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

Choosing an efficient compressor

  1. Oct 31, 2015 #1
    Hello everyone. I'm working those days on my graduation project, and I need a little assistance. I have to choose a compressor that would be driven by diesel engine. The compressor should supply about 1.7 kg air and compress it to 2 atmospheres.

    It seems that centrifugal compressor or axial may fit, but I got a constraint that its efficiency will be about 90% or more which is pretty high, and these compressors typical efficiency may reach 80-85%. Compressors that do have such efficiencies are reciprocating compressors, but they are not able to supply the appropriate air flow.
    Would be glad to hear any suggestions. Tried to find relevant threads in the forum but didn't find one that may help, if you know any, let me know
    Thanks in advance
  2. jcsd
  3. Oct 31, 2015 #2


    User Avatar
    Science Advisor
    Gold Member

  4. Oct 31, 2015 #3
    That's too general...already tried this...
  5. Nov 2, 2015 #4
    Does the diesel have to be directly linked to the compressor? Can you use a diesel-electric setup? Usually compressor flow rates are measured in volumetric flow rates at several given pressures. Do you mean mass flow rate by the 1.7 kg? And why wouldn't reciprocating work? With enough RPMs and displacement, you should be able to get the flow rate you want at the compression you need. Do you mean you can't find any on the market like this?
  6. Nov 3, 2015 #5
    The diesel have to be linked directly. The flow rate is about 1300 liters per seconds. It's a big flow rate, compressors that I've found on the net, supplying such a volume per minut and not second. one of the most important factors for me is high efficiency. in order to supple this flow the compressor must run on a high RPMs, and this may drop the efficiency because of losses doesn't it? What may be the solution for this?
  7. Nov 3, 2015 #6


    User Avatar
    Science Advisor

    What is this for?

    A compressor capable of over 2000CFM will probably be over $100k, have an engine with hundreds of HP and come on a multi axle trailer.

    How long do you need to supply the required flow rate for? If it's short term, a smaller compressor with a reasonable sized receiver (and a high flow regulator) might work.
  8. Nov 4, 2015 #7
    Interesting idea regarding the small compressor, but anyways, it suppose to work continiously without any time limits. Regarding the price, it doesn't realy matter, I'm not going to buy it, it just have to efficient.
  9. Nov 4, 2015 #8


    User Avatar
    Science Advisor

    I don't think this will be an off the shelf item, or at least will be POA and presumably you'll need a written quote for your design report.
    so I think your best bet is to talk to sales engineers at compressor suppliers, IME if you are well prepared and know what you need (ie you need to use industry standard units for a start...) they are more than happy to help even when they know it won't result in a sale, you can sweeten the deal by offering an acknowledgement in your design report or poster. They will want to know what it's for (like we do here :wink: ), they may even have a better solution.
  10. Nov 5, 2015 #9
    Billy, thanks for advice. Today I've talked with engineering department in a compressor supplier company. Term of "isentropic efficiency" sound to him like a curse. He didn't realy know what is it efficiency, may be he used to work with catalogues and efficiency never mentioned there. He said that since I have such a volume to compress, in real life situation, he would suggest a screw compressor.
    In general, purpose of my project is to develop system that would increase efficiency of a diesel engine by use of waste energies. The idea is that engine will drive the compressor, compressed air will be heated by cooling the engine and exhaust gases. Output power will come from kind of turbine\expander and not the engine it self, hopefully to increase total efficiency. This reminds Braiton cycle. In order to achieve this I need that other components would be as efficient as possible.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook