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

Turbocharger Leak Test question from a Physics Layman

  1. Sep 2, 2016 #1
    First off, I know very little about physics and have only taken classes in high school years ago. I'm a car fanatic and I'm mainly joining this forum to settle a debate I'm having with one of my friends about turbo cars and boost leak tests. For those who don't know, in an automotive turbocharged system the turbo spins and compresses fresh air into the cars intake tract creating boost allowing the engine to create more power.

    Part of the maintenance of these cars is something called a boost leak test. A boost leak test is where you seal the intake tract and apply air pressure equal to the amount of boost your car will produce under operating conditions and then look for leaks. Any leaks found equate to loss of power and can throw off the tune of the engine, so it's important to find and fix all leaks.

    My friend thinks that a leak found during a boost leak test will leak the exact same amount of air as it would under the same boost pressure with the car running at operating conditions. I think the turbulence and higher volume of air seen at operating conditions will cause more air to escape the tract than a static test, even if the air pressures are equal.

    I hope this makes sense and I've posted this in the correct place. Basically we're wondering if compressed dynamic air will leak out of a pipe faster than static compressed air, if both are at the same PSI.
  2. jcsd
  3. Sep 6, 2016 #2


    User Avatar
    Science Advisor
    Gold Member

    Can you be more specific as to what you mean by this? Turbulence and volume flow don't typically wouldn't have an effect on air leaking through an orifice unless the geometry causes a higher (or lower) pressure in the region of the leak.
  4. Sep 6, 2016 #3
    This. The geometry may cause a local high or low pressure area next to the leak. And this could change at different flow/pressure levels. I test with a lot of high pressure gas systems at work. I recall one incident where I was emptying a cylinder from about 5K psi and it emptied through the top of a tee. The leg had a ball valve on it. I opened the ball valve in hopes it might speed up the process by allowing an additional flow path, but the high gas velocity caused it to pull air through the leg. The behavior changed as flow decreased. Its similar to the effect that allows a gas mixer for a burner to work. The short answer to this question is "it depends".
  5. Sep 7, 2016 #4
    Sounds like the air being sucked through the leg was because of Bernoulli's principle? Just a guess I've been reading more about physics in the past week than I have for probably a decade, which isn't saying much.

    To be a little more specific let's take the drawing below. Air is flowing from right to left in a straight section of intercooler piping joined by a silicone coupler held on by band clamps. Let's say the band clamp on the left is a little loose and is causing a leak at the red circle. Would static air leak more, less, or the same as air moving at a high velocity directly at the leak? This might be another "it depends" scenario, and that's fine but hopefully this will narrow the scope of my question.

  6. Sep 7, 2016 #5
    I apologise, I certainly dont have the answer, but the idea is making me wonder if an area of leak would cause something similar to a Venturi vacuum?
  7. Sep 7, 2016 #6
    I'd say its possible it may leak more, but it could just as easily leak less if you looked at the other connection in your diagram. Its not a simple question you are asking. There are a number of variables that can affect this, including flow rate, piping wall thickness, pipe diameter, shape of the end of the pipe (rounded, tapered, straight cut), etc. And even then, mathematically solving this isn't trivial. I would suspect most engineers would think about the effects qualitatively. Use computer simulations to model the effects. Then actually test it to confirm. I doubt you are going to find such a specific answer to validate your position. And frankly, being more specific only validates that particular scenario, where the answer to your original question truly is, it depends.
  8. Sep 7, 2016 #7
    I don't really have a position I'm trying to validate at this point. I just thought of that example off the top of my head because it's a common cause of boost leaks and I was trying to be specific. But I understand your point about too many variables. A better example would've been a straight pipe with a small hole in it. But from Mech_Engineer's response, in that example it sounds like the leak will leak the same amount of air whether the air is static or moving with a lot of velocity over the point of the leak.
  9. Sep 7, 2016 #8


    User Avatar

    Actually the leak rate would be reduced as the velocity increased assuming the hole were perfectly perpendicular to the flow because the static pressure at the hole would be reduced just as it is in a pitot type air speed indicator for aircraft with one tube entry aimed directly at the flow to identify the total pressure and the other perpendicular to measure static pressure. At the same time there is the effect of flow impingement against the down stream edge of the hole that complicates the issue.

    With regard to your diagram, a leak at the downstream edge of the flex hose connection should be greater than one at the upstream connection assuming the leak is the same size at both points or the tubing has no clamps and is very flexible so that the increased static pressure at the downstream impingement point would stretch the hose more than at the upstream connection.with the flow away from the connection edge.

    Of course this is all theoretical and the difference relative to the total leakage at both points could be insignificant.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted