Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Exhaust pipe sizing

  1. Sep 22, 2010 #1
    I am trying to get some help in sizing my exhaust pipe for my 1997 Pontiac Trans Am WS-6

    Can anybody point me in the right direction for equations?

    I pulled out my Marks handbook hoping it would help, but I couldnt find anything

    I have been thinking of doing this as a gas dynamics problem and using area ratios to see if the 3" dual or the 2.5" dual pipes give me the best flow for my equipment.

    Currently, I have long tube headers with a 1.75 inch primaries 3 inch collectors then going into a 2.5 inch y-pipe that has 3 inch cut-outs which then has regular catalytic converters then into a 3" Flowmaster catback. The valve is a 1.5in valve with an 8mm stem, and the lift at the valve is 0.5049in, and of course the Bore and Stroke of the engine is 4in by 3.48in respectively.

    I want to redo the y-pipe because it was cheaply made (in college)
    I eventually want to go to a true dual exhaust system.

    If you could please help me out I would appreciate it

    Here is a thread I posted on another forum
  2. jcsd
  3. Sep 22, 2010 #2

    jack action

    User Avatar
    Science Advisor
    Gold Member

    Short answer:

    If space is available, go with the 3 in.

    Long answer:

    The principle behind header design is the following:

    1- Size the primary pipe area as small as possible (without restricting the flow) such that a large pressure wave will travel through it

    2- Once this pressure wave hits the end of the primary pipe, it will return a negative pressure wave (vacuum) back through the primary pipe that, once it reaches the exhaust valve, will assist in cylinder scavenging. The length of the primary pipe will determine the proper timing for the desired rpm and the area ratio at the end of the primary pipe will determine the strength of the negative pressure wave (the larger the area ratio, the stronger the negative pressure wave). The proper way to calculate the area ratio Ar at the primary pipes / collector junction is:



    Ap is the area of one primary pipe;
    Ac is the area of the collector;
    n is the number of primary pipe going into the collector.

    To get a strong enough negative pressure wave, this area ratio should, at least, be equal to 6 (and if you put the number from your header, you'll see that it is equal to 5.94).

    3- if you have a collector, there will also be a (weaker) pressure wave traveling through it. Assuming open exhaust, when this pressure reaches the end of the collector pipe, hence the atmosphere (Ar = [tex]\infty[/tex]), it sends back a negative pressure wave that will reach the primary pipes / collector junction that (if timed properly) will assist the creation of the negative pressure wave that will travel back into the primary pipe (making it stronger). Usually the collector's length is either half or equal to the primary pipe's length for this to happen (that is where the H- or X-pipe may be placed on muffled exhaust, to create that larger area ratio, simulating some sort of "sub-atmosphere").

    That being said, what would be the effect of going from the 3" collector to a 2.5" pipe? At that junction, it would reflect a high pressure wave back into the collector which would reach the primary pipes / collector junction at some point. If this pressure wave arrives in between the arrival of any of the pressure wave coming from one of the primary pipes, it might do no harm. But if it does, it will decrease the strength of the negative pressure wave. So, there is nothing to gain and there's a chance you'll lose something.

    This is really simplified and complex exhaust geometry can do weird unexpected stuff, but if it was me, the short answer is: If space is available, go with the 3 in.
  4. Sep 22, 2010 #3
    This is what I was thinking, I didnt want to have a choke point directly after the collector. I am pretty sure I can tuck a 3" system under there

    I saw a Trans Am with a very nice dual exhaust system, which is the one I wanted to model mine after
    http://i47.photobucket.com/albums/f178/IamStonePhillipsthe1/1997 Trans Am WS-6/2hflyq8.jpg
    http://i47.photobucket.com/albums/f178/IamStonePhillipsthe1/1997 Trans Am WS-6/40evmrl.jpg
    http://i47.photobucket.com/albums/f178/IamStonePhillipsthe1/1997 Trans Am WS-6/2q8c13r.jpg
    http://i47.photobucket.com/albums/f178/IamStonePhillipsthe1/1997 Trans Am WS-6/347vfvr.jpg

