Exhaust Physics

[Drax240z]

No it's not crazy, it's the whole principal behind a venturi. The Romans we're doing it more than 2000 years ago to increase their water flow through piping in Rome.

 

Lots of motorcycle exhaust piping uses this theory, as do many drag racing cars, etc.

[Guest 240zJake]

I think the only problem is the size increase necessary to generate the cooling effect. On my 2-stroke the pipe at least doubles in size by the end of the pipe.

[johnc]

Now as air expands.. it cools. technically.. as the exhuast gas exits the engine... wouldn't it be more efficient to go from let's say a 2.25" to a gradual 3" piping in a relatively small distance? The exhuast gases would "expand" as the piping got larger, resulting in cooling, which means a faster flow??

 

One of the many reasons why a cone on the end of a collector is so effective. Below is a picture of an exhaust I built for a 325is. The merge collector has a cone that goes from 2.5" to 3", although its a little hard to see.

 

[clint78z]

Nice exhaust system JohnC, I bet it sounds as good as it looks. You will notice the flare ends are opposite to exhaust flow, this provides way better sealing than the donuts. That's a pretty slick trick with the cones, I would have missed that one.

 

Haven't dealt too much with 2 strokes, I would imagine why the are so sensitive to exhaust tyuning is because they have twice the amount of exhaust pulses as the 4 stroke. Perhaps someone can chime in on the expansion chamber.

 

Did anyone see TRUCKS show, the project copper head had a 572 with electric cutouts. Sounded quite nasty indeed. There was a big worm drive that opened and closed them.

 

So who here has used header wrap?? Any comments on this from N/A or turbo guys. I have heard on most street manifolds, it can collect moisture and create rust. It can also overheat cast iron, when that happens it lower the strength of it. This can lead to cracking, and failure. I have heard of good things about cermic coatings on pistons and inside headers as well.

[2126]

Clint, Your thoughts on header wrap are correct. It's my understanding that the use of header wrap or the ceramic coated headers is to prevent temperature loss in the exhaust gases. Basically, hot gases flow faster than cooler gases, given the same flow path. Just my two cents worth.

 

What a great post....one of the best in a while.

[Nismo280zEd]

I use headerwrap for a different reason, keep the engine comp cooler, and avoid another fire due to my previous open header. Heat shields are great things... they SHOULD be used.

-Ed

[Scottie-GNZ]

I have used header wrap on the DP of the 3.8L turbo engine for over 4yrs for the 2 reasons listed. However, it should only be used with SS.

[Guest KEINoze]

A very basic way to visualize exhaust flow is to pretend water would be flowing through your piping. At least this is what I do when looking for a new exhaust design. "So how can the water be flowed the fastest through the exhaust?"

Exhaust pulses are generally in sync by the time it exits the header or collector "Y" pipe. The role of the rest of the exhaust to be to minimize turbulence in the flow.

 

Is my logic correct here?

[2126]

I believe your idea is good for descriptive purposes however, the real difference is that gases are compressible and the liquids are not. Theres alot more to designing an effective exhaust system than most people think, but your thoughts about turbulance (restrictions) is correct.

[Guest tony78_280z]

Physics: Liquids can be compressed as well, solids too. It just takes more pressure to compress. So I can see how visualizing a less compressable form could actualy help design the best flowing system. Just $0.02 here

 

Another question. Why are header collector ends considerably smaller than the pipe that they will go into? My headres were advertised as 2 1/4. But a 2 1/4 pipe is a bit bigger around than the collector's end.

[2126]

Physics: Liquids can be compressed as well' date=' solids too. It just takes more pressure to compress. So I can see how visualizing a less compressable form could actualy help design the best flowing system. Just $0.02 here

 

Another question. Why are header collector ends considerably smaller than the pipe that they will go into? My headres were advertised as 2 1/4. But a 2 1/4 pipe is a bit bigger around than the collector's end.[/quote']

 

Ok, technically speaking liquids can be compressed, although the reality of it is that it is so minute that for all practical purposes it is considered non-compressible. I'm sure you are familiar with hydraulics systems!!?

