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

Oil flow through a gallery

  1. May 31, 2013 #1
    Lubricating oil in a motorcycle engine gets pressurised by a pump and goes to an oil gallery, from where it splits into two paths: one to the cylinder head and the other to the crankshaft, via the cover clutch.

    Oil to the cylinder head travels through the entire length of the oil gallery in the crankcase, and then through a small hole at the top of the crankcase (around 2 mm) it fills up and goes to the annular space between the cylinder stud, rises upwards and lubricates the valvetrain and stuff.

    Now, there is a thing called "EFFECT OF FLOW TURBULENCE ON TEMPERATURE" right?

    My doubts are:

    1) How does the length of the oil gallery affect its temperature? My own reasoning goes like this: the longer the length of the oil gallery, the more the friction loss and the more the oil gets heated up. Also, the oil gallery is bent. What happens to the temperature of the oil in the bend? Does it increase, or does it decrease with the bend radius? The more the bend, the higher the temperature the oil gets heated up to, right? Or wrong?

    2) Even if temperature increases with the length of the oil gallery, does not all that heat get dissipated via the conductive surfaces (the walls of the oil gallery)? Ultimately does not the temperature decrease?

    3) How does a sharp bend (around 120 degrees) affect the fluid temperature? Does it increase or decrease?

    4) What is the effect of flow turbulence and bend turbulence on the oil temperature? Is it in any way related to the 3 questions I have asked above?

    I'm confused as hell.

    Please tell me the effect of bends and sharp edges in an oil gallery on the oil temperature.

    Thanks a lot in advance.
     
  2. jcsd
  3. Jun 1, 2013 #2

    SteamKing

    User Avatar
    Staff Emeritus
    Science Advisor
    Homework Helper

    IMO, you are worried about things which have no significant effect on oil temp. The most important factor is making sure that the metal of the engine block and cylinder head are not allowed to overheat, because some of this extra heat will be transferred to the oil by conduction. Remember, if the oil temp in the gallery is lower than the metal temp. surrounding the gallery, heat is going to flow into the oil. This is Thermodynamics 101.

    From a practical standpoint, if the oil path has a lot of twists and turns, it means that more power will be required to pump it through the engine.
     
  4. Jun 2, 2013 #3
    If oil in the gallery passes closer to heated components or thinner walled areas compared to other places in the gallery, than the oil will heat up in those places. This will have more affect on oil temperature than pathway length. In the corners the oil does gather and backlog a little, and creates hot spots and higher oil temps in those spots. There's balance struck between the conductive metal and oil temperature. The higher the RPMs the more heat, but also more coolant is being pushed via the water pump into the water jackets surrounding the cylinders and other places, thus the metal conducts more heat because it's being cooled faster. Around and round we go. However, if the upper limits of the cooling system are reached, and then the metal becomes too hot to conduct heat any more and things go bang!
     
  5. Jun 2, 2013 #4
    "What is the effect of flow turbulence and bend turbulence on the oil temperature?"

    A little tricky to know for sure. Turbulence within an oil gallery general creates friction (heat), but if the turbulence were to push some oil into cooler spots in the gallery, it's then possible the oil could actually cool a little. Perhaps similar to a water eddy (turbulence effect) pushing the water to shaded areas in a river.
     
  6. Jun 3, 2013 #5

    boneh3ad

    User Avatar
    Science Advisor
    Gold Member

    Turbulent heating is going to be negligible here compared to convective and conductive heating going on in a typical internal combustion engine. Turbulence itself can generate heat since turbulent energy cascades from the inertial scales of the flow down to smaller scales, eventually reaching the dissipative scales where it is dissipated into heat. The effect is fairly negligible in most flows. Turbulence is also very effective at mixing a flow up quite well owing to the various scales of eddies. As a result, a turbulent flow transfers heat through itself roughly an order of magnitude better than a laminar flow. this would be the dominant ramification of a turbulent flow in this case given the temperature gradients involved in an engine.

    Of course this is all for naught since I very, very highly doubt that oil flow could go turbulent in an engine. You have a combination of very high viscosity, very low density, very slow flow and fairly small pipes and tubing. That means you are going to have an extraordinarily low Reynolds number and the odds of seeing anything other than a laminar flow are quite slim.

    The end result is that turbulence likely doesn't exist in this situation. Bends in the tubing shouldn't increase the temperature simply because they are bends; if the temperature increases there it would be because the pipe is hot. If the oil gets hotter as it travels farther in the gallery, it is almost certainly due to simply sitting in a hot environment for longer.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: Oil flow through a gallery
Loading...