Liquid ingress in Axial flow compressor 1

[jlbish]

Hi all,

I am interested the reasoning behind liquid ingress being allowed in an axial compressor (on a gas turbine unit), but not in a centrifugal compressor. Obviously liquid carry over can be devastating in a reciprocating compressor, and I believe is quite damaging to a centrifugal compressor as well. I am just wanting to know the theory behind why liquid can enter an axial flow compressor without any negative effect(an example on my plant is compressor blade water washes on gas turbine units, or alternatively aircraft turbojet/turbofan engines which can ingest large amount liquid without damage). The principle of all 3 types of compressor are the same (gas is forced into a smaller space, decreasing its volume), so any liquids, being incompressible, should theoretically not be able to enter the compressor. What am I missing?

Is it just the high mass flow rate that makes liquid ingress negligible, or something else?

Thanks in advance for any responses.

 

[zdas04]

I think you are missing the way a rotary dynamic compressor works. It does not "force gas into a smaller space", dynamic compressor operation is explained best by the Bernoulli Equation (i.e. pressure changes are proportional to the difference in the square of velocity)
[ul]
[li]In a centrifugal compressor the gas enters the eye of an impeller and is rapidly accelerated (at near constant pressure) into a volute. In the volute the high velocity is converted to increasing pressure by putting the gas into an increasing space.[/li]
[li]An axial compressor looks very much like a driven version of a reaction turbine. Within each set of wheels the gas velocity is increased by the driven wheel and then enters a divergent section that trades the increased velocity for increasing pressure in an increasing space[/li]
[/ul]

Centrifugal compressors have several very severe direction changes that gas can make without excessive force, liquid making the same turns has markedly more specific mass and since force is mass times acceleration, the forces increase dramatically at the turns and can break stuff.

Axial compressors do not have the dramatic direction changes and can tolerate liquid ingestion without excessive local forces.


David Simpson, PE
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

 

[jlbish]

Thanks very much for your insight. It's a bit of a broad question, but is there a rough limit for the amount of liquid that an axial compressor can ingest (as a ratio of gas to liquid intake) before potential component damage is an issue? Or do the forces from the lateral momentum of the liquid not reach a high enough point for this to be a real issue? (Lets assume the liquid is water with s.g of 1)

Thanks again.

 

[rotw]

Agree with Zdas, especially on having non excessive local forces effect. The question I guess is how non excessive?
On centrifugal compressor, the liquid ingestion is tolerated however this may create erosion and/or corrosion. actually a certain fraction of liquid carry over through machine is normally specified by machine manufacturer as an acceptable limit. But centrifugal was not the question as I understand.
So for axial blades, the example that comes to my mind is condensing type steam turbines. I think that when the steam has some mist there is a risk of damaging the blades by erosion phenomenon. I think 80 percent humidity is a sort of acceptable limit I have in mind from some project, but I have to confirm it. Actually this sets the condensing pressure level. With air cooled condenser, the condensing pressure is a sizing criteria for the condenser, together with the humidity limit specified at the condensing stages (therefore mitigation of erosion problem) these things are intimately connected.

"If you want to acquire a knowledge or skill, read a book and practice the skill".

 

[jlbish]

Thank you also rotw..

I've never looked for any manufacturers specified tolerances for liquid carry over..I've just always been told 'no liquid in the compressor', which I think is a good rule to live by in general. I'll check out the data sheets for some centrifugals and find out.

I don't know too much about steam turbine units, but the principle is the same so I get what you're saying, although that answer seems like it is more in relation to mist/saturation. How about liquid slugs or a constant liquid water stream through an axial compressor? How much is too much?

An example that comes to mind is Qantas flight 32 a few years ago...the aircraft engine wouldnt shut down (Rolls Royce trent 900 series, running full speed) and fire fighters at the airport introduced a high flow of water into the engine trying to cause flame out. How would internal components handle this kind of large liquid volume?

It's getting very hypothetical now, I know, but I am interested in whether water/liquid is an issue for compressors in regard to stator/rotor damage etc.

Thanks again.

 

[zdas04]

rotw,
I think you mean "steam quality of 80%" instead of "Relative Humidity 80%" since by definition in a saturated system the RH is always 100%.

The how much liquid is too much question is something that the manufacturer of a given machine would have to answer. I occasionally see centrifugal air compressors (most dynamic air compressors are axial), and each manufacturer has very detailed and specific limits on liquids ingress--and they are all different.

David Simpson, PE
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

 

[jlbish]

I figured that was probably the case. I'll try and find some data on it..although if anyone has some anecdotal evidence I would be interested in hearing it.

 

[jlbish]

In case anyone is interested..I found a paper on the subject (relating to aircraft engines, but same principle). It seems like water ingestion will just cause steaming in the latter compressor stages and combustion chamber, eventually resulting in flame-out. Doesn't say anything about component damage, but since the water is vapourising in the compressor I would assume it is negligible.

 

[rotw]

rotw,
I think you mean "steam quality of 80%" instead of "Relative Humidity 80%" since by definition in a saturated system the RH is always 100%.

Yes, correct, this is what I meant.
Thanks for correcting me.

"If you want to acquire a knowledge or skill, read a book and practice the skill".

 

[jlbish]

Ah yes..that makes a bit more sense to me now (I know next to nothing about steam turbine units). haha

 

[gruntguru]

RH is not a relevant term in a system containing no air.

I wonder whether the higher PR in centrifugal machines accounts for the lower tolerance of liquids? The relative velocity would need to be significantly higher to achieve the required PR in a single stage whereas the multi-stage axial machine accelerates then decelerates the working fluid in each stage.

je suis charlie