 

To shed light on your second question about collectors...by reducing the exit size of a collector (compared to the inlet side of a collector) it basically creates a venturi. It increases the velocity of the exhaust gases to improve upon a desired scavenging of the gases. I'm sure there is more to it than that, but I'm not a thermodynamics professor. Another obvious reason for the smaller diameter is so you can attach an exhaust pipe. Why yours is sized the way it is is beyond me....maybe poor quality control?

[SHO-Z]

As a compressible gas, exhaust, travels down a straight pipe the speed of the gas in the center accelerates. There is a layer of gas on the sides of the pipe called the boundary layer that for all practical purposes is not moving. Of you looked at a velocity profile it would have a bullet shape. As the velocity in the center increases the boundary layer increases in thickness and the bullet shape gets more pointed. There is a point on a long straight pipe where the gas reaches sonic velocity and the flow is choked, the gas will not flow over sonic velocity. This will not happen on the length of an exhaust system. Any type of change of direction will disrupt the flow and cause the boundary layer to decrease.

 

I would not be concerned with the bends required to install an exhaust system. Most of the ones used are at least 3 diameter bends and really do not effect the back pressure greatly.

 

There is a whole lot more going on with the pressure waves and such than I would like to think about.

[Guest tony78_280z]

Another common thing I hear is "Backpressure creates more torque" This is not directly true, and I will explain why. Backpressure due to friction loss is an evil thing, and will always steal the amount of power and engine can make. It is true that a 2.5" exhaust will usually create a bigger torque number at a lower rpm than a 3". It is not the backpressure which gives you the more torque it is the velocity of the gas at that rpm that gives you the bigger number.

Once again the topic of backpressure has come up with my associates. They say some (undescribable yet magical) amount of backpressure is good for an engine. And I go on to tell them that it is a myth, and I don't believe it. Is there some irrefutable source that I can quote facts from to end or support this myth? I don't take "I saw it on discovery channel", or "I read it on the internet." As bases for facts, so I can't expect others to take word from this site as fact.

[2126]

It's always interesting, regarding technical discussions, the nomenclature being used! What I'm getting at is, more times than not people will use a particular word or phrase to describe something and in actuality they are using the wrong descriptive word (nomenclature), even though they are correct about the topic they are discussing. So, what happens to the discussion? It usually turns into a lengthy debate because the people in the discussion are actually saying the same thing but using the wrong words to describe something. Now, I'm not saying this discussion is going that way, but thought it would be appropriate to bring up this bit of information for better future discussions. The definition "backpressure" is one of those words that seems to be over used in discussions regarding exhaust systems and is likely selected to describe a particular phenonema that is not fully understood by the individual trying to describe it. So, for the techie tip of the day.....learn and use proper nomenclature when dicussing technical issues. What do you all think? Sorry for the sidetrack!

[johnc]

As I've said before, there's no such thing as "backpressure." The exhaust system gets pressurized to some degree because you are restricting the flow of a gas. The pressure doesn't go "back" somewhere.

 

A pressure number of 2 to 3 psi measured immediately after the header collector is a good number to shoot for.

[clint78z]

Perhaps TimZ can chime in with some numbers he was seeing preturbo in the exhaust manifold. I think he did some testing to size his exhaust turbine. The pressure seen can help you size the A/R ratio on the housing. This can tell you if your housing is restricting the flow on the top end.

 

It is not a good idea to put a restriction in a system for the sake of getting a certain pressure number. The only benifit you get from that is lowering the noise, that can eb done other ways. A restriction is just that it restricts the flow of exhaust gases. Then engine wastes more power trying to pump the exhaust gas out, instead of turning your wheels. It is more important to focus on gas velocity and scaveging effect, tuning for a rev range.

[1 tuff z]

i've had a header "blanket" on my header for about 10 years now. i did it in an attempt to keep heat away from my jcr supercharger intake manifold which is in very close proximity [less than 0.5"] to the header. the 6 into 1 header is coated using swain technology's "white lightning" thermal barrier coating. no corrosion, rust or other problems noted so far...

[pparaska]

Great thread!