    Anyways, I really appreciate your input, your concurrence really helped me think it through, I kept getting all sorts of different answers/comments/advice from some of my fellow engineers here at work
  5. Sep 22, 2010 #4
  6. Sep 23, 2010 #5

    Ranger Mike

    User Avatar
    Science Advisor
    Gold Member

    for street applications go with a tri-Y design..use a balance tube to connect both collectors as close to the end of the collector as possible. i assume you have a dual plane intake. so engine will not be as smooth as a single plane. with dual plane and stock exhaust the engine diles like it is two 4 cylinder engines connected with one crank shaft. single plane intake and the collector pipe syncs the engine and it idles a lot smoother.
  7. Sep 23, 2010 #6


    User Avatar
    Gold Member

    Although I've never referred to a fuel injected setup like that as single plane, unless the OP has changed it out, it is a single plane (one common plenum at least), large plenum, short runner intake manifold.

    Very interesting valve lift. Did you just change the rockers (I can't come up with that lift based on my known specs) or go with something small like a hotcam and keep the stamped steel 1.5s?

    Unless you're just wanting to try and play with the numbers yourself, there is plenty of data available for that vehicle (the f-body platform was extremely popular in those days). On those vehicles you're also pretty limited in ways you can run the exhaust, so you'll be able to figure out lengths pretty well. My suggestion is to dig for other people's research. On these cars with such a broad torque range, and especially at close to stock power levels, specific exhaust tuning (also difficult due to engine bay space and things like floorpans) provides pretty minimal gains.

    Is the car stock ride height, and if so are you going to keep it that way?
    Is the drivers side collector on your headers oval shaped or was there no real collector?
    I'm just wondering because it appears that you want to fabricate the system yourself.

    If you really want some answers, my suggestion is to also find a more specified forum. There are people like Larry Meaux (who designed the software PipeMax) that can give you more specific formulas/answers than you could ever ask for.
  8. Sep 23, 2010 #7
    Do you have the rest of the engine specs? Compression ratio, cam specs, etc? Was the only internal change the 1.65 rockers?
  9. Sep 23, 2010 #8
    yes internally the engine is stock minus the 1.65 roller rockers and beehives

    the compression is 10.5:1
    Valve Duration 279 intake, 276 exhaust (w/1.5 ratio)
    Valve Lift, degrees .447 intake, .459 exhaust (again w/ 1.5)
    Valve Overlap, degrees 41

    All the following specs are with 1.5 ratio rockers
    Valve Lift @ .050"(1.3mm)
    Intake .447"(11.35mm)
    Exhaust .459"(11.66mm)

    Lobe Lift @ .050"(1.3mm)
    Intake .298"(7.57mm)
    Exhaust .306"(7.77mm)

    Duration @ .050"(1.27mm)
    Intake 205 deg.
    Exhaust 207 deg.

    Lobe Separation 117 deg.
  10. Sep 23, 2010 #9
    As stated in the post above, yes it has the stock cam with 1.65 Roller rockers

    The pictures I posted are a project by Kooks, which make some pretty nice headers for the F-bodies.
    Yes its stock height, and no Im not going to lower it, it scrapes with the stock exhaust as it is.

    I have the FLP headers and they have a 3" v band collector so its not oval, and I just wanted to fabricate the exhaust from the collector back
  11. Sep 23, 2010 #10


    User Avatar
    Gold Member

    I'm pretty familiar with this platform. I've owned 3 camaros (2 LT1 cars) and helped build quite a few, mostly LT1 and LS1.
  12. Sep 23, 2010 #11
    Right on man.

    Well right now the main reason I am trying to rebuild my exhaust is because the one I had fabbed in college is pretty restrictive and it leaks, which brings up the CES light with a EGR flow code, Im pretty sure its because one of my cutouts is leaking OR i have a bad weld somewhere. AND with no O2 sensors downstream of the catalytic converters my gas mileage has really suffered, I use to get high 20s average and 30+ on highway trips, now its gone down to low 20s average and ~25 on highway trips

    For now I wanted to just redo the y-pipe going from the headers to the flowmaster catback that was on the car when I bought it, and then when I had extra money replacing the catback with a full true dual system
  13. Sep 23, 2010 #12
    At this power level there will be very little difference on a chassis dyno. Unless you are planning to make more hp in the future, the 2.5" system will do just fine. Your choice really.
  14. Sep 23, 2010 #13
    Initially I was planning on doing lots of mods to get bigger power, but not really so much anymore. I just wanted to get the better sound, get rid of the crossflow flowmaster, and fix the CES light

    On one hand the 2.5" system will be cheaper, and it would tuck under the car better, BUT like you said there is the bigger flow area and bigger potential down the road with the 3"

    Anyways we will see
  15. Sep 23, 2010 #14


    User Avatar
    Gold Member

    Has the EGR code been on ever since you installed the headers and/or rest of the exhaust? OBD-2 cars are pretty sensitive to throwing an EGR code due to exhaust changes.

    The sensors located downstream of the catalytic converters are purely for determining if each converter is working or not. They can throw a code, but won't affect how the vehicle runs.
  16. Sep 23, 2010 #15

    What is your exhaust port flange area or equivilant area (assuming the 1 3/4 id? headers are actually larger in diameter than the port), primary tube length minus the collector, and valve open, closing specs, any dyno runs?

    I am mainly just curious about your header design vs secondary pipe diameters. They should complement each other.

    If you want to learn more about the performance aspect of exhaust and intake design, search for papers written by GP Blair.
    Last edited: Sep 24, 2010
  17. Sep 24, 2010 #16

    If your loss of millage was immediately after installing your exhaust system, it is much more likely that you took a serious loss in horsepower with that system, than it being restrictive, and causing the problem.
    Last edited: Sep 24, 2010
  18. Sep 24, 2010 #17
    No the EGR code started going off about 6 months after I moved to New Mexico from the Texas Gulf Coast, which pretty much coincided with the exhaust leak I can hear and smell, I am thinking that the terrible New Mexico roads vibrated something to the point of either crack propagation or loosened my cutout flanges, I already checked the actual headers and they are still on tight and the dead soft aluminum gaskets dont seem to have blown out.

    As far as the downstream O2, they have a HUGE impact on the fuel ratio, WAY back when I was doing ASE Engine performance stuff my instructor use to call the downstream O2 the king sensor
    Basically it dictates whether or not the car is near 14.7, in my situation the O2 simulators are always reporting to the computer that all is fine, even though the upstreams might be black
    Now I maybe wrong about this, but before I went to college to get my BSME I was trying to get my ASE certification, and thats what they taught me
  19. Sep 24, 2010 #18
    it wasnt immediate
    and no I didnt have a serious loss in HP, before I couldnt spin the wheels on dry road because of the traction control, and after the switch it shows the low trac light but I can spin them stopped and even rolling

    Im fairly certain the culprit is the down stream O2 simulators
  20. Sep 24, 2010 #19


    User Avatar
    Gold Member

    Incorrect, certainly in this application. The upstream sensors are used for closed loop, the downstream ones (only on 96-97 LT1 cars and maybe 95 California cars since they had dual cats) are simply there to monitor the catalytic converter operation.

    It would be foolish to run 02 simulators in place of the sensors you're using to control that bank (and trust me, you're not). You do not always (or almost ever actually) have the car run at 14.7:1. It varies depending on certain inputs, a main one being throttle position. At low throttle you will want to run leaner than 14.7 and at high throttle you will be running quite a bit richer, almost certainly in the 12.5-13:1 range. I don't have any stock tables from my 94 otherwise I'd check and tell you how it came directly from GM.
  21. Sep 24, 2010 #20


    User Avatar
    Gold Member

    Actually, after some digging, it appears for low TPS outputs the STOCK tables do actually call for 14.7:1. I run mine a little leaner than that. Wide Open Throttle calls for 11.7:1, although it is in a different mode of operation at this point (enrichment vs closed loop feedback). I have mine at 12.5:1.

    It's been about 5 years since I played with stock LT1 tables. If you've heard of C.A.T.S. (TunerCAT) and datamaster, those are the pieces of software I used to tune and scan my OBD-1 LT1 cars.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook