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

Ram air question

  1. Jun 25, 2010 #1
    I have read up on the ram air/ram jet of NACA times, and how its really only useful at supersonic speeds.

    my question basically is for clarification because I honestly don't know how to work the math out to really figure out how beneficial it would even be.

    This pertains to drag racing. I have a motorcycle and currently trap around 135mph at the end of the track, so i would make more use than most cars.

    Basically I was trying to figure out, if i had a cone inlet, say 7" in diameter that tapers down to 4" in diameter over 1' length linearly, would I see a notable increase in air velocity? at 100mph, if my manifold vacuum is say 3.5" H2O, how much less vacuum would I have, or would I possibly even see a slight presurization with that same cone?

    this question is benefical for me because as we go faster we experience more drag, our acceleration rate slows, and we plateau. If i can raise the plateau, even slightly, it would be great. something to counteract the drag, i.e. more horsepower.
  2. jcsd
  3. Jun 25, 2010 #2

    Ranger Mike

    User Avatar
    Science Advisor
    Gold Member

    Waddel Wilson, one of Nascars all time great crew chiefs was asked about air one time..he said " anytime the car is moving, yer moving air" so the supersonic crap is just that..small hp gain is realized with a slight ram effect

    Ram ( or scavenging) effect is most efficiently accomplished at a given engine RPM. this is based on the fact that the velocity of the pressure waves traveling in intake and exhaust passages is variable with engine speed.
    see Coanda effect of dropping pipe pressure..

    current thinking is a bell mouth inlet flair will add 25% to the ram effect of square cut tubing. in round numbers this means the ram boost of 45 psi will drop to 3 psi if no bell mouth opening is used. overall flare diameter is found by multiplying the uncut pipe diameter by 1.5
    The radius of the bell mouth is found by dividing the uncut pipe diameter by 4.
    basic ram tube formula is
    L=N(1100) / R where L is length of ram tube measured in inches from inlet tip of the tube to intake valve seat through the carb.
    R is desired RPM
    N is number of degrees intake valve closes past BDC intake ( for values up to 75 degrees use the specific value, for values greater than 75 degrees, use 75 degrees.
    watch the spark plug..you will lean the intake mix and may burn a piston so read the plugs..also you need to have a tuned header to remove the mix.
    Last edited: Jun 25, 2010
  4. Jun 25, 2010 #3
    Header pipe tuning isnt as readily available here. As it sits, there is minimal back pressure, just enough to create some exhaust scavenging. And the plugs should be fine, the bike is speed density, so between the o2 sensor, intake temp sensor, and the MAP sensor, I believe the ECU should be able to extrapolate an appropriate fuel mixture.

    so, if i did the numbers right in your runner length calculation, im looking at needing a runner length of 3.8 inches... valve timing is set to 38* ABDC, and im looking to make peak power at roughly 11000 rpm. that just doesn't seem right for some reason though. Its been my understanding and experience that longer runner lengths tend to reduce turbulence and increase velocity for each given cylinder, causing the power curve to slightly shift higher into the rpms. for example, some of these big power all motor big block cars are running tunnel ram intake manifold and are utilizing runner lengths well into the 25" range, whereas the formula provided would make a typical runner length for those type engines only 12-15"

    And im going to say 45 was a typo...i think you meant 4..makes sense with the math you provided, but this is definately not my strong point. but I would think the effect would be proportionate to the ratio of inlet vs outlet and the degree at which it has to transition.
  5. Jun 25, 2010 #4
    OP supersonic is not for internal combustion engines. The fastest the air can get past your valves is Mach 1, so sending the air in faster to an engine is pointless.

    Ramjets are designed for supersonic combustions so you can have more thrust from the exhaust at higher speed. Since you aren't using exhaust thrust, ignore it.

    You need to be careful, as RAM effect can have a noticable gain in power. However you can't just stick on a big inlet somewhere and hope for the best. Adding something like that will hugely increase drag, it needs to be sculpted into the bike.

    The inlet also neesd to be carefully designed so that the flow is distributed equally to all cylinders. Adding a ram and the n some ducting to the inlet runners may casuse the ones closest to the front to be starved of air (and run rich) and the ones at the rear to run lean as they get too much air.
  6. Jun 25, 2010 #5

    Ranger Mike

    User Avatar
    Science Advisor
    Gold Member

    like everything in racing, it is about balance. engineers have spent millions on developing a good power plant that will cover a wide rpm band...once you get away from this compromise you narrow up the power band for strictly 1/4 mile high end blasts and you loose any street drivability. As usual Chris x is right on..but its your ride so good luck and get a checker!
  7. Jun 25, 2010 #6
    on the bike the engine sits perpendicular so the cylinders are equidistant from the front.the headlight assembly will be where the cone(s) will be put. have to work around the front suspension so I may need to do 2, each one feeding 2 cyl.

    I have also contemplated layering the front end, similar to "<<" so it retains its shape and factory aero....or possibly using something on top of the radiator, as it catches air from above the tire but below the front end, the area that is spaced for the suspension travel, there will be excess air deflection above and below because of the pressure gradient as it hits the radiator but only a little bit passes through..

    think my most conclusive thing will be putting a vac/boost gauge into the air box, and seeing where i am at for a baseline, then put something together and see if the pressures change and if it coorelates with anything on the track time/mph wise. just some trial and error...but id like to take some of the error out before i get started ya know?
  8. Jun 25, 2010 #7
    I am willing to sacrifice the streetability, but that will be down the road when i can actually afford to build a bottom end to support boosting it.
  9. Jun 25, 2010 #8

    jack action

    User Avatar
    Science Advisor
    Gold Member

    The ram air boost (not to be confused with ram charging due to pressure wave) is what we call the dynamic pressure due to air velocity. Assuming no compressibility effects for simplification (which is true for air speeds << speed of sound, i.e. about 770 mph), it can be found by (in SI units):

    [tex]p = \frac{1}{2}\rho v^{2}[/tex]

    Where [tex]\rho[/tex] is the density of the fluid and [tex]v[/tex] is the air velocity.

    If we assumed the air density is 1.23 kg/m³, velocity is in mph and pressure is psi then:

    [tex]p = \frac{v^{2}}{56120}[/tex]

    If we assume that the air enters a 4" inlet @ 135 mph, the pressure boost will be 0.3 psi. At half-speed (67.5 mph) you get a 0.08 psi. Not much.

    Now, if we install a cone shape at that 4" inlet with a 7" base, the air speed will be 135 mph at that new inlet. Always assuming no compressibility effects, to determine the speed at the 4" inlet, it is a matter of area ratio, so:

    [tex]v_{@d} = \left( \frac{D}{d}\right) ^{2}v_{@D}[/tex]

    So the new air speed will be 413 mph (getting closer to speed of sound!). The new dynamic pressure will be 3 psi. At half-speed, you'll get 0.76 psi.

    This is like having a turbo with a 3 psi boost. You might need some engine tuning, but it can definitively make a power difference.

    One big problem: nothing is free. All that compression didn't happen out of thin air :biggrin:. You might have increased your drag. The big question is: Is the power increase high enough to overcome the drag induced? The answer will depend not only on the engine output but also on the new vehicle' shape and frontal area. So I guess this means a trial and error process on your part as determining the shape of the inlet and its location.

    By the way, if space is limited, when reducing the air inlet, a cone with an included angle of 60° is enough. So the length (L) of your cone don't need to be much longer than:

  10. Jun 25, 2010 #9
    wow, 3psi... i know that it isnt exactly the same as a turbo, less heat, and turbo's have varying efficiency at different pressures/rpm but typically, at 3psi with a appropriately sized turbo, its not uncommon to see 25-30whp. I know aero comes into play, but with the inlet being at the front of the bike with no protrusion, i would think that once a pressure had built at the front, drag wouldnt be an issue as what was built up at the front would cause the excess air to streamline around the rest of the bike as if the headlights were still there. If both my assumptions are correct, that is a dramatic improvement, but I have to be skeptical because if some little know nothing like me could pick up that much power, someone else in the years of racing should have come up with something like this aswell.
  11. Jun 25, 2010 #10

    jack action

    User Avatar
    Science Advisor
    Gold Member


    Almost all sport bikes have ram air nowadays.

    Read these 2 posts that contain an article from "Sport Rider":


    Here there might be a good explanation of the limitation on the stock bikes:

    http://www.hayabusa.org/forum/2102762-post4.html" [Broken]

    I think the problem is more one of engine tuning. IMHO, definitively worth a try, especially considering the low cost of the modification.
    Last edited by a moderator: May 4, 2017
  12. Jun 25, 2010 #11
    You certainly won't see a 20-30hp increase. It's best to just think in % gains. 3-8% is probably sensible depending on tuning. You'll probably lose a set % of that depending on the extra drag.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook