faro From Egypt, joined Aug 2007, 1327 posts, RR: 0 Posted (1 year 8 months 2 weeks 4 days 8 hours ago) and read 10949 times:
A number of questions that have been on my mind recently:
- Why are composites not used more extensively in the cooler bits of jet engines like the low pressure compressor (blades, stators), spool shafts and HP/LP casings?
- Why are the HP compressor blades not cooled, I understand that the air exiting the HP compressor stage can be at over 500°C? Wouldn't active cooling like with turbine blades increase the pressure ratio?
- Would a variable-cycle engine like YF120 hold any benefits for civil turbofans?
- Why are variable stators not used in the turbine?
- Are blisks feasible in the turbine stages?
- If the air coming out of the HP compressor is slowed by the diffuser (leading to lower pressure I think?), why is it not blown out the front of the engine by the force of the combustion in the chambers?
- What % of the oxygen in the air coming out of the HP compressor is not used in the combustion process? Is 100% oxygen burn desirable/achievable?
- Why not use coatings like Diamond-Like Carbon on compressor blades to enhance long-term surface smoothness and efficiency (reduction of surface abrasion from particulate matter)?
- What is the lower limit of the number of fan blades (the GENx present has only 18) that can be used on an engine? What drives this limit?
- What are the pros/cons of reverse-flow combustors as implemented on the earliest RR jet engines?
- Why not have a hollow spinner admitting air that would flow through the insides of the spool shafts (and be ejected out the back end of the nozzle) to enhance cooling of compressor/turbine blades from the inside?
- Is ambient air used to cool the engine casing? If not why?
A rather long list I'm afraid; thanx in advance for your input.
dl757md From United States of America, joined May 2004, 1554 posts, RR: 20 Reply 1, posted (1 year 8 months 2 weeks 4 days 6 hours ago) and read 10923 times:
Quoting faro (Thread starter): Why are composites not used more extensively in the cooler bits of jet engines like the low pressure compressor (blades, stators), spool shafts and HP/LP casings?
I'm sure they're looking at utilizing composites in at least some of those areas.
Quoting faro (Thread starter): Why are the HP compressor blades not cooled, I understand that the air exiting the HP compressor stage can be at over 500°C? Wouldn't active cooling like with turbine blades increase the pressure ratio?
Even at 500degC (probably a little higher than most engines) temp is not much of an issue.
Quoting faro (Thread starter): Why are variable stators not used in the turbine?
The mechanical complexity combined with the high temps would be a maintenance nightmare. VSVs in the LPC cause enough problems with bushings wearing out the I wouldn't think that a similar setup in the turbine would work.
Quoting faro (Thread starter): If the air coming out of the HP compressor is slowed by the diffuser (leading to lower pressure I think?), why is it not blown out the front of the engine by the force of the combustion in the chambers?
The diffuser slows down the airflow and therfor raises the pressure. In fact the highest pressure in the enigine is the at the outlet of the diffuser.
Quoting faro (Thread starter): What % of the oxygen in the air coming out of the HP compressor
is not used in the combustion process? Is 100% oxygen burn desirable/achievable?
Roughly only about 25% of the air entering the combustion liner is used to support combustion. The rest is used for cooling.
Quoting faro (Thread starter): Why not use coatings like Diamond-Like Carbon on compressor blades to enhance long-term surface smoothness and efficiency (reduction of surface abrasion from particulate matter)?
Erosion on the compressor blades is not much of an issue. Dirt buildup and FOD are the main issues with compressor blade shape and resulting efficiency.
Quoting faro (Thread starter): What are the pros/cons of reverse-flow combustors as implemented on the earliest RR jet engines?
Pro - Shorter more compact engine.
Con - reversing the airflow twice reduces efficiency.
Quoting faro (Thread starter): Why not have a hollow spinner admitting air that would flow through the insides of the spool shafts (and be ejected out the back end of the nozzle) to enhance cooling of compressor/turbine blades from the inside?
There are already provisions for cooling air throughout the engine. Additionally the spinner is designed to deflect FOD outward and around the compressor inlet rather than let it get ingested into the compressor. A hollow spinner would likely not be as good at doing this.
Quoting faro (Thread starter): Is ambient air used to cool the engine casing? If not why?
No. The compressor casing is not cooled and the HPT is cooled by bleed air from the HPC and the LPT is cooled by bleed air from the LPC. This is a sytem called ACC or active clearance control which helps to maintain the case to turbine blade tip clearance at optimum values.
tdscanuck From Canada, joined Jan 2006, 11032 posts, RR: 72 Reply 2, posted (1 year 8 months 2 weeks 4 days 2 hours ago) and read 10858 times:
Ditto everything DL757Md said. To expand a couple of points:
Quoting faro (Thread starter): - Why are the HP compressor blades not cooled, I understand that the air exiting the HP compressor stage can be at over 500°C? Wouldn't active cooling like with turbine blades increase the pressure ratio?
Active cooling the blades generally lowers the pressure ratio, because you're taking high pressure air from somewhere and dropping the pressure down as you cool the blades. Intercooling (actually cooling the airstream) can raise the possible pressure ratio, but it fairly impractical in aircraft engines.
Quoting faro (Thread starter): - Would a variable-cycle engine like YF120 hold any benefits for civil turbofans?
It could increase performance a little bit in some corners of the envelope, but it likely wouldn't pay for the increased complexity. Variable cycle is best used when you need to increase performance at some point far from the engine's design point...in the fighter case, you want acceptable performance all the way from takeoff (high bypass ideal) through supersonic (low bypass ideal). Commercial jets live in a much tighter speed range so the benefits of variable cycle are much less pronounced.
Quoting faro (Thread starter): - If the air coming out of the HP compressor is slowed by the diffuser (leading to lower pressure I think?), why is it not blown out the front of the engine by the force of the combustion in the chambers?