 

SHO-Z really hit it I think - don't worry about the bends too much. Make sure the size transitions are slow (like Johnc's cone), the pipes are of adequate size to not have too much pressure change from atmosphere to in the pipe, etc., but the big issue is the wave tuning and scavenging.

 

David Vizard did a really nice article recently on exhaust system theory and testing in the May 2005 Popular Hotrodding Magazine. "Exhaust Science Demystified" (byline: "The fact is most cars are leaving horsepower on the table. We show you how to get it all back.")

 

He talks about how the (then) Cyclone Sonic Turbo muffler design he and an acoustics expert friend designed made more horsepower in a dual 2.25" system than all the other dual 2.5" systems they tested using other mufflers.

 

Also intersting is his discussion on building a no-loss exhaust system - that's right - an exhaust system that saps NO power compared to a correctly designed open header setup for racing.

 

He talks about the 5 cycles of a racing engine - the added one being the exhaust-driven induction event - what you have heard called the scavenging effect that pulls intake mixture into the cylinder during overlap.

 

He describes a comparison of induction pressures between the peak suction on the intake port due to the piston going down the bore (about 0.5 to 1 psi) and the partial vacuum that an exhaust scavenging event has on the intake port during overlap (about 4-5 psi).

 

David goes on to talk about primary header pipe diameters, saying a bit too small is better than a bit too big, and that the only real way to find out what the engine needs is to do dyno testing. However, he provides a nomogram showing the relationship between the primary pipe diameters that have been proven to work well versus exhaust port flow at peak lift, for street, hot street, and race Normally Aspirated engines. He adds that the primary pipe AREA must be increased by 6 to 7% for every 50hp worth of N20 injected., but that for street use, it's better to not do the increase as the engine is mostly in non-N20 use.

 

Primary length is also addressed. David always boohoos the "equal length" argument. In this article, he gives some good reasons - you may not know the optimal length, so if the pipes are all of equal, but incorrect length, that may be worse than having some variation in the lengths. He notes that an exhaust system can scavenge at or near maximum intensity over a 4000 rpm range. His reasoning is then that if one pipe is 1000 rpm different in its rpm range of scavenging from the rest, pipe length differences of as much as 9 inches different have little effect on performance.

 

Then he gets into an issue that came up in my mind recently when reading "Headers-by-Ed"'s header information. He's a big proponent of very equal length headers. As I was reading his theory, it seemed based on the assumption that the pulses were evenly spaced in time between the pipes. (He mostly does regular V8 headers). But as Vizard points out, the timing of the pulses on one bank of a two plane crank V8 are unevenly spaced (90, 180, 270, 180 degrees, etc.). Vizard likens this to the the collector on one bank of a two-plane crank V8 acting like it were on a 3 cylinder engine with different sized cylinders. He notes that the varied spacing makes the system less sensitive to primary length. He goes on to say that dyno tests on two-plane crank V8s show little sensitivity with primary lengths between 24 and 36".

 

Collectors (secondaries): Start with a ratio of collector to primary diameter of 1.75:1. As for collector length, performance is more sensitive to it than primary length, and collector diameter and length have more effect on performance than primary length. Basic rule: shorter, larger diameter collectors favor top end, longer, smaller diameter collectors favor bottom end. He also says that for an engine with peak hp from 6000-8500 rpm, the collector length of 10-20 inches is effective. Again, shorter for higher rpm band (4500-8500).

 

Mufflers: Other than noise reduction, get a high enough flowing muffler to handle the HP. Don't use a muffler that flows less than the correctly sized secondary piping. For each peak HP, have the total muffler flow be 2.2cfm. He gives a graph of percentage of maximum power retained with a muffler on versus cfm per open pipe horse power to show that at 1.5cfm/hp, you lose about 17% of your peak hp, 5% loss at 1.75cfm/hp, and 2% loss at 2.00cfm/hp. At 2.25cfm/hp and beyond, the loss is less than 1%! On the same graph is a curve for backpressure versus cfm per open pipe horsepower:

 

1.50cfm/hp gives 2.0psi backpressure

1.75cfm/hp gives 1.1psi

2.00cfm/hp gives 0.3psi

2.25cfm/hp gives 0.1psi

 

(sorry, Johnc, David Vizard uses the term backpressure - it's just the difference between the average pressure in the pipe versus atmospheric pressure).

 

David goes on to talk about pressure waves and mufflers, the flowmaster, glasspacks, etc. I won't go into that part...buy the magazine!

 

What is REALLY interesting is the "pressure wave termination box" (or resonator box). What this is is a large volume box (muffler, empty, etc.) that is at least 8 times the volume of one of the engines cylinders. In the V8 case, as large as the engine displacement! It's inlet pipe extends into the box, and the end of the pipe is the end of the collector that the open header would have for best performance. What this does is make the rest of the exhaust system virtually have no effect (pressure wave wise) on the headers. If you have little backpressure behind the resonator box (piping, muffflers, tailpipes, etc.) the engine will lose no power. Without this box, the rest of the exhaust just causes the collector to get REALLY long, as far as the engine/header combo is concerned.

 

He covers X and H pipes. Less noise and increased power is the result. Read the article!

 

This is a really great exhaust science article. I urge anyone interested in this stuff to get the mag! What I like about David's stuff is that he has the theoretical background to make most any hotrodder bow down, but then backs up all his theory with TEST data, from tests designed by someone who understands the scientific principle and knows how to keep from chasing your tail by not changing too many (seemingly unrelated) variables at once. His review in this article of magazine muffler tests is a great example of this insight.

 

I don't know if even now in early April that you'll find the magazine on the rack, but to get the May 2005 as a back issue, you can do this:

MAGAZINE BACK ISSUES

To order back issues, call McMullen Argus' back issues department at:

U.S. Customers (866) 601-5199

International Customers (714) 712-2130

For back-issue questions, including availability, please send your query, including the magazine name, month and issue, to: backissues-mailorder@primedia.com

 

Sheesh! I think my post count should be triple incremented for this one

[Guest tony78_280z]

Great, now I gotta see if I can get a copy of that magazine! How come you couldn't have mentioned this back in May?

 

How about you make up for it by breaking copywrite laws and scan and post the article! (Just Kidding)

 

(sorry, Johnc, David Vizard uses the term backpressure - it's just the difference between the average pressure in the pipe versus atmospheric pressure).

The differance in in-pipe pressure versus atmospheric should allways be in-pipe > atmosphere, unless the motor is off and then the pressure would equalize. Right?

 

Since this thread resurfaced I'll ask a few question I've been wondering lately.

 

1> Is there a product that acts like header wraps in displacing and moving heat out of the engine bay that can be added without risking the moisture and rust trap that these wraps are known for? Ive heard of ceramic coating acting like header wraps, but my headers/exhaust is not ceramic coated. I hadn't heard of this technique until after I had my exhaust purchased and installed. What would be realy cool if there was some sort of dip/brush/spray on ceramic coating a person could add to their exhaust system.

 

2> Someone said something about somehow (don't you hate it when a question starts off like that) the exhaust valves could be open at the wrong time and actualy suck air/exhaust back into the combustion chamber. How could this be? I think it'd only be possible on miss matched components particularly in cam and heads. Is it possible for a stock motor (like mine) to have this effect? Also, this would be simply a horrible perfomance loss, would it not?

[pparaska]

Great' date=' now I gotta see if I can get a copy of that magazine! How come you couldn't have mentioned this back in May?

 

How about you make up for it by breaking copywrite laws and scan and post the article! (Just Kidding)[/quote']

 

Sorry, I just saw this thread last night. No, not interested in breaking copyright laws . Besides, David deserves the royalties on each copy of that article sold!

 

The differance in in-pipe pressure versus atmospheric should allways be in-pipe > atmosphere, unless the motor is off and then the pressure would equalize. Right?

 

Yes, while it's running. A typical gage shows positive pressure or negative pressure (vacuum) that is in relation to the ambient pressure around the gage. So just hooking the gage to the exhaust pipe shows the pressure difference to atmosphere. This is typically called gage pressure, versus absolute pressure which is gage pressure plus the atmospheric pressure.