exhaust knowledge

[johnl]

Oh goody, an excuse for more theorising! Sorry that this is another long post, and may go over some of the same ground as my previous post, but I feel I need to do this in order to make a cogent argument. It’s still just my own understanding, which may not hit the target exactly in the bull’s eye. I’m not trying to suggest that what Camurai has said is incorrect (I don’t think it is, but IMO may be a little misleading due to omission), just following the rabbit he released down the hole.

an exhaust can reach a point where it is so high-flowing that it will begin to draw unspent air/fuel mixture out the exhaust valve during overlap, resulting in reduced fuel mileage.

This occurs when significant valve overlap exists, and when a strong negative pressure wave arrives at the open exhaust valve, and is a good thing for power (though not necessarily economy) because more waste exhaust gas can flow into the low pressure zone that momentarily exists in the exhaust port, and more intake mixture can then be sucked into the cylinder (or more correctly, 'pushed' in by atmospheric pressure, there being no such thing as 'suction', but that’s an argument for another day!).

This isn't due to an exhaust being “high-flowing” due to large diameter and lack of back pressure so much as being ‘high-flowing’ due to harmonic pressure waves in the exhaust pipe creating a strong scavenging affect (mostly as a result of the header dimensions). If the pipe ID is too large then the pipe may be ‘free-flowing’ in a non dynamic / non resonating manner, but the scavenging affect will be diminished due to weakening of the pressure waves, and this may have an adverse affect on ‘flow’ and thus power despite lack of back pressure. The pipe being too large shouldn’t have a direct adverse affect on economy, though it would on power (which itself might indirectly affect economy due to the need to use larger throttle openings to produce the required power, sorry to split hairs).

Having said that, the beneficial affects of pressure waves may be diminished if back pressure is substantial. This is because the back pressure strangles exhaust flow, increasing ‘ambient’ pressure in the exhaust port even when the negative pressure wave arrives at the valve, so power is lessened despite correct harmonics (i.e. the high back pressure will prevent the pressure of the negative wave being as low as it otherwise would be if back pressure were also lower).

When scavenging works very well not only is nearly all exhaust gas removed from the cylinder (though not all as some % will remain whatever because of turbulent mixing of the fresh and spent gasses in the cylinder), but some of the incoming intake mix will also flow through the cylinder and out through the exhaust valve (exhaust gas being expelled from the cylinder by it's own high pressure, the pumping action of the rising piston, and by the inertia of the exhaust gas moving at high speed as it escapes out the exhaust valve into a zone of lower pressure).

This is because the intake valve opens before the exhaust valve closes (overlap), allowing the incoming mixture to be exposed to the low pressure that exists in the cylinder and exhaust port as the last of the exhaust gas is leaving the cylinder, some intake mix then following the exhaust out the port and lost to combustion and power production (the affect being stronger the lower the momentary pressure in the exhaust port is and the greater the valve overlap). This is unfortunate for economy but the price you pay for better cylinder filling.

This low pressure in the cylinder lasts longer the greater the valve overlap, allowing the incoming mix to more easily continue flowing into the cylinder (and some % continuing out the exhaust port) at an ever increasing speed until the exhaust valve closes. But, since the incoming mix is now flowing at high speed (higher speed still due to the initial exposure to a lower pressure in the cylinder) and has inertia it continues to 'ram' into the cylinder despite the exhaust valve now being shut (more so than would be the case just from the piston falling in the cylinder), and in perfect conditions can mean that when the inlet also closes (trapping the intake mixture) the pressure inside the cylinder can be slightly higher than atmospheric even before the piston starts to compress it (i.e. it's like a slight supercharging effect).

This 'ram' effect of the incoming mix can indeed be so strong that the inlet valve can remain open (for a very short time) even after the piston starts rising and compressing the mixture. In this case the mixture is effectively being compressed from two ends, one end being from the rising piston and the other being from the inertia of the still incoming mixture 'ramming' into the cylinder. Of course this can't last too long or pressure rise caused by the rising piston will overcome the inertia of the incoming mix and start pushing it back out the inlet valve.

This all only works well at a particular point in the rpm range, the affect dropping away to each side of that point in the range. At 'wrong’ rpm the negative pressure wave won't be there (or will be but won’t be so low in pressure) when the exhaust valve is open and the affect will be less, and indeed at some rpm (usually much lower rpm) there may be a positive wave actively 'pushing' some exhaust gas back into the cylinder, diluting the incoming mix with inert and hot gasses (which needless to say isn't a good thing for ‘quality’ cylinder filling and thus power, or even smooth running in extreme cases, being why many racing engines idle so poorly and require higher idle speed just to avoid stalling). To make things worse, if intake gas speed isn’t quite high (i.e. rpm quite high) and thus the kinetic energy in the incoming intake gas also quite high, an engine with ‘adventurous’ intake timing may find the piston pushing intake mix back out the intake port as the piston rises, further reducing power when not in the ideal rpm range.

Some fuel is wasted (i.e. not burned) because of this exhaust gas scavenging affect, but on the plus side the fuel / air mix that is trapped in the cylinder is at a higher than otherwise possible pressure (before and after compression), and is 'cleaner' than it otherwise would be, i.e. more of the inert gasses (as far as combustion is concerned) are eliminated from the cylinder, leaving a more potent combustible charge.

With large valve overlaps etc, even when the engine is in the 'sweet' rpm range power can be very good but economy can be poor (compared to more conservative overlap) due to enthusiastic scavenging affects drawing unburned mixture out through the exhaust port, but scavenging affect is less outside the 'sweet' part of the rpm range so it would be easy to think that economy ought to be OK when not in the ideal rpm range. However such engines also tend to have poor economy (compared to an engine with more conservative overlap etc) even when driven fairly sedately.

I suspect what happens is that large valve overlap causes the engine to lose a significant amount of intake mixture through the exhaust port even when the engine is being used on a lighter throttle opening at a lower rpm, where some unwanted scavenging is occurring but not as much as at higher rpm (and indeed not required as high power isn’t being required of the engine, therefore scavenging at lighter throttle openings is wasteful). I tend to think that poor average economy with ‘sportier’ engines is more likely to be more a function of valve timing than pipe design, though pipe design will have some interactive affect.

It may be that in such circumstances where a significant scavenging affect is causing poor economy with sedate driving, that in these conditions (lower rpm and / or lighter throttle openings) a higher back pressure may prevent some of the intake mix loss down the exhaust by counteracting the scavenging strength to some degree, and this might improve economy and possibly may also increase low rpm torque (and response at lighter throttles), but at the expense of strangling the engine higher up, which kind of defeats the purpose of larger valve overlaps and tuned length headers etc…

[johnl]

Good stuff, John... and as far as my own studies go, every word of it seems correct. I'm no expert either, but it's good when educated amateur knowledge coincides.

Or a worry! I’ve commented on some of your statements below, but keep in mind that all of this is IMO and some is speculation based on background understandings and not based on practical experimentation on my part. If I make statements that sound like I’m stating ‘fact’, then it’s only because continually qualifying every statement with ‘IMO’ becomes tiresome. Feel free to disagree!

The best way to look at the entire engine is that it's an air pump. You want to be able to flow as much air through the engine as efficiently as possible. The whole idea of fuel and combustion complicates this a bit, as well as the mechanical efficiency of the engine and transmission... but overall, in it's simplest form, the engine is an air pump.

Of course I agree, in it’s simplest concept an engine is a self powered pump that fortunately for us creates more power than is required just to run itself. Getting gas to pass through the pump as efficiently as possible and thus produce the greatest power is complicated by the addition of internal combustion, but the greatest complication is created by the intermittent nature of the flow required both into and out of the pump. This creates stop / start flow and resonant affects that are very complex, and can be very disadvantageous or very advantageous depending on how well we understand and use them.

Reversion is when the spent gasses push their way back into the combustion chamber due to strong backpressure. Never good.

Unless truly excessive I doubt back pressure alone would be enough to cause a substantial reversion, though it will reduce the effectiveness of negative pressure waves.

The exiting exhaust gas will be at a much higher pressure than the ambient back pressure, except perhaps for the last bit coming out of the chamber, and this bit could be ‘reversed’ by back pressure. Of course every bit of exhaust gas not ‘extracted’ will dilute the incoming charge, and potentially indirectly cause less efficient induction flow. I think more likely to cause substantial reversion would be positive pressure waves reflected back toward the exhaust port from a constriction somewhere in the exhaust after the header pipes, as you might get from a reduction in pipe size or baffles in a muffler.

High performance two stroke engines typically have much higher back pressure than four stroke engines, but this is used deliberately to create strong reflected positive pressure waves (reflecting off the reverse cone toward the end of the muffler) through a particular range of rpm in order for intake mixture that is coming out of the exhaust port to be pushed back into the cylinder (i.e. reversion). The amount of intake mix coming out the exhaust in the first place is largely due to the slightly earlier negative pressure wave being ‘reflected’ back from the first divergent cone, and can be a quite substantial amount of intake mix with a 2 stroke (much more than a 4 stroke). The back pressure itself is still disadvantageous in itself, but it does allow the production of a stronger positive pressure wave that in a 2 stroke application is more advantageous than the disadvantage of the higher back pressure.

Stagnation can occur with TOO wide of an exhaust system. Think of it as a wide hallway with people standing around talking on either side... The hallway might be wide, but it's wide enough for people to gather and hang out "out of the way"... when a crowd of people that would nearly fill the hallway wants to come running through, it's not going to work! That wide hallway, thanks to it's size accommodating so many idle people, seems to be a much more narrow space than it really is! This creates a good deal of turbulence as well... which leads to increased backpressure.

I have some difficulty with the concept of ‘stagnation’ occurring in larger pipes. I suspect it may be one of those convenient terms sometimes used as a kind of shorthand to avoid more fully expressing a more complex concept, useful but possibly not entirely accurate, like ‘suction’.

I think your analogy is flawed with regard to demonstrating your point. The actuality will be that the wider the hallway the more people will be able to easily pass through it even if there are people loitering near the walls. If the hall is narrower then these people wouldn’t be able to ‘loiter’ as they’d be swept up in the general flow if the flow is highish, but the hallway is still narrow so only a limited number can pass through in a given time. In a wider hall they’ll still get swept up when flow volume increases, they won’t be able to stand firm and push against the flow to push it back or even restrict it, unless their feet are nailed to the floor. If flow doesn’t increase to the point at which these people are swept up, then the hall is still wide and people will still pass through easily. My understanding is that problems with boundary layers creating ‘drag’ in pipes increase as the pipe gets smaller, not bigger. Even if the boundary layer in a larger pipe may be ‘thicker’, it tends to be less ‘intense’.

The same is true for back pressure in large or small pipes, the larger pipe will always present less restriction to a non fluctuating flow. With a fluctuating flow in a smaller pipe we can arrange for the flow to speed up and slow down substantially more than we can with a larger pipe, even if the potential for non fluctuating flow in a larger pipe is greater. A fluctuating flow can momentarily speed up over and above what’s possible with non fluctuating flow, but also will conversely momentarily slow down below a non fluctuating flow. Arranging the fluctuating flow so it’s speed is fastest in a particular section of the pipe at the appropriate moment for greatest cylinder filling and slowing when we don’t need it to be fast is the art of exhaust tuning.

An overly large pipe will I think cause weaker cylinder filling and thus power loss due to weakening of the pressure wave strength, not because it has less potential for non fluctuating flow (it has more). This is more so for the header pipes, and a very large main pipe may in fact be beneficial to the strength of a wave reflected back from the end of the header pipe itself, creating a stronger returning negative pressure wave back up the pipe and thus a stronger surge in power at a particular point in the rpm range. However, if the main pipe is very large there might be a reduction in the strength of secondary negative pressure waves generated further along the pipe (than the header pipe ending) that take longer to pass back up the system than the primary pressure waves, and might have been be useful at lower rpm, therefore resulting in a power loss at lower rpm. Therefore, if the headers are a good size, but the main pipe too large, there may be a good power spike higher in the rpm range but a loss of useful power lower down.

I'm just adding on to what John said above. Exhaust theory isn't difficult in the beginning stages. When you get to sound waves, pressure changes, and all that crazy stuff, THEN it gets beyond my knowledge.

I reckon exhaust theory starts off pretty hard to understand and just gets more complex from there! It’s one of those things where a little knowledge is probably a dangerous thing, there are so many interacting variables. I keep finding this when trying to explain something (as I see it) in this thread, i.e. I explain X, but then have to add more to explain why X is only the case in certain conditions and not in others. You end up explaining the explanation, then expanding on it further to avoid misleading statements. It’s a bloody can of worms!

For the non expert (myself included), it’s better to select exhaust components designed and tested (important) by experts for the application we might have. It has to be the right system for the application though, quite similar looking systems may perform very differently, and may or may not be completely unsuitable for our particular engine set up and intended usage. A racing exhaust on a non racing engine may look cool, but you’ll probably be going backwards compared to even the stock system.

PS We do need to keep in mind that the harmonic pressure waves are indeed sonic waves, i.e. pressure fluctuations created by high energy sound. The reason why both negative and positive sonic waves don’t both travel at exactly the ‘speed of sound’ in the confines of an exhaust pipe is because the confines of the pipe allow substantial pressure fluctuations, and sound travels faster in higher pressure gas and slower in lower pressure gas.

[PakaloloHonda]

OK! I think I'm finally startin to get some of this .. a bit .. maybe, a jist ..
I think if I read it a few more times .. it'll sink in a tad better .. ***faint***

A Hui Hou !!!
Tomi

[93 CB7 SE]

Oh yeah.

Honestly, an H23 with that kind of nitrous on it would probably benefit from 3". But I don't know that I would want that much exhaust off the bottle.

thanks

[verderacer]

Well said !

Or a worry! I’ve commented on some of your statements below, but keep in mind that all of this is IMO and some is speculation based on background understandings and not based on practical experimentation on my part. If I make statements that sound like I’m stating ‘fact’, then it’s only because continually qualifying every statement with ‘IMO’ becomes tiresome. Feel free to disagree!



Of course I agree, in it’s simplest concept an engine is a self powered pump that fortunately for us creates more power than is required just to run itself. Getting gas to pass through the pump as efficiently as possible and thus produce the greatest power is complicated by the addition of internal combustion, but the greatest complication is created by the intermittent nature of the flow required both into and out of the pump. This creates stop / start flow and resonant affects that are very complex, and can be very disadvantageous or very advantageous depending on how well we understand and use them.



Unless truly excessive I doubt back pressure alone would be enough to cause a substantial reversion, though it will reduce the effectiveness of negative pressure waves.

The exiting exhaust gas will be at a much higher pressure than the ambient back pressure, except perhaps for the last bit coming out of the chamber, and this bit could be ‘reversed’ by back pressure. Of course every bit of exhaust gas not ‘extracted’ will dilute the incoming charge, and potentially indirectly cause less efficient induction flow. I think more likely to cause substantial reversion would be positive pressure waves reflected back toward the exhaust port from a constriction somewhere in the exhaust after the header pipes, as you might get from a reduction in pipe size or baffles in a muffler.

High performance two stroke engines typically have much higher back pressure than four stroke engines, but this is used deliberately to create strong reflected positive pressure waves (reflecting off the reverse cone toward the end of the muffler) through a particular range of rpm in order for intake mixture that is coming out of the exhaust port to be pushed back into the cylinder (i.e. reversion). The amount of intake mix coming out the exhaust in the first place is largely due to the slightly earlier negative pressure wave being ‘reflected’ back from the first divergent cone, and can be a quite substantial amount of intake mix with a 2 stroke (much more than a 4 stroke). The back pressure itself is still disadvantageous in itself, but it does allow the production of a stronger positive pressure wave that in a 2 stroke application is more advantageous than the disadvantage of the higher back pressure.



I have some difficulty with the concept of ‘stagnation’ occurring in larger pipes. I suspect it may be one of those convenient terms sometimes used as a kind of shorthand to avoid more fully expressing a more complex concept, useful but possibly not entirely accurate, like ‘suction’.

I think your analogy is flawed with regard to demonstrating your point. The actuality will be that the wider the hallway the more people will be able to easily pass through it even if there are people loitering near the walls. If the hall is narrower then these people wouldn’t be able to ‘loiter’ as they’d be swept up in the general flow if the flow is highish, but the hallway is still narrow so only a limited number can pass through in a given time. In a wider hall they’ll still get swept up when flow volume increases, they won’t be able to stand firm and push against the flow to push it back or even restrict it, unless their feet are nailed to the floor. If flow doesn’t increase to the point at which these people are swept up, then the hall is still wide and people will still pass through easily. My understanding is that problems with boundary layers creating ‘drag’ in pipes increase as the pipe gets smaller, not bigger. Even if the boundary layer in a larger pipe may be ‘thicker’, it tends to be less ‘intense’.

The same is true for back pressure in large or small pipes, the larger pipe will always present less restriction to a non fluctuating flow. With a fluctuating flow in a smaller pipe we can arrange for the flow to speed up and slow down substantially more than we can with a larger pipe, even if the potential for non fluctuating flow in a larger pipe is greater. A fluctuating flow can momentarily speed up over and above what’s possible with non fluctuating flow, but also will conversely momentarily slow down below a non fluctuating flow. Arranging the fluctuating flow so it’s speed is fastest in a particular section of the pipe at the appropriate moment for greatest cylinder filling and slowing when we don’t need it to be fast is the art of exhaust tuning.

An overly large pipe will I think cause weaker cylinder filling and thus power loss due to weakening of the pressure wave strength, not because it has less potential for non fluctuating flow (it has more). This is more so for the header pipes, and a very large main pipe may in fact be beneficial to the strength of a wave reflected back from the end of the header pipe itself, creating a stronger returning negative pressure wave back up the pipe and thus a stronger surge in power at a particular point in the rpm range. However, if the main pipe is very large there might be a reduction in the strength of secondary negative pressure waves generated further along the pipe (than the header pipe ending) that take longer to pass back up the system than the primary pressure waves, and might have been be useful at lower rpm, therefore resulting in a power loss at lower rpm. Therefore, if the headers are a good size, but the main pipe too large, there may be a good power spike higher in the rpm range but a loss of useful power lower down.



I reckon exhaust theory starts off pretty hard to understand and just gets more complex from there! It’s one of those things where a little knowledge is probably a dangerous thing, there are so many interacting variables. I keep finding this when trying to explain something (as I see it) in this thread, i.e. I explain X, but then have to add more to explain why X is only the case in certain conditions and not in others. You end up explaining the explanation, then expanding on it further to avoid misleading statements. It’s a bloody can of worms!

For the non expert (myself included), it’s better to select exhaust components designed and tested (important) by experts for the application we might have. It has to be the right system for the application though, quite similar looking systems may perform very differently, and may or may not be completely unsuitable for our particular engine set up and intended usage. A racing exhaust on a non racing engine may look cool, but you’ll probably be going backwards compared to even the stock system.

PS We do need to keep in mind that the harmonic pressure waves are indeed sonic waves, i.e. pressure fluctuations created by high energy sound. The reason why both negative and positive sonic waves don’t both travel at exactly the ‘speed of sound’ in the confines of an exhaust pipe is because the confines of the pipe allow substantial pressure fluctuations, and sound travels faster in higher pressure gas and slower in lower pressure gas.

[johnl]

But anyways, you want some real exhaust knowledge, look up David Vizard.

I would recommend reading anything by David Vizard. Some of my first reading on car performance and theory etc was authored by him.

[71_307]

Wow, that's great. I would rather have a car that isn't a plastic peice of shit as opposed to a SS exhaust manifold.

The metal on them is so thick, that it would take DECADES for them to rust through. The exhaust system will rust through long before these do. Of course I bet there is a good chance your full SL2 exhaust wasn't stainless...

wowowo I was just asking... and SL2's are great little cars... polymer is a great product, couldnt tell you how many times it saved me.. and ZERO dings... I had a 94 and it still looked brand new 10 years later.

[owequitit]

wowowo I was just asking... and SL2's are great little cars... polymer is a great product, couldnt tell you how many times it saved me.. and ZERO dings... I had a 94 and it still looked brand new 10 years later.

On average, they do not last as long as your typical Honda. That is why they are very hard to find nowadays.

My point was that no car is perfect. I would rather have a mild steel EM, than a car that has many other problems.

The SL2 was thrashy, crude, and depending on who dealt with it, a PITA. Polymer was all fun and games until a panel shattered that shouldn't have. Sometimes dents are better. Not to mention the paint quality issues, and the fit and finish.

I don't dislike Saturns, but there is a lot more to a quality car than an SS exhaust manifold. I think it has been proven that Honda knows how to build a quality product.

[Camurai]

On average, they do not last as long as your typical Honda. That is why they are very hard to find nowadays.

My point was that no car is perfect. I would rather have a mild steel EM, than a car that has many other problems.

The SL2 was thrashy, crude, and depending on who dealt with it, a PITA. Polymer was all fun and games until a panel shattered that shouldn't have. Sometimes dents are better. Not to mention the paint quality issues, and the fit and finish.

I don't dislike Saturns, but there is a lot more to a quality car than an SS exhaust manifold. I think it has been proven that Honda knows how to build a quality product.

X2. My 15 year old car is tight, quiet and comfortable, while every Saturn I have ever ridden in that even comes close to that age is a noisy, harsh and uncivilized ride. I'll take my cast iron exhaust manifold, TYVM. It's a long sight better than the poor quality of an just about everything on an older Saturn.

[d112crzy]

Holy fuck, I think we have a new owequitit.

[fatboy1185]

after all these years, why are people still talking about exhaust size on a car that is almost 20 years old??? its not like this is some new technology that just came out . look at what the big boys do, copy cat them, and be happy.

[Camurai]

Holy fuck, I think we have a new owequitit.

Why? Because I'm not a big fan of a car with poor build quality? If that's the case, there are a shitload of owequitits out there.

[F22HB]

Umm, hi exaust master. Hows it goin..
What can you tell me about unequal length headers? I've been thinking about making a set for my accord to give it the subi sound.

[d112crzy]

Why? Because I'm not a big fan of a car with poor build quality? If that's the case, there are a shitload of owequitits out there.

calm down there buddy. My post wasn't even referenced to you.

Anyone who's been here long enough knows what I was talking about.

That john l guy writes novels everytime he posts, just like owequitit.

[owequitit]

Holy fuck, I think we have a new owequitit.

LOL! Thank God. My hands are wearing out. LOL!

Edit, on average lately, I think my posts are about 10% of what they used to be.