The Brayton cycle (the thermodynamic cycle under jet engines) is a constant pressure combustion. In an ideal cycle, the pressure in constant all the way through the combustor. In real engines, the pressure actually drops going through the combustor. Total pressure stays constant through the diffuser but, since the flow is slowing down, static pressure is increasing. For a lot of reasons, you want to do your heat addition (combustion) at low Mach number. Combuster design is all about minimizing pressure loss (both static and total) while adding as much heat as you can.
Quoting faro (Thread starter): - What % of the oxygen in the air coming out of the HP compressor is not used in the combustion process? Is 100% oxygen burn desirable/achievable?
100% burn isn't possible with today's materials...the turbine inlet temperature would be far too high. From a thermodynamic point of view it would be great, but we're not there with the technology yet.
Quoting faro (Thread starter): - What is the lower limit of the number of fan blades (the GENx present has only 18) that can be used on an engine? What drives this limit?
Two is the absolute lower limit (balance reasons). There are lot of competing factors on number of blades...aerodynamics drives you to low numbers, structures drives you to high, production cost wants to land somewhere in the middle, and noise wants to control number of blades vs. number of stators.
Quoting faro (Thread starter): - Why not have a hollow spinner admitting air that would flow through the insides of the spool shafts (and be ejected out the back end of the nozzle) to enhance cooling of compressor/turbine blades from the inside?
Cooling the shaft isn't typically a problem, since it's relatively far from the gas path and has a robust oil system providing a lot of cooling.
Quoting faro (Thread starter): - Is ambient air used to cool the engine casing? If not why?
In some designs, fan air (basically ambient) is used to ventilate the space between the core and the cowl. This provides some bulk cooling, but isn't doing much for components in the gas path.
lightsaber From United States of America, joined Jan 2005, 8537 posts, RR: 100 Reply 3, posted (1 year 8 months 2 weeks 1 day 20 hours ago) and read 10577 times:
Cracks knuckles, but the current asmswers are pretty darn good.
Quoting faro (Thread starter): - Why are composites not used more extensively in the cooler bits of jet engines like the low pressure compressor (blades, stators), spool shafts and HP/LP casings?
As already noted. Not required and composites are 'new technology.' For most of the compressor, aluminum blades could be used, but titanium is still used as that is the 'know material.' If you do not have to switch, there is a high cost to switching. The ability to utilize efficiently a higher pressure ratio is keeping down the temperatures from the compressor.
Integrated blade rotors (BLISKS) will cut out enough weight to make composites have very little (if any) benefit in a compressor. Also note, the shapes that can be made from composites are very limited due to the nature of the stress concentrations. So for now, metals are *far* cheaper in the compressor.
Quoting faro (Thread starter): - Why are the HP compressor blades not cooled,
As already noted, there is plenty of temperature margin with metals.
Quoting faro (Thread starter): Would a variable-cycle engine like YF120 hold any benefits for civil turbofans?
I liked tdscanuks answer.
Quoting faro (Thread starter): Why are variable stators not used in the turbine?
Risk. Moving parts in a hot environment are tricky. If anything failed, it requires a full engine rebuild. The high turbine determines the engine overhaul interval. No other component should have a shorter life. Adding variable stators in a *favorable pressure gradient* doesn't improve efficiency much (it would some), but it would make maintenance a headache and force engines to be pulled to the shop far too often.
Recall the compressor is in an adverse pressure gradient. In other words, it wants to flow backwards (a compresosr surge). So variable stators provide far more benefit to the engine.
Quoting faro (Thread starter): Are blisks feasible in the turbine stages?
In the low turbine... maybe. For the actively cooled turbine blades, the friction welding process would seal the cooling holes dooming the turbine to a short life with an ugly failure mode. Why do I say maybe for the low turbine? Right now the blade materials and the hub materials are different due to the different properties required. There are alloys that might work... but there is a lot of R&D to be done before it could work.
On this point I think you have a great idea. I'll have to do more research on integrated blade rotor turbine feasibility.
Quoting faro (Thread starter): If the air coming out of the HP compressor is slowed by the diffuser (leading to lower pressure I think?), why is it not blown out the front of the engine by the force of the combustion in the chambers?
The slowing of the air by the diffuser converts 'dynamic pressure' to 'static pressure.' There is always a favorable pressure gradient across the combustor (1.75% to 2.75% pressure loss across the liner and 0.15% to 0.2% pressure loss due to the combustion process or more properly "Rayleigh losses").
Now, fuel flow has to be ramped up to prevent the effect you are describing. On the PW4170A, there was a typo in the fuel flow map. We loved it! Boom! A 15' diameter fireball at ignition out the fan provided 'proof of light.' It did no harm to the engine, but for some reason the executives made us 'remove that feature.'
Quoting faro (Thread starter): What % of the oxygen in the air coming out of the HP compressor is not used in the combustion process? Is 100% oxygen burn desirable/achievable?
About 2/3rds of the oxygen. However, there is always cooling air entering the flow path that increases the oxygen content of the core as it goes through the turbine.
Quoting faro (Thread starter): Why not use coatings like Diamond-Like Carbon o
Not worth it. Some roughening of the blades actually helps.
Quoting faro (Thread starter): What is the lower limit of the number of fan blades (the GENx present has only 18) that can be used on an engine? What drives this limit?
Good question. I would have said 20 before the GEnX. Since the flow path needs to be covered, I really doubt we'll go below a dozen. But I Could be proven wrong. Ideally the number will drop to a prime number for vibration control reasons, but that requires getting down to 13 or 11 blades which will be tough.
Quoting faro (Thread starter): What are the pros/cons of reverse-flow combustors as implemented on the earliest RR jet engines?
Cons:
1. High pressure loss (due to turning a hot high velocity flow) which increases fuel burn.
2. More liner area to cool (increased cooling air means work to compress the cooling air without burning it with fuel). Thus an increase in fuel burn.
3. A horrid velocity and temperature profile into the turbine. This forces a compromise in turbine blade shape and an increase in cooling. Both result in lower high turbine efficiency and thus greater fuel burnl
4. Liner cooling air ingestion will increase the CO (carbon monoxide) output of the engine
Pros:
1. Shorter engine which reduces weight
2. Ability to re-adjust the radius the main flow path is on significantly. With centrifugal compressors, this allows the combustor to economically bring the 'flung out' air back into the optimal radius of the high turbine. To be blunt, due to the increase in fuel prices and lower manufacturing costs (today) of an axial compressor I do not see any new design, other than VLJ's and cruise missiles there is no advantage to a reverse flow combustor above 5,000 lbf of thrust.
Quoting faro (Thread starter): Why not have a hollow spinner admitting air that would flow through the insides of the spool shafts (and be ejected out the back end of the nozzle) to enhance cooling of compressor/turbine blades from the inside?
The shafts already are somewhat cooled by ambient air. But for most cooling, using 3.5 psia ambient altitude air would require *huge* pipes to cool anything. So slightly compressed air, even though it took fuel to compress the air, is usually a better trade off. Besides, compressed air needs to go to the air 'thrust bearing' anyway. It is best to re-use the air.
Quoting faro (Thread starter): Is ambient air used to cool the engine casing?
Fan air is used to cool the core casing. The fan casing is dang cold anyway! Also, compressed air is used to cool the turbine casing. Even fan air doesn't have enough pressure to provide sufficient cooling (the boundary layer needs to be compressed, this is the basic theory behind impingement cooling). Note: The turbine casing cooling must be actively controlled to keep turbine tip clearances at the optimum. Too little is abraded (worn) turbine blades. Too little is too much 'blow by' air that kills turbine efficiency.
faro From Egypt, joined Aug 2007, 1327 posts, RR: 0 Reply 4, posted (1 year 8 months 2 weeks 23 hours ago) and read 10402 times:
Thanx to all for your time and detailed replies, very much appreciated.
Quoting tdscanuck (Reply 2): Total pressure stays constant through the diffuser but, since the flow is slowing down, static pressure is increasing.
Quoting lightsaber (Reply 3): The slowing of the air by the diffuser converts 'dynamic pressure' to 'static pressu
Quoting lightsaber (Reply 3): ow, fuel flow has to be ramped up to prevent the effect you are describing.
If we try to reason in terms of forces and momentum:
1) The gas flow is slowed in the diffuser
2) At constant mass, speed loss (=loss of dynamic pressure) results in (net) rearward momentum (increase in static pressure) being imparted to the combustion chamber
3) This rearward momentum is sufficient to resist the forward component of combustion pressure and prevent it from going out the front end
Is this accurate? What about the combustion chamber/liner itself? Doesn't the confining/directing effect it has also direct the combustion pressure towards the rear? What is more important in preventing the combustion process from exiting the forward end à la PW4170 (lol), this liner confining effect or the increase in rearward static pressure imparted by the diffuser?
Quoting lightsaber (Reply 3): For most of the compressor, aluminum blades could be used, but titanium is still used as that is the 'know material.'
What about titanium aluminide? Any progress with that recently in compressors?
Also, some complementary questions for your kind consideration:
1) On military engines and some of the older low-bypass civil engines one always found variable inlet guide vanes. Why are these necessary on these applications and not on high bypass turbofans?
2) How much forward deflection (with respect to the plane on which the spinner sits) is experienced by the tips of the fans on big engines like the Trent at full power (the spinner is fixed because it's attached to the rotor and the fan tips are the area experiencing very high forward force; they must necessarily deflect forward)? How is the force generated by this forward deflection handled/routed if the main forward thrust force is borne by the thrust bearings in the engine's bowels? The fan containment ring looks rather flimsy in this respect (ie, handling significant forward force on the rim portion which sits forward of the fans) and seems mainly designed to handle centrifugal forces, not longitudinal ones.
3) Do the latest big fan engines use supercritical fan tip airfoil sections? Is this feasible in such very thin airfoil sections? Is it desirable in the first place? Are the fan tips also going around at transonic speeds in the cruise or at full thrust only?
4) One thing I have never understood: why did the TF39, the grand-daddy of big fan engines, have a bypass ratio of around 8 when all subsequent big fan engines (until quite recently I believe) hovered around 4-6? Higher bypass means higher efficiency, right? Does this mean the TF39 was more efficient than all its descendants in the CF6 lineage? I think not, so what gives?
5) Why are compressor blades uniformly graded in size towards the rear and not modified on an ad hoc basis for those stages which sit adjacent to bleed valves (where conceivably there is a pressure drop)? Some of those bleed holes/valves are rather big, why are the bleedless engines on the 787 not significantly more efficient than their non-bled brethren?
6) Has anyone ever tried running the engine gearbox off gears on the rim of the compressors rather than from the rotor? Wouldn't this reduced complexity/weight somewhat?
tdscanuck From Canada, joined Jan 2006, 11032 posts, RR: 72 Reply 5, posted (1 year 8 months 2 weeks 22 hours ago) and read 10377 times:
Quoting faro (Reply 4): If we try to reason in terms of forces and momentum:
Is this accurate?
I'm not really sure...you're using momentum and force in a way that I'm finding difficult to follow. Working with static pressure in engines is tough, usually you just stick with dynamic pressure (aka total pressure) so you don't have to worry about all the speed change. The short short version is that the flow goes backwards through the combustor because of a favorable total pressure gradient.
Quoting faro (Reply 4): 1) On military engines and some of the older low-bypass civil engines one always found variable inlet guide vanes. Why are these necessary on these applications and not on high bypass turbofans?
They're still there in high-bypass turbofans, they're just behind the fan...they're the variable stator vanes in the compressor.
