the firing order and the design of the exhaust manifold. Since the resistance to flow is an exponential equation over speed and not linear we will see that a long small diameter pipe will equalize the pressure curve within this gaspillar. A short thick pipe will equalize it much less. <br>Since the turbine housing with the turbine wheel can be considered as a short piece of small pipe after a long section of big pipe it will equalize the pressure curve within the gas pillar a lot, creating backpressure. (OK we use this backpressure we usually try to avoid for something useful) <br>There is a factor not mentioned yet, that's the effective pipe diameter. I see only the cross section of pipe as effective where the flow is laminar. Turbulent flow on the sides of the pipe decreases usable diameter, and therefore the speed of the exhaust gas and the area of the turbulent layer in comparison to the area of laminar flow changes. A sudden gradient of pressure (a sound wave) will effect the movement of the molecules, and therefore will also influence the position of the turbulent layer. <br>Naturally if we have a very big and very smooth pipe the effects will be less noticeable than with a small and rough pipe. <br><br>AlpineRAM

HOHN

I agree with the part about laminar flow. There is, however a point where turbulence can be good.<br><br>The point where the airflow is against the tube walls is one of very high drag. The air molecules that immediately touch the walls are actually not moving. Then, as you move away from the walls, the molecules speed up, until you hit the max velocity somewhere near the center of the flowstream.<br><br>There is a &quot;velocity gradient&quot; within the flowstream. It ranges from zero at the tubing walls to maximum in the center (or near it).<br><br>In Aero, they call this the &quot;boundary layer&quot; effect. As implied, the boundary layer is the layer of the flowstream that lies directly against the surface.<br><br>The boundary layer effect increases drag on the air moving over it. However, making this boundary layer turbulent DRASTICALLY lowers the friction on the stream trying to flow past it. This is part of the reason why golf ***** have dimples-- the turbulent boundary layer reduces aerodynamic drag (though this a mixture of pressure drag and boundary layer drag).<br><br>True, also that the highest- hp cylinder heads in racing do NOT have mirror-polished ports, as you might expect. They are actually finished to a 60 grit texture or so. This also benefits the &quot;wet flow&quot; of the gas-air mixture.<br><br>So, if you could make an exhaust pipe that had a bunch of little holes in it (kinda like an air hockey table in a tube), and you could either pressurize the pipe via these holes, or draw a vacuum via these holes, the flow would go WAY up. Air moving through these holes (in or out) makes the boundary layer turbulent, and reduces boundary layer drag.<br><br>So, the ideal exhaust pipe would resemble the surface of a golf ball on the pipe's inside. And it would have VERY laminar flow throughout the pipe, with a MUCH better drag gradient at the pipe's walls.<br><br>You can't really say that the pipe effectively has NO cross section where the flow is turbulent, but it IS fair to say that the pipe has reduced efficiency where the flow is not laminar.<br><br>Aero is just too complicated to be reduced simply, ESPECIALLY when you add the pulsed wave energy of exhaust to the mix!!!<br><br>Man, I wish we could just find out the skinny by trying out a bunch of different things and measuring EGT, HP, air velocity, etc.<br><br>HOHN

AlpineRAM

HOHN: I agree on the dimple statement. I think we concur about the effective diameter being the one with laminar flow- Actually you create a kind of cushion with microturbulences to &quot;lube&quot; the laminar flow. If you evacuate or blow into the boundary you can increase the laminar diameter of a pipe. I should have expressed myself better than saying &quot;smooth pipe&quot; I should have said &quot;smooth flowing pipe&quot;<br>I have built lots of heads with different textures on the intake to achieve best flow. Most exhaust ports were extremely smooth to reduce heat import from the exhaust gas. Sometimes we did do a ceramic coating on them. <br>Another boundary can exist between gasses of different temperatures. They won't mix readily and sound can be deflected by this boundary. And since any muffling device will convert kinetic energy into heat we have a completely new zoo of phenomena in there. <br>One of the main challenges in designing an exhaust IMHO is that in daily driving so many parameters change over very wide maps that a great deal of work is involved in finding a solution that will give good results under any driving condition and will fit under the vehicle. In circuit race car building we had at least the only goal to optimize for power under WOT over a small powerband. <br>Since we are in the high perf. forum here we can tune the engine to a more specialised spec than a manufacturer who can not target only us loose nuts behind the steering wheel. <br>And we do not want to pay much for these trucks to have some money left for bombing. <br><br>AlpineRAM

doug

[quote author=Gary - KJ6Q link=board=7;threadid=14778;start=45#140027 date=1053385530]<br>&gt;...comments about steady vs chaotic and standing waves, SWR, etc.&gt;<br>Yeah. I know that was what you really meant.... Just wanted to make it clearer. ;D ;D <br>[/quote]<br><br>well, I don't think we disagree here, except perhaps with regard to the time factor. my point was that for an mechanical standing wave to develop, energy has to be present at the correct wavelength for a time greater than zero. this is because kinetic energy has to be converted into potential energy and vice versa, the medium in which this occurs (air) is lossy, and the boundaries at which the wave is reflected (sudden density changes) are not 100% reflective. <br><br>In a practical sense, and for the purpose of illustrating my point, this (time delay) is largely not true of an electromagnetic wave, for example one that would be supported by a 1/2 wavelength antenna and for which we would be accustomed to measuring the SWR. since air is not involved in the wave propogation, the time it takes the electromagnetic standing wave to develop is near zero, mostly because the energy travels at approximately 1 million times faster than the speed of sound in air.<br><br>also the standing wave ratio is a voltage measurement performed in a transmission line that feeds a load impedance, and the measurement is a description of how closely the load impedance matches the characteristic impedance of the transmission line and the output impedance of the source. Mathematically, this is a different expression than how we describe mechanical wave reflections, but I do understand what you are getting at: there is some ratio of transmitted to reflected energy in both the electromagnetic and the mechanical examples. You are right about that.<br><br>another part of my point was that the spectra of acoustic energy emitted by the exhaust manifold is both rich (very broad) and to a significant degree, constant. that is, at any given point in time there is energy at just about any arbitrary frequency you want to name --- for the most part, only the relative amplitudes in the spectrum will change. so the task of the exhaust designer is difficult in the sense that (virtually) *all* frequencies are present all the time and are capable of exiting a mode in the exhaust system, producing a standing wave.

doug

[quote author=HOHN link=board=7;threadid=14778;start=60#140156 date=1053432433]<br>In some respects, we can NOT separate the sounds waves from the flow aspect. Why? Well, what CAUSES the sound? It's the rapid increase in pressure as the exhaust tries to force itself into the manifold and pipes.<br>[/quote] be careful here. We MUST analyze the sound waves separately from the mass flow because their velocities are an order of magnetude different and their behaviors are described by different mathematics. Sound is described by wave theory. mass air flow is described by fluid dynamics. the first describes what we hear. the second describes how the air moves throught the system to impact cooling. we cannot ignore the certain interaction, but the behaviors are distinctly different. lets be precise here. the statement, &quot;ALL sound is ... our perception of wave energy...&quot; is not correct. sound is a particular form of wave energy, namely a mechanical wave in air. period. it is not the human perception of any arbitrary wave energy between 20 and 20kHz, such as 60hz hum from the lights. we hear that because the balast happens to vibrate mechanically at 60Hz and this exites the air to propogagte a sound wave which in turn vibrates our eardrums. we absolutely cannot hear electromagnetic energy itself, which is present all around us, even at 60Hz. period. If that were true than you would hear the wires in your walls along with the lights.<br><br>We feel subwoofer frequencies at high energy levels simply because it generates a mechanical wave in air at high intensity, and because human tissues other than the eardrum vibrate in response to the wave. There's no magic or any correlation or association with 60Hz buzz from the lights. <br>trying to keep this discussion subject to physical laws here an not to get distracted. we must look at two things: air in the exhaust system moves VERY SLOWLY compared to the speed of sound. even the big momentary spikes. the piston peak velocity is easy to calculate, and is only on the order of 30 miles per hour at 2000 rpm. So these pulses of air do NOT, in and of themselves, define the sound wave(s). What happens is that air rushing around corners, through the valves, etc, etc. produces sound waves that propogate at more like 750 mph towards the tailpipe, and these waves must themselves travel through air that is traveling below 100 mph. So no, the air pulses themselves are NOT the sound waves. <br><br>but you are right that sound is (in part) the result of air trying to flow. yes, it pulses with a total of three consecutive 1-liter pulses every revolution, at the aggregate rate of about 100 L of air per second (no boost). These 1 liter pulses, occuring at 33 times every second (2000 rpm) rush around the valves and into the manifold with great turbulance, all contributing to the spectra of sound waves that launch outward at 750 mph through what ever is there. But this does not completely describe the sounds present in the system. Air trying to flow is part of it. The rest comes from the engine as it generates sound waves in the exhaust air even before it is expelled through the valves. that is, there is acoustic energy present in the exhaust air before it pulses and rushes around in the exhaust system.<br><br>so this is very important: air pulsing and flowing through the exhaust pipe is NOT ITSELF SOUND. sound is created in part as a consequence of these pulses and flows (along with other things), and travels an order of magnetude faster than the air itself does. Even faster than the instantaneous air pulses. (if that were not true, then we would indeed have sonic booms in our trucks).<br><br><br>certainly, peak flows, pulses of air and instantaneous pressure spikes do occur. They cannot be decribed as peak versus rms, but I know what you're getting at with the analogy: the peaks are much larger than the &quot;average&quot; Also, to be precise, the term 'average' is a specific, mathematical expression in the context of a sine wave that is different from its RMS value. So I prefer not to use that term <br><br>maybe you were not speaking to the subject of cooling in this context (of pressure spikes). I still think that cooling depends on average mass air flow, not peak or instantaneous pressures. you have to move hot air molecules out the tail pipe and THAT occurs at average molecule speeds of well under 100 mph. I'm not saying the peak or instantaneous pressures do not impact mass air flow. they certainly do. my point is that we describe cooling efficiency by the ability of the system to move large quantities of air through the system over LONG periods of time (long compared to a single stroke). that is, the important thing to measure is the average mass air flow, not the instantaneous pressures themselves which may or may not contribute. <br>

