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

Compressor Performance

  1. Apr 19, 2006 #1
    Hello all.

    I have the following questions regarding the analysis of the performance of a rotary vane compressor.

    1) Why can the specific kinetic and potential energy changes normally be neglected?
    2) If an adiabatic compression is considered, what is the primary cause of energy loss by heat transfer from the system using a (water) cooling jacket?
    3) What possible causes could explain a lack of energy balance in the SFEE?

    Thanks.
     
  2. jcsd
  3. Apr 24, 2006 #2

    Mech_Engineer

    User Avatar
    Science Advisor
    Gold Member

    I'll try to answer these, it all has to do with trying to simplify the thermodynamic state of the air in the compressor to more easily decide where it is.

    1)
    a) The changes in kinetic energy can be ignored because the air is travelling relatively slowly. The air would have to be travelling at speeds similar to a trubine for the kinetic energy to make a significant difference. For a compressor, the energy is being stored as pressure work, not kinetic.
    b) For this, the fact is that a compressor is small, not several hundred feet tall. because of this, even if the air moves to a larger height, it's not enough to need to take into account. Even with water, it takes a column height of about 667 feet to store one BTU of energy, a lot of height for a little energy.

    2) This question doesn't really make sense, because an adiabatic compression means that there is a perfectly insulated control volume enclosing your compression process, which means a water cooling jacket cannot apply... A water cooling jacket would usually be used in conjunction with a very slow compression to try to approximate an ISOTHERMAL compression, where the air in the compression cylinder stays as close to constant temperature as possible. The problem with this is that the compressions take a very long time, and usually an ISENTROPIC compression is used to approximate a compressor. However, for an isentropic compresison with a water cooling jacket, the main losses would be in the form of heat conducted into the water.

    3) Lack of energy balance is easy to guess at. Losses in the entire assembly due to friction should be a first guess, since to have a good seal in the piston you muct have some frictional losses. Additionally, an isothermal compression is theoretically the most efficient compresison you can have. Any heat that is lost from the compressor is an energy loss. So, the farther you are from this form of compression, the less efficent you are. Usually, compressors follow closer to an isentropic compression, but even then you are not folowing it exactly due to real-world losses.

    EDIT: Since it is a rotary vane compressor, another possibly significant loss could be from the viscous losses in the fluid, depending on how fast you are compressing the fluid and its associated Reynold's number.
     
    Last edited: Apr 24, 2006
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?



Similar Discussions: Compressor Performance
  1. Compressor as engine (Replies: 2)

  2. Compressor power (Replies: 1)

  3. Multistage Compressor (Replies: 1)

Loading...