Contact US

Log In

Come Join Us!

Are you an
Engineering professional?
Join Eng-Tips Forums!
  • Talk With Other Members
  • Be Notified Of Responses
    To Your Posts
  • Keyword Search
  • One-Click Access To Your
    Favorite Forums
  • Automated Signatures
    On Your Posts
  • Best Of All, It's Free!

*Eng-Tips's functionality depends on members receiving e-mail. By joining you are opting in to receive e-mail.

Posting Guidelines

Promoting, selling, recruiting, coursework and thesis posting is forbidden.

Students Click Here

Exhaust debate

Exhaust debate

Exhaust debate

am having a friendly debate.

if you have a 3" y pipe tied into a
3.5" peice of straight exhaust pipe...no muffler..6' long..

would it have any advantage over a 3" y pipe tied into a
3" pipe of same configuration.

ie...would not the restriction at the 3" collector cancel any advantage of the 3.5" pipe.

ie...my "opponent" states it will require more HP with the 3" pipe.


RE: Exhaust debate

Here's my .02, It seems to me that it would be hard to judge w/o some measurements.  While the 3" pipe has more restriction than the 3.5", we don't know just how much.  I don't however see much problem going from a 3" Y pipe to a 3.5" tailpipe.  I've heard people recommend 3" pipes in places so I'm thinking they may offer some help with the exhaust flow dynamics and scavenging.  Therefore, the 3" collector may not cancel any benefit of the 3.5" tailpipe.  In any case, the short 3" pipes will have less restriction than a 3" pipe running all the way back.

RE: Exhaust debate

Without a LOT more information there is just no way to give you an answer.  Even then it would be a guess without a dyno test.  As an example---Lotus engined road racing car with a 2 1/2 inch collector dumping into  a 50something inch 3 inch tailpipe GAINED hp , torque, and broadened it's rev range by going to a 2 inch collector and a 2 inch tailpipe WITH a 5 inch Supertrapp with 16 plates!!!
A case of bigger is NOT better!  It made it under the 105db limit, too.  Louder is not faster!


RE: Exhaust debate


Isn't it amazing what sells? More of anything isn't always better in reality but the concept sure sells!

I have one question about exhaust that I've yet to see discussed.  If say 1000 cfm of exhaust gas leaves an engine at 1100°F, by the time it reaches the end of the tailpipe, its temperature will have dropped to about 300°F and there will be a corresponding significant loss in volume as well as a slight increase in its viscosity.  So, if you have a 3" diameter pipe at the exhaust manifold, why do you need the 3" pipe all the way to the tail pipe?  Seems like those coffee can size stainless exhausts the kids love today are more decorative than anything else.


RE: Exhaust debate

What you described could be due to not only the diam but also the length of the pipes so that the exhaust pulses and their timing and inertia may have helped scavenge more gas as it exited from the cylinders leaving more room for the incoming fuel and air.

RE: Exhaust debate


Of course.  That was the purpose in building a "tuned" exhaust system.  Believe me, we tried a multitude of different systems but ended with the one I described.  I don't recall the length, it HAS been a long time, but it exited the rear.  The problem is not knowing how to configure an exhaust system for a particular application, it's choosing the right formulae to find the correct dimensions.  There are great works available that give, in detail, how to find all this stuff.  THEY DO NOT ALL AGREE!
Trial and error works great IF you have lots of time and money.  Choose any book. Do the math.  Build the system. (So far so good, huh?)  Now, how do you know you have the ultimate unless you try a few more systems?  Shucks!  Now we are back to 'trial and error'.  The solution is  to engineer the very best system you are capable of and TEST, TEST, TEST!!!  What I have found is that the best engineered system is often NOT the best performing system.

Chumley, dammit, when are you going to give it up and join us?  We need more people in here that I agree with.  


RE: Exhaust debate

..wow i must be totally off.
I had assumed that if you have a 3" collector...then rather you have a 3" dia straight piece of pipe or no pipe at all after the collector that there would be no difference in hp/torque output.

ie...friction loss is not a factor here..and since there are no bends in the pipe then static pressure loss is out.

i just don't see where pipe after the restriction is a factor.

RE: Exhaust debate

Islander: "..wow I must be totally off."---

Could be!  Perhaps a trip to the bookstore and a few weeks of study could bring you up to speed?   


RE: Exhaust debate

even a pipe without bends has "restriction" just not nearly as much.
for instance a 3" pipe with a constant radius 90deg bend.
the 90 deg bend has approximately the same restriction as 9 feet of straight pipe.
I don't know weather it affects the wave dynamics timing in the same way, I would assume not.

RE: Exhaust debate

Oooooh the 'Black Art' of exhaust design.