TXRAM2

jeeesh oil wars are gone now we have sound war you guys are great ;D

doug

[quote author=AlpineRAM link=board=7;threadid=14778;start=60#140186 date=1053439814]<br>Well I can throw in some cents here now ;D<br>When we look at an exhaust cycle of a single cylinder we get a certain distance of pipe filled with hot gas travelling at some speed. (Depending on the diameter of the pipe compared to the size of the cylinder, speed of piston movement, pressure at the end of the exhaust pipe and filling of the cylinder) [/quote]<br>more great comments, AlpineRAM. just to throw some reality checks in here:<br><br>* the CTD bore is 4.02&quot;. Stroke is 4.72&quot; (for the 02s)<br>* the speed of piston movement (its peak velocity) is found by the expression:<br><br>v = A(w) where <br> (w) is angular velocity in radians per second<br> (A) is the peak amplitude of a body in harmonic motion<br> (v) is peak velocity<br><br>at 2000 rpm, the crank makes 33.3 revolutions per second, and this is (33.3)(2*pi) = 209.33 radians per second. The peak amplitude of the piston is 1/2 its stroke (2.36 inches).<br><br>so (A)(w) = (209.33)(2.36) = 494 inches per second, which is about 28 mph.

geer

Wow ??? ;D

AlpineRAM

Doug: Just a moment! I think you left out where I said in correlation to the diameter of the exhaust pipe. I see that you can calculate an average piston speed for the exhaust cycle, but since the exhaust pipe has a smaller diameter than our cylinder I'd like to keep it like that:
About 1l of gas going in a pipe of 5 cm diameter ( 2.5[sup]2[/sup]*3.1415 = 20cm[sup]2[/sup]
Hence one exhaust cycle will fill about 50cm of that pipe.
So gas speed is about 16meters/second at 2000rpm (about 36 mph) average. To get a real world figure for max speed against a certain backpressure the room of the combustion chamber , the area of the valves etc goes into the equation.
This is not taking into consideration that we have a non negligible rest pressure at the end of the working cycle, that we have some gas temperature etc. The practical gas speeds in an exhaust manifold are quite higher.

AlpineRAM

HOHN

Doug-- you said what i was trying to, but only said it better. i should have thought it through a little more.<br><br>Thanks for the great posts, all you guys. I am learning a bunch here.<br><br>HOHN

KATOOM

We seem to be repeating ourselves here. <br><br>In short: You dont need 4&quot; unless you are around 40psi or higher (in conjuction with better intake system). Hot exhaust is bad. We dont know how sound pulses effects exhaust flow. And we will never know untill someone does some serious testing. <br>Am I missing anything?<br><br>Great posts here!!!