Exhaust - beating a dead horse

[soycory]

If this has been discussed, sorry. I have been reading about exhasts here today and I understand why a huge exhaust isn't good for fe. But, how about a straight standard size one. For example say a certain car has 1" exhaust pipe that bends, goes through a muffler and bends some more etc. It seems like a 1" straight pipe would keep the "harmonics" or scavaging or whatever, but, at the same time be a smoother path to exit thereby helping fe. So, reading the answers some of you guys posted on other threads, I know you are more learned than me on these things, so, please let me know if my idea has any merit please.

[Bicycle Bob]

IMHO, there is no benefit to backpressure except within overall systems that are still tuned to work with it, instead of with free flow. At the RPMs we run, there might be some extra scavenging due to resonance ahead of the cat converter, but I've never seen a system designed for that. I'd look up the standard tables for ductwork losses if I knew the gas volumes and could estimate the temperature changes.

[Ryland]

fewer bends is going to help, also keep in mind that your exhaust system has a resonator and steps the tubing size after that point, it's not a straight size of tubing.

[Frank Lee]

The muffler on my car has a smaller outlet than inlet. I don't know why.

[Piwoslaw]

Originally Posted by Frank Lee

The muffler on my car has a smaller outlet than inlet. I don't know why.

Maybe the gasses are cooler (taking up less volume) by the time they exit the muffler.

There was a link somewhere here to someone who redid the exhaust in a turbodiesel and claimed that power and torque improved at lower rpms. I can't find it at the moment, I have the link somewhere under my other OS. I'll post it later.

EDIT: Here's the link. But I still can't find the post where I found it.

[aerohead]

The I.D. of your throttle should closely mirror that of your exhaust.Your system should already have mandrel bends at all turns.A little bit of crimping won't significantly restrict the flow,and as mentioned by others,some back-pressure is ESSENTIAL to overall performance.Check me on this,but for a 4-stroke engine,if you'll divide your RPM by 2,then multiply that value times the engines displacement,this will give you your flow volume.If you divide the volume by the area of your exhaust pipe,this would give you the gas velocity inside.ASHRAE(American Society of Heating and Refrigeration Engineers) has some tables,as BicycleBob has mentioned that will give recommendations for max velocities inside pipes/ducts based on Reynolds Number.You would really have to continuously operate your car at wide-open -throttle,everywhere you went to benefit from an idealized exhaust system.Pressure taps and manometer readings along the entire system would prove it to you.

[Tygen1]

Avoiding the science, my expirience is going from a 1"7/8 crush bent pipe to a 2"1/4 mandrel bent pipe and didn't really see any noticeable improvement in FE. However, going from a CastIron log style exhaust header to a tuned, tubular, header did make an impact on FE, that is in addition to the larger exhaust pipe. Not a big impact on FE, mostly because of better torque down low has helped to decrease the amount of time in the Pulse phase of the Pulse and Coast routine. Consequently my throttle body is about 2"1/4, however the exit from the header is 2"1/2 back to the flex pipe.

[Jim Allen]

The outflow of an I.C. engine is exactly the same as the inflow with one vital difference... gas temperature. It might go in at 100 degrees but it comes out at 6-1200 and greatly expanded. The problem is that the "ideal" exhaust size various almost infinitely according to engine speed, load and exhaust temp. All you can do is test, determine the size that's beneficial the majority of the time, and lose a little on either side of it. For economy, smaller is better but one factor to keep in mind in dealing with OE systems is that one of the major criteria/factors for the OE in designing a system is sound tuning. They will sacrifice some performance and economy to meet that goal by installing an overly restrictive muffler. There are better and worse exhaust system examples out there and the price of the car usually dictates how the OEM did the job and the compromises made.

Contrary to what was said above, backpressure is not necessary or desirable in and of itself but, especially with a carbureted engine, if it and engine was tuned/calibrated with a restrictive system, it must be retuned for the less restrictive system. EFI systems will adapt (most times) to a free flow exhaust but even they should undergo some measure of recalibration to optimize a free flow exhaust.

You can calculate exhaust flow if you have the intake flow in CFM, and the exhaust temp using the formulae below.

First, you need to know how much air is going in. You can calculate a maximum, which is done for performance systems, or you can calculate at a specific rpm at which you want to tune. We'll use 3000.

Naturally Aspirated Intake Airflow in CFM

cubic inches x rpm
________________ x volumetric efficiency* = airflow in cfm
3456

* Most engine are 80-85 percent efficient on average. At lower rpms, they are generally at the higher end, the good one getting to 90 percent and sometimes more, at peak torque.

Example using a 2.3L (140ci) eng-

140 x 3000
_________ x .90 = 109.37 cfm
3456


To get exhaust flow, you need to know the exhaust temp at the speed and engine load you want to tune for, which is highly variable. I have a lot of data on diesels (max economy in a deisel usually occurs at 550 F EGT, for example) and could make an average "guesstimate" but not much on gassers. The only way to know is to install a pyrometer on your engine and find out. I'll use 600 degrees as an average for a gasser at a moderate load. The "460" and "540" are constants and I don't have time to go into what they are.

exhaust temp in Fahrenheit + 460
___________________________ x intake airflow cfm = exhaust flow
540

Example for our 2.3L at 600 degree EGT (at the previously determined 3000 rpm)

600 + 460
________ x 109.37 = 214.69 cfm exhaust flow
540


You can find the exhaust flow of many performance mufflers at various manufacturers sites and sometimes in performance magazines. OEM parts... you'll have to test yourself. I have done a bit of that and some OEM systems are surprisingly good... some abysmal. Most are OK for low rpm use. As to pipe size, the flow will vary according to the number of bends but see below for some average number for straight pipe. At a given diameter, number of bends and airflow, a mandrel bent system will have 27 percent less backpressure than a typical crimped system. This I have seen personally on a flow bench.

Straight Pipe Flow By Diameter

5-inch: 2200 cfm +
4-inch: 1800 cfm
3.5-inch: 1400 cfm
3-inch: 1200 cfm
2.5-inch: 900 cfm
2.25-inch: 600 cfm

I do not have data below 2.25 inch pipe.


As someone said above, scavenging is the ideal but that effect is elusive and dependent on gas velocity. With small pipes, relative to displacement/gasflow, the effect comes on at a lower rpm and the larger the pipe the later it comes. The trick is finding a balance between scavenging at a workable, efficient rpm and restriction when you have to rev higher. The OE strives for a balance, but we are free to focus as narrowly as we want and lie with whatever compromise you decide to make.

[theunchosen]

To answer the OP, you want the pipes as physically straight as is possible. Sun Tzu would say as is easily possible, but I doub thats necessary. If its too difficult you won't do it.

I have considered it for 2 reasons, 1 if you turn the engine around your AIT is not pre-heated by the engine and 2 the exhaust doesn't have to do a 180 turn to go backwards.

I'm looking at going lean burn and so the lowest possible AIT is desireable to keep it away from engine damage. On that note the only way I know to make my exhaust not do the 180 turn is to mount the engine backwards with an additional gear to change the direction of movement, but I doubt thats conducive to FE.