Quoting faro (Reply 4): 2) How much forward deflection (with respect to the plane on which the spinner sits) is experienced by the tips of the fans on big engines like the Trent at full power (the spinner is fixed because it's attached to the rotor and the fan tips are the area experiencing very high forward force; they must necessarily deflect forward)?
If you look at the rub patterns on the abradables on large fans, you can see that it's pretty small...fractions of an inch.
Quoting faro (Reply 4): How is the force generated by this forward deflection handled/routed if the main forward thrust force is borne by the thrust bearings in the engine's bowels?
Aerodynamic forces pull the blades forward. This pulls the fan hub, which pulls the shaft, which is riding on the thrust bearing. The other side of the thrust bearing is carried by the engine frame. For most engines, the thrust force goes through thrust links from the engine frame to the back of the strut.
Quoting faro (Reply 4): The fan containment ring looks rather flimsy in this respect (ie, handling significant forward force on the rim portion which sits forward of the fans) and seems mainly designed to handle centrifugal forces, not longitudinal ones.
The fan case typically isn't carrying thrust loads.
Quoting faro (Reply 4): 5) Why are compressor blades uniformly graded in size towards the rear and not modified on an ad hoc basis for those stages which sit adjacent to bleed valves (where conceivably there is a pressure drop)?
Bleed air isn't always on, and the amount being extracted isn't under the control of the engine. Compressors do sometimes have sized gaps between stages to accomodate bleed.
Quoting faro (Reply 4): Some of those bleed holes/valves are rather big, why are the bleedless engines on the 787 not significantly more efficient than their non-bled brethren?
They are...that's one of the reasons they went bleedless.
Quoting faro (Reply 4): 6) Has anyone ever tried running the engine gearbox off gears on the rim of the compressors rather than from the rotor? Wouldn't this reduced complexity/weight somewhat?
Not that I'm aware of. Compressors don't have a rim on any design that I'm aware of right now (not counting centrifugals in small engines), so you'd have to add a large spinning rim to the compressor. Also, you'd be focussing the entire power offtake onto a single stage, rather than the whole spool.
faro From Egypt, joined Aug 2007, 1327 posts, RR: 0 Reply 6, posted (1 year 8 months 2 weeks 21 hours ago) and read 10368 times:
Quoting tdscanuck (Reply 5): Quoting faro (Reply 4):
1) On military engines and some of the older low-bypass civil engines one always found variable inlet guide vanes. Why are these necessary on these applications and not on high bypass turbofans?
They're still there in high-bypass turbofans, they're just behind the fan...they're the variable stator vanes in the compressor.
Sorry, I meant inlet guide vanes *ahead* of the fan, not behind. May be wrong but I think that the vanes ahead of the fan do not have the same function as those in the compressor.
lightsaber From United States of America, joined Jan 2005, 8537 posts, RR: 100 Reply 7, posted (1 year 8 months 2 weeks 20 hours ago) and read 10329 times:
Quoting tdscanuck (Reply 5): The short short version is that the flow goes backwards through the combustor because of a favorable total pressure gradient.
Except in the diffuser, which is a custom component designed to convert rho*u^2/2 back into static pressure, there is a static pressure drop through the compressor. Fuel flow rate (on and off) is controlled to prevent a reverse pressure gradient of the type faro was asking about.
Momentum isn't the way to look at it. Use Bournoulli's equation Faro. It is actually energy.
Quoting faro (Reply 4): 3) Do the latest big fan engines use supercritical fan tip airfoil sections?
Yes. They are required as tip speeds are Mach 1.1 for older engines to Mach 1.35 for the planned fan of the LEAP-X (at the fan tips).
Quoting faro (Reply 4): why did the TF39, the grand-daddy of big fan engines, have a bypass ratio of around 8 when all subsequent big fan engines (until quite recently I believe) hovered around 4-6?
Notice how different thrust ratings of the same engine have different bypass ratios? If you 'under-run' an engine it will not accept all the air that should go into the core (there will be a backpressure that deflects more of the air to the fan). Thus the same engine will have a higher bypass ratio when run at a 'gentle thrust setting.'
Gran daddy wasn't made to run very hard...
Quoting faro (Reply 4): Why are compressor blades uniformly graded in size towards the rear and not modified on an ad hoc basis for those stages which sit adjacent to bleed valves (where conceivably there is a pressure drop)?
There is an 'ad hoc' modification near the bleed flows. It is just subtle. One cannot stall the flow and bleed flows are a fraction of the main flow.
Quoting tdscanuck (Reply 5): This pulls the fan hub, which pulls the shaft, which is riding on the thrust bearing.
But it should also be noted that the thrust bearing is actively controlled to position the low spool at an optimum position.
Quoting faro (Reply 6): Sorry, I meant inlet guide vanes *ahead* of the fan
Those were there for structural and not aerodynamic reasons. With better rotor dynamics (includes shaft design, bearings, etc.) they are not needed for structure. Aerodynamically they were a loss. Not to mention a favorite place for ice to build up to ruin that lovely fan. If you look at those vanes ahead of the fan, they had some wicked schemes to melt the ice... Schemes that cost a bit in fuel burn. So they are gone! For commercial engines that is.
faro From Egypt, joined Aug 2007, 1327 posts, RR: 0 Reply 8, posted (1 year 8 months 1 week 6 days 22 hours ago) and read 10203 times:
Quoting lightsaber (Reply 3): A horrid velocity and temperature profile into the turbin
I don't quite understand why reverse-flow combustion engines have a higher speed gas flow going into the turbines. Do the two 180° turns accelerate the flow?
Quoting lightsaber (Reply 7): Quoting faro (Reply 4):
3) Do the latest big fan engines use supercritical fan tip airfoil sections?
Yes. They are required as tip speeds are Mach 1.1 for older engines to Mach 1.35 for the planned fan of the LEAP-X (at the fan tips).
Wow, so they are actually going supersonic and not just transonic! Supercritical airfoils may lessen the strength of the shock wave forming off the top of the fan blades but will not attenuate it altogether; what about those shock waves bouncing about behind the fan and into the nacelle duct? What is their effect if any? Also, is this supersonic fan tip speed responsible for the highly-swept outer fan blade profile implemented on the latest big fan engines like the Trent 900?
Quoting lightsaber (Reply 7): Those were there for structural and not aerodynamic reasons. With better rotor dynamics (includes shaft design, bearings, etc.) they are not needed for structure. Aerodynamically they were a loss. Not to mention a favorite place for ice to build up to ruin that lovely fan. If you look at those vanes ahead of the fan, they had some wicked schemes to melt the ice... Schemes that cost a bit in fuel burn. So they are gone! For commercial engines that is.
Are military engines the exception because of the high G forces imposed during maneuvering and the need to keep the rotor centered when at high G? Any other reasons?
lightsaber From United States of America, joined Jan 2005, 8537 posts, RR: 100 Reply 9, posted (1 year 8 months 1 week 6 days 17 hours ago) and read 10141 times:
Quoting faro (Reply 8): I don't quite understand why reverse-flow combustion engines have a higher speed gas flow going into the turbines. Do the two 180° turns accelerate the flow?
It is not the gas speed going into the turbine that changes, but rather the radial distribution.
High velocity at the turbine blade tip is bad. It cooks the end of the turbine blade and shortens the life of the turbine dramatically. So that is to be avoided (and is with reverse flow combustors).
High velocity at the root of the turbine blade is also bad. It cooks the highest tress area of the turbine blade.
So for optimal fuel burn one wants high velocity in the middle of the combustor air stream and low velocity at the walls. A reverse flow combustor cooks the root (inner diameter) of the turbine blade. That requires more metal (that screws up the aerodynamic shape of the blade) and cooling (this is the most costly, from fuel burn, cooling in the engine).
That that is what I mean by 'velocity profile.'
Quoting faro (Reply 8): Supercritical airfoils may lessen the strength of the shock wave forming off the top of the fan blades but will not attenuate it altogether; what about those shock waves bouncing about behind the fan and into the nacelle duct? What is their effect if any?
True. I'm not an expert in that area, so I'll let someone else comment. There is sound absorbing material that will dampen the shock waves, but it is not a region of the engine I spend time engineering.
kl671 From Canada, joined Jul 2005, 128 posts, RR: 0 Reply 10, posted (1 year 8 months 1 week 6 days 16 hours ago) and read 10132 times:
Quoting faro (Thread starter): What are the pros/cons of reverse-flow combustors as implemented on the earliest RR jet engines?
The P&WC PT6 family all have reverse-flow combustors and with some 40,000 examples delivered, in all of its variants, this engine has to be the among the most produced gas turbine engine in the world. In the PT6, the pros of reverse-flow combustors obiviously out weight the cons.
Does anyone out there know of a gas turbine family that has sold in greater numbers than the PT6?
tdscanuck From Canada, joined Jan 2006, 11032 posts, RR: 72 Reply 11, posted (1 year 8 months 1 week 6 days 15 hours ago) and read 10114 times:
Quoting faro (Reply 8): Supercritical airfoils may lessen the strength of the shock wave forming off the top of the fan blades but will not attenuate it altogether; what about those shock waves bouncing about behind the fan and into the nacelle duct? What is their effect if any? Also, is this supersonic fan tip speed responsible for the highly-swept outer fan blade profile implemented on the latest big fan engines like the Trent 900?
The shocks dissipate quickly...the flow in the duct itself isn't supersonic, it's just supersonic relative to the fan blades. And yes, this is partly why the fan blades are swept strongly.
Quoting kl671 (Reply 10): The P&WC PT6 family all have reverse-flow combustors and with some 40,000 examples delivered, in all of its variants, this engine has to be the among the most produced gas turbine engine in the world. In the PT6, the pros of reverse-flow combustors obiviously out weight the cons
The PT6 turbine is *tiny*, so the velocity gradients along the blade probably aren't large...I suspect the compactness of the engine massively outweights any inconvenience of the reverse flow combustor.
kl671 From Canada, joined Jul 2005, 128 posts, RR: 0 Reply 12, posted (1 year 8 months 1 week 6 days 13 hours ago) and read 10096 times:
Quoting tdscanuck (Reply 11): The PT6 turbine is *tiny*, so the velocity gradients along the blade probably aren't large...I suspect the compactness of the engine massively outweights any inconvenience of the reverse flow combustor.
When did we introduce the size of the engine into this discussion.? I was simply pointing out that the the PT6 is probably the most succussful gas turbine engine in history and it has a reverse flow combustor.
Does your "tiny" theory also apply to the 5000 hp PW150A which I suspect is the most powerful turboprop engine flying in commercial passanger service today, unless there are a few TU-114's still out there. The PW150A has a reverse flow combustor.
BMI727 From United States of America, joined Feb 2009, 11547 posts, RR: 27 Reply 13, posted (1 year 8 months 1 week 6 days 12 hours ago) and read 10083 times:
Quoting faro (Reply 8): Supercritical airfoils may lessen the strength of the shock wave forming off the top of the fan blades but will not attenuate it altogether; what about those shock waves bouncing about behind the fan and into the nacelle duct?
They probably will not have too much effect on air entering the core, since the center parts of the fan are most likely not supersonic. But, the blades going supersonic does give a really annoying buzzing sound, at least on some engines.
Quoting faro (Reply 8): Are military engines the exception because of the high G forces imposed during maneuvering and the need to keep the rotor centered when at high G? Any other reasons?
Military aircraft sometimes have to use tricks in the inlets to control shockwaves. Also, military aircraft are more likely to be operating at odd angles of attack, which can be hard for the inlets. (See the problems early F-14s had with compressor stalls)
Edit: Just to add on a bit, the Concorde also have variable geometry intake ramps to control the flow before it got to the engine. They were also designed to snap shut if an engine failed, as the drag from a windmilling engine at Mach 2 could induce such a yaw that the aircraft could break up due to aerodynamic stresses.