As has been said above there are many different factors that have effects and interactions. Things such as the surface roughness of the pipe itself, speed of the test engine, valve timing etc etc

I think that Ricardo invented their WAVE suite of software was an attempt of trying to quantify things like this.


RE: Exhaust debate

We use WAVE. A lot. We also use a gas axe. A lot. Guess which we use when we're in trouble?

Chumley is right, the drop in temperature down the pipe has a big effect on the dynamics, including the wave speed. Getting your temperature profile right to the n'th degree doesn't seem to be that important, but then it could be part of the reason for my first paragraph!

Carnage - I don't think bends would provide much of an impedance mismatch to bounce waves off - I can't remember what the critical angle is for a diffuser, but I think if you run a 20 degree taper then the gas can't 'see' the join. I think a bend would be like that. To get reflection you have to create the equivalent of a hydraulic jump if you know what that is, ie an abrupt acceleration or deceleration.


Greg Locock

RE: Exhaust debate

I've just remembered the obvious - the diffuser on a cat is at the correct angle to provide good flow characteristics - what you want to do is to get the percentage rate of change of area per unit length to be the same. I have a suspicion that you might need to factor in the local sound velocity in there... but maybe not.


Greg Locock

RE: Exhaust debate

The answer depends on the engine you're running, and how much gasflow runs through the system.
If your engine is making 100 HP,  there'll be absolutely no difference between a 2" pipe, or a 3" pipe, or a 4" pipe downstream of the collector: the frictional losses at that flow velocity will be negligible.
My guess is your engine would have to making upwards of 1000 hp for you to see a measurable difference between 6' of 3" pipe and 6' of 3.5" pipe.

I do know that I developed a system for the ZR-1 Corvette that reduced the system backpressure while reducing system pipe diameter from 2-3/4" to 2-1/4".
Because most of the losses in this system are at the entrances to and exits from the mufflers and resonator.
And of course the smaller diameter system weighed significantly less, while having better subjective sound quality.

RE: Exhaust debate

ah, well then i am back to square one.
you decreased the size of the exhaust pipe by reducing the other restrictions and reduced backpressure at the motor.

SO Unless someone shows me otherwise...
If I have a 3" "restriction" at the manifold or at the Y of the exhaust pipe of a V8 engine making 270 hp...
then increasing the 3" pipe downstream of the restriction to 3.5" will aid no benefit other than by reducing the restrictions at the muffler and any bends.

or more simply..
If the pipe downstream of the 3" restriction was just a straight run out say 10' long,  no muffler...then going with 3.5" diameter pipe downstream of the 3" restriction gains me nothing..


Because "someone" is trying to tell me there is some sort of gas pulse loss or something like that.

I contend there is an irrelevant amount of friction loss of any type involved in my example.

RE: Exhaust debate

Islander, I was being a bit humoruos on my last post but---in all seriousness, a good book on exhaust theory (many available) will get you farther than what a 'friend' thinks.  The dia of the pipe is only one factor in designing a proper system.  The tuned length of the system is affected not only by the pipe dia but the temp. of the gas and the gas speed.  Too large dia gives too slow gas speed and even though the back pressure may be low and, since the sonic speed varries with temp and pressure, the wave tuning goes down the 'tube' (pun).  There is a 'perfect' system for every application and any variance either way of that  will result in less performance.  Finding that 'perfect' setup is the subject of MANY books and hours of, as Greg put it, 'gas ax' engineering.
Just because your headers have a 3 inch collector is NO guarantee that a 3 inch tailpipe is needed.  The length of the headers, their configuration, length of collector, duel/single pipe, 'X' pipe, 'H' pipe, type of silencer , number of resonators, cats, -------on and on. The only factory car I  ever owned with a 3 inch exhaust system OEM was my 64 Ford Fairlane Hy-Po 271hp 289cu.in. and it was only a SINGLE pipe system.  Seemed to work ok , though.

Bottom line is that you may find performance increase with SMALER dia system that a full 3 inch.  Maybe not, but a VERY good bet you will.


RE: Exhaust debate

Another thing that always "tickles" me is hearing from some engineer that "you gotta have some backpressure to make it work right!"


Did a dyno test on a 318 Dodge engine about 8 years ago and used the stock exhaust, complete with no alterations.  Got 185 bhp.  Took off the manifolds and installed headers 2.5" down pipes to 2.5 inlet 2.5 outlet cats, then to 2.5 inlet 2.25 outlet mufflers,(two of em) and gained 25 bhp.

We used a pressure transducer in the EGR port to measure backpressure.  Had 4 psi at 4000 with the stock system, 1.5 psi with the headers, and we gained 35 bhp too.  At no time did we ever obtain positive scavaging.

General logic tells me that if you have to increase backpressure to make it run right, you have a problem with camshaft profiles, ignition timing rate, fuel induction (my best guess) or engine mechanicals.  Otherwise, we would see Top Fuel 7000 bhp engines with butterflys in the exhaust headers!


RE: Exhaust debate

I read a superb article by David Vizard on exhaust design published in an Austrailian magazine but emailed to me by a correspondent. The basics are as follows:-

Back pressure is bad and if you can get the exhaust to exit in a low pressure area on the vehicle all the better.

For no or very little power loss the exhaust must flow 2.2 cfm per horsepower (0.2 psi back pressure, 5 to 6 inches of water)

To achieve open pipe power with a silenced system you require an expansion chamber at the end of the collector. This expansion chamber must have a volume of 8 to 15 times the size of one cylinder and have an abrupt change in cross section. This then feeds into the rest of the system and silencer/s and as long as it flows 2.2cfm per hp no power loss will result.

The expansion chamber basically replicates an open pipe.

The article also discusses manifold design but is not easily summarised here.

There is a picture of my expansion chamber here www.BeardmoreBros.co.uk\website pages\new_project_engine_page5.htm


RE: Exhaust debate

In referance to the Vizard exhaust ideas---We just recently tested the David Vizard (David Anton designed?) RC-40 exhaust system on the Evanspeed Clayton chassis dyno and the results were somewhat unexpected.  The muffler/expansion chamber/resonator system produced just 4% less hp at 7500rpm than an equal length 2 in. open  pipe. The problem came with the torque readings.  The system produced 15% less torque from 3500 to 6500 rpm.  One of our forum members, MaxRaceSoftware, analized the numbers and concluded that the  problem was either a wave tuning problem or restriction but since the  top end figures were nearly the same the restriction theory is probably not valid.  That leaves the wave tuning problem.  This would not normally come up on a street machine but on a car that is 'dual'   purpose (supposedly) race/street it is cause  to abandon the muffler system in favor of open exhaust, noise regulation permitting.  Seat of the pants dyno on track validates this idea.  I found a VERY big difference at 3500 rpm on the track (since the Mini has terrible gear ratios I found myself out of the optimum powerband a LOT) between the  two systems. Bottom line here is that while I have several of Vizard's publications at hand and they have  been the 'bible' for sports car tuners and deservedly so, I might add but, I do have other publications that tend to dissagree  with some of his 'expansion chamber' ideas.


RE: Exhaust debate

Islander, what no one has bothered to explain to you because it is well known to experts in exhaust systems is that flow though an exhaust pipe is not uniform and can't be viewed like most engineers are used to. When an exhaust valve opens it allows a pulse of high pressure gas pass out of the cylinder at high velocity. Gas at high velocity has momentum. If the exhaust pipe is tuned properly you can time things so that gas that is still moving away from the exhaust valve due to it momentum down the pipe creates a partial vacuum at the valve. If this happens before the valve closes again this can assist in removing exhaust from the cylinder. Of the low pressure pulse could be timed to arrive at the exhaust valve when it opens again at the next cycle to offer less resistance to flow.

Buzz jets (e.g. the German V1 during WWII) use this principle.

RE: Exhaust debate

Thanks for the reply(s).

Anyhow..I have installed headers and 2.5" true duals.(gsmotorsports.com).lessening backpressure, increasing torque/hp.(even at low rpm)..maybe the x-pipe synched the pulses feels like a nice gain..have'nt dyno'd yet.

RE: Exhaust debate

One other not-so-minor point: some engines have a much greater sensitivity to exhaust backpressure than others.
In general,  the higher the specific output(and therefore the greater the valve overlap) the greater will be the power loss per unit increase in backpressure.
More backpressure produces more charge dilution and/or reduced scavenging during the valve overlap period.

For this same reason, 4-valve engines are generally much less sensitive to exhaust backpressure.
Ford produced an excellent but not widely-disseminated internal-only paper on this topic during the MN-12 program.

Red Flag This Post

Please let us know here why this post is inappropriate. Reasons such as off-topic, duplicates, flames, illegal, vulgar, or students posting their homework.

Red Flag Submitted

Thank you for helping keep Eng-Tips Forums free from inappropriate posts.
The Eng-Tips staff will check this out and take appropriate action.

Reply To This Thread

Posting in the Eng-Tips forums is a member-only feature.

Click Here to join Eng-Tips and talk with other members! Already a Member? Login


Close Box

Join Eng-Tips® Today!

Join your peers on the Internet's largest technical engineering professional community.
It's easy to join and it's free.

Here's Why Members Love Eng-Tips Forums:

Register now while it's still free!

Already a member? Close this window and log in.

Join Us             Close