Compressor Stall?

[avbug]

Back to the topic at hand...a compressor stall occurs when the airflow through the compressor doesn't occur as it should. It can take several forms, and may have several different causes.

Ariflow through a turbine engine starts with pressure elevated slightly above ambient ("ram air rise") in the engine inlet, and then continues to rise under successive stages of compresssion...as the air is both accelerated and compressed. From there, the air is dumped into a "diffuser" that serves to further increase the airpressure before it's admitted to the burner section of the engine.

Air pressure in the compressor should be higher than that in the burner section. If airflow in the engine slows and pressure in the burner can increases above that of the compressor section that's feeding it, the airflow "backs up." A change in this flow, and a reduction in speed, changes the angle at which the airflow meets the various vanes and blades in the compressor section. These blades and vanes function very much like any airfoil to which you may be accustomed. They stall, too.

During engine start when airflow is low and burner can pressure is rising, "acceleration bleeds" are used to control the air pressure being developed in the compressor section. Too little airflow isn't good as the engine can overtemp, and a hot start will develop. Too much airflow isn't good, as the flame may be blown too far aft in the burner can and can go out...a flame-out.

Accleration bleeds automatically regulate the air pressure and flow during lower speed engine operation. Traditionally, compressor stalls can occur when some type of blockage or restriction occurs to the airflow into the engine, during operation of the engine at high angles of attack with low speed and high power settings, and during rapid power transients when temperature and burner pressure may increase faster than the compressor can keep up.

Compressor stalls vary from a low grade hooting hoise to shotgun like bangs and barks with flame coming out the front of the engine. The soloution, or the start of the soloution, is the same; reduce power to the point where compressor stalling ceases, then attempt to correct the condition. If the condition was low speed with high angle of attack and a high power setting, you've just corrected it by reducing the power. If the problem was too rapid an advance on power, then retarding power has corrected the situation.

While a compressor stall may include actual aerodynamic stalling of compressor blades, generally it's not really a compressor stall so much as it is an engine stall...the engine is "stalling" by slowing or reversing the airflow momentarily. Surging in a compressor stall occurs as the situation rights itself and then creates itself again, and again. During these surges, the engine may flame itself out by moving the flame in the burner cans toward the rear of the can and ultimately eliminating the ability for it to propogate itself. It also creates a situation in which not enough cooling airflow through the burners is available, and the burner cans can be torched and crack.

Approximately 75% of the airflow through the comprressor gets used to cool the burner cans and keep the flame in the proper spot inside of them...to protect them from being torched. Reducing this airflow, such as occurs during a compressor stall or surge, can result in the flame touching the think burner can walls and damaging, eroding, or cracking them. Again, an important reason why a reduction in power is necessary during a compressor stall.

 

[Workin'Stiff]

I've always thought of it like a boat... Have you ever heard an outboard engine cavitate when you're sitting still and you firewall it??? You throw it to the boards, tons a fuel is sent to the engine, in turn spinning the prop extremely fast. However, there's in sufficient "grip", so to speak, to produce any thrust. The prop just spins and make a helluva sound...

 

[mayday1]

Bad news. Best solution: reduce power, lower pitch and increase airspeed in order to regain a nice smooth airflow back into the engine.

thanks, good info... is recovery that easy? can you develop unrecoverable compressor stalls? for instance, that RJ that crashed earlier this year..?

 

[avbug]

A mechanical compressor stall, wrought on by damaged compressor blades or stator vanes, inlet blockage, excessive compressor contamination or FODding,or other mechanical interruption, may be unrecoverable, or the engine may be operated in some cases at a reduced power setting.

If the engine flames out due to the airflow interruption or surging, then it may or may not be restartable, depending on a number of variables.

 

[mar]

Just a little more

Ariflow through a turbine engine starts with pressure elevated slightly above ambient ("ram air rise") in the engine inlet, and then continues to rise under successive stages of compresssion...as the air is both accelerated and compressed. From there, the air is dumped into a "diffuser" that serves to further increase the airpressure before it's admitted to the burner section of the engine.

These topics are always difficult to generalize because we operate these machines over such a vast range of settings and conditions.

But I wanna add just a little more.

I'd say the first part of Avbug's statement is true during "cruise" conditions (let's generalize and say more than 250 kts). At speeds greater than 250 kts (generally) you have to factor in "compressibility" in everything aerodynamic.

Therefore, yeah, the air pressure in the inlet is "above ambient." But that's not the only reason.

At speeds less than 250 kts (when you can consider the inlet air being "sucked" in rather than "crammed" in) the inlet will *still* induce a pressure rise by the act of diffusion (which is the exact opposite of Bernoulli's constricted flow idea).

I'm not arguing with Avbug's point that a "diffuser" exist behind the compressor section but I didn't want to gloss over the importance of the design of the inlet and spinner that the engineers use to mitigate compressor stalls.

Not only that but I'd also like to expound just a little more on Avbug's use of the word "accelerate" in the context of what's occuring in the compressor section.

Is the air "accelerated"? In the sense that it's direction has changed (centrigually), Yes! A physicist would consider that an acceleration.

But would a pilot consider the air flow (axially) to be accelerated? I'm not so sure. I would certainly admit that the compressor blades are small airfoils and require smooth airflow to operate efficiently and, thus, conform to Bernoulli's theorum. But I'd be careful to consider the airmass (in general) as "accelerating" through the compressor section.

In fact, quite the opposite is occuring. The airflow, in general, from the lip of the inlet until combustion in the hot section, is slowed and the pressure rises along with temperature.

One final point: In the original question it was asked if the blades in the compressor can change their "angle of attack" or are they fixed.

As far as I know, compressor blades are fixed (in the angle they're mounted) but they actually "float" in fittings. Sort of like loose teeth that wiggle and can be removed if you're clever about it, but the centrigual force from the rotation of the compressor rim actually holds them in place during flight.

What does change is the angle of the stator vanes (on some engines). The purpose of the stator is help direct the air flow from a previous compessor stage on to the next compressor stage. Stators do not rotate and do not exactly compress the air but they do slow it down and present it for further compression.

Like I said, on some engines, the angle is changed on the stator vanes pneumatically by engine bleed air. For the most part they change angles based on power setting.

To answer mayday's question about the last RJ crash and "unrecoverable compressor stalls"...that wasn't just a compressor stall. That was a seized engine where all rotation stopped.

In a plain vanilla compressor stall you won't find a seized "spool" but as Avbug says, there may be more damage than you counted on: http://www.tsb.gc.ca/en/reports/air/2002/A02P0021/A02P0021.asp

 

[avbug]

Actually, the airflow does speed up and "accelerate" in terms of both velocity and direction, in the compressor section. It is in the diffuser section that airflow slows and pressure rises, prepatory to introduction to the burner cans.

Ram air rise begins the moment the aircraft begins moving down the runway, and hits an equal point for most engines about 70 KIAS. Above that speed, ram air increases inlet pressure such that air is being "crammed down it's throat" instead of it "sucking." In reality, both are happening, but forward speed is creating an increase in inlet air pressure well above ambient from about 70 KIAS in most aircraft engine installations, on up.

"Ambient" for most engines isn't the outside pressure, as we would normally think of the term ambient. Turbine engines work by doing one of two (or both) things...producing torque to move things, and/or producing thrust. In either case, many engines utilize pressure probes as part of the fuel metering system or fuel control unit which make comparisons between inlet air pressure and air pressure in other parts of the engine. These are referred to as pressure stations. A few, but not many, aircraft also referece true ambient air pressure, though it's meaningless for most installations. A few use a separate external pitot tube to sample undisturbed (relative) airflow, but these also account for ram air rise, and the "ambient" that the engine sees isn't what we'd see from an independent gauge outside the aircraft somewhere. It's a relative term, and ambient to a jet engine can be considered the pressure inside the inlet prior to introduction to the first stages of the compressor. In some jet engines using EPR, this is the first element of the pressure ratio, or comparison between what the engine is taking in and what it's putting out. In short, it's the basis for the engine knowing how much power it's producing, and how to regulate it.

Some stator vanes are variable, but most are fixed.

 

[User546]

Here's a video to go along with the discussion of an A330 in flight with some serious "back-firing" going on:
http://www.bartnet.net/~jmt/a330_engine_stall.mpeg

 

[mar]

Nice video of a compressor stall.

Actually, the airflow does speed up and "accelerate" in terms of both velocity and direction, in the compressor section. It is in the diffuser section that airflow slows and pressure rises, prepatory to introduction to the burner cans.

Ram air rise begins the moment the aircraft begins moving down the runway, and hits an equal point for most engines about 70 KIAS. Above that speed, ram air increases inlet pressure such that air is being "crammed down it's throat" instead of it "sucking." In reality, both are happening, but forward speed is creating an increase in inlet air pressure well above ambient from about 70 KIAS in most aircraft engine installations, on up.

I'm far from an expert on any of these things but I'm having a hard time reconciling these statements against my, admittedly, "theoretical" knowledge and personal experience.

I'll tell you why I *think* there's more 'sucking' than 'cramming' during the T/O roll. Sometimes, on very humid days you can see moisture condense in the inlet. That indicates to me a strictly Bernoullian condition where the pressure is dropping along with temperature.

As you know, (I only mention it for those who are new to jet theory) a high bypass engine develops most of its thrust from the fan at low altitudes. It's operating like a huge ducted turbo-prop. In other words, it's taking a large mass of air and accelerating it backwards.

Therefore, I'm visualizing a big low pressure area in front of the compressor fan and a large airmass being sucked in until the engine reaches a cruise regime and the laws of high speed aerodynamics take over (Compressibility).

I'm certainly not trying to be argumentative here (after all we're not talking about the President). I'm here to learn as much as to teach so if there's some gap in my knowledge I'd certainly appreciate it if someone points it out...

...that said...

...I'm still fairly convinced the air flow *in general* MUST slow down in the compressor section, otherwise how in the world would it ever get compressed?!?!?!?!

Teach me something.

 

[aucfi]

...I'm still fairly convinced the air flow *in general* MUST slow down in the compressor section, otherwise how in the world would it ever get compressed?!?!?!?!

The airflow (velocity) in the compressor section is regulated by the use of stator vanes to maintain a relatively constant velocity after entering the compressor section. As the rotors rotate, they increase both the pressure and velocity of the air.

The stators do three things.
1. Slow the airflow off of the rotors.
2. Increase the pressure just a little bit more.
3. Smooth the airflow so it has very little "rotation" (spiraling slip stream).

The compressor section itself is designed like a venturi to utilize bernoulis principal but since the air cannot speed up due to the stators, the air is further compressed (restricted).




au

 

[jafo20]

Hey guys, if you want a good rundown on compressor stalls, read "Aerodynamics for Naval Aviators" and "Fly the Wing".

In a macroscopic sense, you want uniform airflow (uniform velocity in speed and direction) reaching the first stage compressor. Whatever produces non-uniform flow, high TAS+altitude+AOA, damage or contamination of inlets, screwed up stators, etc. may lead to a compressor stall. Non uniform flow, means one part of your compressor may be stalled (cavitating), and is no longer able to sustain a pressure barrier. Also, if pressure inside the engine exceeds the ability of the compressor to...uh...compress, you can get stall surge. For example, a surge valve failure can lead to stall-surge although airflow reaching the compressor is more or less uniform in property.

 

[DC8 Flyer]

........

 

[Regul8r]

I know compressor stalls can chew up an engine internally, anyone know if it's an automatic removal from wing if an engine has a compressor stall?

 

[avbug]

No, it's not an automatic removal from the wing if there's a compressor stall. A company may institute such a policy, but I've never seen it, and I've never seen any engine manufacturer that dictates engine removal or overhaul due to a compresor stall.

Things associated with the compressor stall may dictate removal or teardown. An engine can easily overtemp when airflow changes or reverses, and severe surging, backfiring, and stalling can damage vanes, stators, discs, blades, and crack or burn through burner cans, as well as even damage the engine pylon or mounting webbing and structure.

An engine experiencing a low grade compressor stall may be run at lower power settings in some cases, rather than being shut down, if the power is needed...reducing the power to the point that the hooting, surging, or other indications goes away, may permit continued operation so long as other signs of trouble aren't present.

Each airframe and engine manufacturer will prescribe specific special inspections that may be warranted under certain conditions...eg, hard landing inspections, lightening inspections, and hot start inspections. Hard surging or stalling may also warrant a particular inspeciton, or just a general maintenance investigation.