Compressors vs pumps

[mech212]

I know that centrifugal pumps produces head which has a value equals to the system resistance ,, but what is the case of compressors ? do they have curves like pumps ? Does their head or pressure changes with the system resistance ? How the compressibility factor of gases is involved ?

 

[zdas04]

A centrifugal pump or compressor is a polytropic machine that uses a rotating element to increase the velocity of the fluid and then pass the fluid into a volute with an increasing cross-sectional area. As long as the the fluid is in the range described by the Bernoulli equation the pressure will increase as the velocity decreases in the volute.

The most significant performance difference between a pump (liquid) and a compressor (gas) is that at some increasing velocity the gas will stop acting as an incompressible fluid and start exhibiting choked flow which does not follow the relationships described in the Bernoulli equation. At that point the volute causes an increase in velocity at nearly constant pressure.

A compressor curve will show a much smaller envelope with a "choke" line to the right and a "surge" line to the left. The surge line is the point where the pressure at the end of the volute is less than the system pressure and the system flows back into the compressor.

Another important difference is that liquids can cavitate and damage the pump internal parts, while gases cannot cavitate, but that difference is not clearly shown on a pump curve.

System resistance controls the discharge pressure of any centrifugal machine, a pump just has a less dramatic response to changes in that resistance.

[bold]David Simpson, PE[/bold]
MuleShoe Engineering

In questions of science, the authority of a thousand is not worth the humble reasoning of a single individual. Galileo Galilei, Italian Physicist

 

[LittleInch]

zdas04 has it correct.

"do they have curves like pumps ?" Yes see below for a typical. compressors also often change speed and pressure is often shown as a ratio of pressure in to pressure out, somewhere between 1.4 to 2.5 normally.

Does their head or pressure changes with the system resistance ? Of course it does.

How the compressibility factor of gases is involved ? This is involved with the system design or calculations, e.g. If you're pumping gas into a 100km long pipeline and the end valve closes, you can continue to pump gas into the pipeline for a long time as the pressure increases and the line "packs up". The flow through the compressor will track along the curve relevant to your speed until you hit the surge line You couldn't do the same thing with a liquid system to anywhere near the same extent.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[rotw]

To add a little to the valuable replies already given...

The first feature that is very characteristic of centrifugal compressors, and which is - to my opinion, a game changer relates to the so-called flow coefficient. Because generally speaking, gases are compressible, the actual volume of gas reduces through the successive compression steps. So the flow passage (impeller exit section) becomes narrower from one impeller to the next as the flow proceeds though the machine. This is something that can be observed physically; for instance a H2 compressor which operates upon a gas that is typically "hard" to compress will have successive impellers of exist sections that are almost equal while the differences in impeller's width is immediately noticeable on, say a Propylene compressor or even a natural gas compressor.
Ultimately, when suction inlet flow shifts when compared to the design flow, not only it offsets the 1st stage operation with regard to the best efficiency point and available margins to surge/choke, but that very effect stacks up across the successive stages and becomes quite dramatic on the last stages. All in all, the entire machine performance (efficiency, turndown) can be heavily effected by any shift occurring at the inlet. The phenomenon can be more or less accentuated depending upon the shape of the curves under consideration and the available turn down. As example, a CO2 machine would typically exhibit quite steep curves with narrow turn down / operating range. A similar outcome can happen due to fouling (take as example a coker gas compressor); because fouling narrows the flow passage section, even though the operating conditions remain at design, the machine / operating envelop is not anymore fitted to the conditions it has been designed for; performance degradation is one notable consequence. Some of the common implications encountered in contract negotiations have to do with guaranteeing performance outside of best efficiency area because of the aerodynamic peculiarities above described.

The second feature that is also remarkably specific to compressors and which you would not encounter in pumps, is that the compressor performance curves are set for a specified set of inlet conditions (namely pressure, temperature and molecular weight). Thus when a machine operates at different inlet conditions than design, a new set of interpolated curves is required based on the inlet conditions under consideration. Some techniques for estimate do exist, but in general you are left to the diligence of the OEM to provide you with the needed input, when not available in the original set of curves.