Quoting kl671 (Reply 12): When did we introduce the size of the engine into this discussion.?
Because the compactness of the PT6 is part of what makes it attractive, and that same small size helps mitigate the cons of reverse flow combustion.
[Edited 2010-09-14 22:59:11]
Why do Aerospace Engineering students have to turn things in on time?
faro From Egypt, joined Aug 2007, 1327 posts, RR: 0 Reply 14, posted (1 year 8 months 1 week 6 days 10 hours ago) and read 10055 times:
Quoting kl671 (Reply 12): Does your "tiny" theory also apply to the 5000 hp PW150A which I suspect is the most powerful turboprop engine flying in commercial passanger service today, unless there are a few TU-114's still out there. The PW150A has a reverse flow combustor
Maybe it has something to do with centrifugal compressors. Perhaps it's difficult to feed straight-through combustors directly from centrifugal compressors? The Kuznetsov NK-12 powering the Tu-95 has a straight-through combustor but is a pure axial turboprop.
lightsaber From United States of America, joined Jan 2005, 8537 posts, RR: 100 Reply 15, posted (1 year 8 months 1 week 6 days 1 hour ago) and read 9980 times:
Quoting tdscanuck (Reply 11): .I suspect the compactness of the engine massively outweights any inconvenience of the reverse flow combustor.
And allowed for a far lower cost compressor.
Quoting faro (Reply 14): Perhaps it's difficult to feed straight-through combustors directly from centrifugal compressors?
Exactly. It is possible, but the deficiencies of a reverse flow combustor are 'noise' at that point with many benefits.
tdscanuck From Canada, joined Jan 2006, 11032 posts, RR: 72 Reply 17, posted (1 year 8 months 1 week 6 days 1 hour ago) and read 9976 times:
Quoting kl671 (Reply 12): When did we introduce the size of the engine into this discussion.? I was simply pointing out that the the PT6 is probably the most succussful gas turbine engine in history and it has a reverse flow combustor.
We introduced radial flow variation at the turbine inlet as one of the issues with reverser flow combustors. The PT-6 has a very small turbine (the blades are less than an inch long) spinning on a fairly descent sized disk, so it's harder to get a large radial gradient, which mitigates some of the issues of the reverse flow. At the same time, compactness is a very desirable attribute for engines in the PT-6 class, which biases the trade study heavily towards a reverse flow combustor.
Quoting kl671 (Reply 12): Does your "tiny" theory also apply to the 5000 hp PW150A which I suspect is the most powerful turboprop engine flying in commercial passanger service today, unless there are a few TU-114's still out there. The PW150A has a reverse flow combustor.
There are other reasons for reverse flow combustors, as lightsaber said in Reply 3. These types of decision are always a trade between multiple competing factors and the design goals between different engines can heavily alter where the optimum point is.
faro From Egypt, joined Aug 2007, 1327 posts, RR: 0 Reply 18, posted (1 year 8 months 1 week 6 days 1 hour ago) and read 9970 times:
Another question, with judicious routing of the air from the diffuser, it is possible to have a liner-less combustor, one where there is no (solid) containment of the combustion process, neither liner nor combustion chamber apart from the outer combustor casing? Can you encase the combustion flames in a jet of air that will provide guidance/insulation/cooling and not have the flames go out?
tdscanuck From Canada, joined Jan 2006, 11032 posts, RR: 72 Reply 19, posted (1 year 8 months 1 week 6 days ago) and read 9943 times:
Quoting A342 (Reply 16): Quoting lightsaber (Reply 3):
Ideally the number will drop to a prime number for vibration control reasons
Now you got me interested. Could you please explain a bit why that is the case or refer me (us?) to a good explanation?
Any time a blade lines up with a stator, you get an impulse to the structure as the blade's wake passes over the stator. You want to smooth out those impulses as much as you can, which means you do not want every stator to get hit at the same time. The worse case would be the same number of blades as stators...you get a big "whack" as ever stator gets hit at the same time. So you want the minimum number of blades (minimum number of wakes interfering) and the minimum number of common factors between blade number and stator number (minimize the number of stators that get hit at the same time). If you have a prime number of blades and a different number of stators, it's impossible to hit more than one stator at the same time because there are no common factors. There are other reasons that prevent you from going to the practical minimum (two blades and three stators), like fan solidity, but from a vibration point of view that's where they'd like to go.
Quoting faro (Reply 18): Another question, with judicious routing of the air from the diffuser, it is possible to have a liner-less combustor, one where there is no (solid) containment of the combustion process, neither liner nor combustion chamber apart from the outer combustor casing?
In theory, yes, but it would be very difficult to achieve the level of mixing you need for good combustion. You want lots of mixing in the combustor to promote even and total combustion, which mostly precludes the nice smooth airflow you'd get without any kind of liner or shaped chamber.
Quoting faro (Reply 18): Can you encase the combustion flames in a jet of air that will provide guidance/insulation/cooling and not have the flames go out?
Yes...this is basically how GE's TAPS combustor works:
A342 From Germany, joined Jul 2005, 4627 posts, RR: 4 Reply 20, posted (1 year 8 months 1 week 5 days 23 hours ago) and read 9915 times:
Quoting tdscanuck (Reply 19): Any time a blade lines up with a stator, you get an impulse to the structure as the blade's wake passes over the stator. You want to smooth out those impulses as much as you can, which means you do not want every stator to get hit at the same time. The worse case would be the same number of blades as stators...you get a big "whack" as ever stator gets hit at the same time. So you want the minimum number of blades (minimum number of wakes interfering) and the minimum number of common factors between blade number and stator number (minimize the number of stators that get hit at the same time). If you have a prime number of blades and a different number of stators, it's impossible to hit more than one stator at the same time because there are no common factors. There are other reasons that prevent you from going to the practical minimum (two blades and three stators), like fan solidity, but from a vibration point of view that's where they'd like to go.
faro From Egypt, joined Aug 2007, 1327 posts, RR: 0 Reply 21, posted (1 year 8 months 1 week 5 days 19 hours ago) and read 9877 times:
Quoting tdscanuck (Reply 19): Any time a blade lines up with a stator, you get an impulse to the structure as the blade's wake passes over the stator. You want to smooth out those impulses as much as you can, which means you do not want every stator to get hit at the same time. The worse case would be the same number of blades as stators...you get a big "whack" as ever stator gets hit at the same time. So you want the minimum number of blades (minimum number of wakes interfering) and the minimum number of common factors between blade number and stator number (minimize the number of stators that get hit at the same time). If you have a prime number of blades and a different number of stators, it's impossible to hit more than one stator at the same time because there are no common factors. There are other reasons that prevent you from going to the practical minimum (two blades and three stators), like fan solidity, but from a vibration point of view that's where they'd like to go.
I agree with A342, fascinating explanation there Tom, thanx!
speedygonzales From Norway, joined Sep 2007, 608 posts, RR: 0 Reply 22, posted (1 year 8 months 1 week 5 days 4 hours ago) and read 9799 times:
Quoting tdscanuck (Reply 19): If you have a prime number of blades and a different number of stators, it's impossible to hit more than one stator at the same time because there are no common factors.
lightsaber From United States of America, joined Jan 2005, 8537 posts, RR: 100 Reply 24, posted (1 year 8 months 1 week 3 days 18 hours ago) and read 9603 times:
Quoting speedygonzales (Reply 22): None of the two numbers need to be prime to achieve that, they only need to be coprime.
But as one gets to very low blade counts, it is likely to be a prime number.
6 and 8 blades can share some harmonics, but that can be tuned out.
As to the source... I'm trying to recall which textbook I learned the theory from...
faro From Egypt, joined Aug 2007, 1327 posts, RR: 0 Reply 25, posted (1 year 8 months 1 week 3 days 6 hours ago) and read 9566 times:
Quoting lightsaber (Reply 24): Quoting speedygonzales (Reply 22):
None of the two numbers need to be prime to achieve that, they only need to be coprime.
But as one gets to very low blade counts, it is likely to be a prime number.
6 and 8 blades can share some harmonics, but that can be tuned out.
As to the source... I'm trying to recall which textbook I learned the theory from...
My initial impression re the prime number of *fan* blades (where blade wake passes are perhaps less of an issue given that the nacelle flow de-swirler stators are some distance back from the plane of the fan) was if the number is prime, you do not have any radial symmetry. Perhaps this may lessen the number of possible vibration modes/harmonics of the fan itself?
In the above link to a turbofan cut-away drawing, the disks that bear the HP compressor/turbine blades on their outer edges and which extend inwards in the direction of the rotor shaft have a distinctive thickening on the inner edges which is triangular in cross-section. What is the purpose of this thick inner edge design?
tdscanuck From Canada, joined Jan 2006, 11032 posts, RR: 72 Reply 27, posted (1 year 7 months 3 weeks 4 days 23 hours ago) and read 9049 times:
Quoting faro (Reply 26): What is the purpose of this thick inner edge design?
To maintain approximately constant stress. As you move inwards the centrifugal load is going up (since there's more mass outside your reference point to restrain) and the circumference of the disk at your reference point is dropping. The only way to hold the material stress constant is to increase the area.
jetlife2 From United States of America, joined Jul 2006, 187 posts, RR: 22 Reply 28, posted (1 year 7 months 3 weeks 4 days 19 hours ago) and read 8983 times:
Quoting faro (Reply 25): My initial impression re the prime number of *fan* blades (where blade wake passes are perhaps less of an issue given that the nacelle flow de-swirler stators are some distance back from the plane of the fan) was if the number is prime, you do not have any radial symmetry. Perhaps this may lessen the number of possible vibration modes/harmonics of the fan itself?
There are three families of modes to consider
1 The modes of the fan blade itself. These modes are defined by the geometry of each blade. For example, first flex mode is the mode where a fan blade is flexing at the root only (there is one node only and it is at the root). The frequency of these modes changes as the fan speeds up, because of centrifugal stiffening (tends to raise the frequency) and because the fan blade is subjected to higher temperatures in operation (tends to lwer the frequency)
2 The system modes of the fan blades in assembly with the disk. These modes are a function of all the variables in 1. plus the number of blades, the spacing, the disk geometry and material. Think of an umbrella as an example.
3. The modes of the OGVs (downstream airfoils).
There are an infinite number of all of these modes - but the ones that matter are the ones that are excited (energized) during normal operating range of the engine. If there is no exciting force, the presence of a mode at a certain frequency is benign. Also very high order modes (like say 5th flex etc) have very little energy in them and are almost never of concern.
So the number of airfoils in the fan stage and OGV stage will control the blade passing frequencies, and hence whether any of the modes in 1, 2 and 3 are excited.
All of this is studied on a Campbell diagram. Probably there are some good links out there that describe the construction and use of this tool.
faro From Egypt, joined Aug 2007, 1327 posts, RR: 0 Reply 29, posted (1 year 7 months 3 weeks 4 days 8 hours ago) and read 8873 times:
Quoting tdscanuck (Reply 27): Quoting faro (Reply 26):
What is the purpose of this thick inner edge design?
To maintain approximately constant stress. As you move inwards the centrifugal load is going up (since there's more mass outside your reference point to restrain) and the circumference of the disk at your reference point is dropping. The only way to hold the material stress constant is to increase the area
Thanx for the explanation; just one thing I don't understand though, on a conceptual level. If the disk is not attached to the rotor, there is no centripetal force acting on the disk's thickened inner edge to counter the centrifugal force acting outward, is there? In this case, my understanding would be that the only centripetal force to be met is the one at the disk's slim outer edge which is restrained by the outer ring where all the blades attach. If this is the case, then greatest centrifugal stress loading would be at the disk's outer end where it meets the outer (restraining) ring. Not at all sure about this though...
Quoting jetlife2 (Reply 28): The system modes of the fan blades in assembly with the disk. These modes are a function of all the variables in 1. plus the number of blades, the spacing, the disk geometry and materia
Thanx Jetlife; in this case, do you also do a prime number of blades to reduce the possible modes of vibration of the fan baldes in assembly with the disk? Does radial symmetry matter or is less of a consideration given all the other variables?
Some of the blade pairs used to balance the compressor stage have a weight difference of up to 3.5% which seems quite a lot for the sophisticated contraptions that are jet engines. Are these used blades or is it normal to get such weight differences on new ones too? If used, are these diffs due to erosion from particulate matter or is there some other factor like thermal cycling, etc involved? When rebalancing the fan, are the weight differences even greater due to direct exposure to the environment?
jetlife2 From United States of America, joined Jul 2006, 187 posts, RR: 22 Reply 31, posted (1 year 7 months 3 weeks 4 days 1 hour ago) and read 8828 times:
Quoting faro (Reply 29): in this case, do you also do a prime number of blades to reduce the possible modes of vibration of the fan baldes in assembly with the disk? Does radial symmetry matter or is less of a consideration given all the other variables?
No. The GE90 has 22 and the GEnx has 18. All else being equal (i.e assuming that the blade can be designed to meet performance, impact, etc etc) it is usually desirable to reduce the number. Then adjust the number of OGVs accordingly. No need to stick to a prime number, there are not enough of them. As you get to very low blade counts though the blade is getting longer in chord and heavier, so there is a system weight trade to be done.
tdscanuck From Canada, joined Jan 2006, 11032 posts, RR: 72 Reply 32, posted (1 year 7 months 3 weeks 4 days ago) and read 8800 times:
Quoting faro (Reply 29): If the disk is not attached to the rotor, there is no centripetal force acting on the disk's thickened inner edge to counter the centrifugal force acting outward, is there?
The disk and the rotor are the same thing; we might have a terminology problem here. The centrifugal force (calm down Newtonians, we're working in a rotating reference frame here) is what's required to keep the blades going in a circle. That's countered by hoop stress in the disk. You need to have enough material to resist the hoop stress, which goes all the way through the disk. The edge of the disk isn't capable of restraining the blades all by itself (not enough material).
Quoting jetlife2 (Reply 31): The GE90 has 22 and the GEnx has 18. All else being equal (i.e assuming that the blade can be designed to meet performance, impact, etc etc) it is usually desirable to reduce the number.
A non-noise issue that tends to prefer even numbers is balancing...with even numbers, you have blade pairs opposite and you can do matched sets for balance. It's doable, but more complicated, if you don't have opposing pairs.
kl671 From Canada, joined Jul 2005, 128 posts, RR: 0 Reply 33, posted (1 year 7 months 3 weeks 3 days 16 hours ago) and read 8717 times:
Quoting tdscanuck (Reply 32): The disk and the rotor are the same thing; we might have a terminology problem here.
A rotor in a gas turbine engine refers to a rotating assembly, which would consist of a shaft, one or more disks, (depending on the number of stages), blades, oil scoops and every thing else that turned with the shaft. Bearings and seals would be included if installed on the shaft before installation.
The term is used to differentiate between rotating and static assemblies, thus rotors and stators.
In some instances, disk and rotor can be used to describe the same component, such as in brakes for example.
faro From Egypt, joined Aug 2007, 1327 posts, RR: 0 Reply 34, posted (1 year 7 months 3 weeks 3 days 9 hours ago) and read 8688 times:
Quoting tdscanuck (Reply 5): Quoting faro (Reply 4):
1) On military engines and some of the older low-bypass civil engines one always found variable inlet guide vanes. Why are these necessary on these applications and not on high bypass turbofans?
They're still there in high-bypass turbofans, they're just behind the fan...they're the variable stator vanes in the compressor.
Quoting lightsaber (Reply 7): Quoting faro (Reply 6):
Sorry, I meant inlet guide vanes *ahead* of the fan
Those were there for structural and not aerodynamic reasons. With better rotor dynamics (includes shaft design, bearings, etc.) they are not needed for structure. Aerodynamically they were a loss. Not to mention a favorite place for ice to build up to ruin that lovely fan. If you look at those vanes ahead of the fan, they had some wicked schemes to melt the ice... Schemes that cost a bit in fuel burn. So they are gone! For commercial engines that is.
aircellist From Canada, joined Oct 2004, 1302 posts, RR: 9 Reply 35, posted (1 year 7 months 3 weeks 1 day 1 hour ago) and read 8424 times:
Quoting jetlife2 (Reply 28): There are an infinite number of all of these modes - but the ones that matter are the ones that are excited (energized) during normal operating range of the engine. If there is no exciting force, the presence of a mode at a certain frequency is benign. Also very high order modes (like say 5th flex etc) have very little energy in them and are almost never of concern.
I feel there is something musical in this...
"Mort aux cons!" (anonyme) "Vaste programme..." (Charles de Gaulle)
faro From Egypt, joined Aug 2007, 1327 posts, RR: 0 Reply 36, posted (1 year 7 months 3 weeks 9 hours ago) and read 8325 times:
Why do engine manufacturers implement active compressor blade tip clearance instead of having some sort of thin, flexible tip shroud that would attach to and encircle the blades tips and afford zero tip clearance like with turbine blades? Perhaps the weight of such a shroud would be less than the active tip clearance set-up and it would certainly be less complex. Maybe CFRP materials could work in such an application?
tdscanuck From Canada, joined Jan 2006, 11032 posts, RR: 72 Reply 37, posted (1 year 7 months 3 weeks 3 hours ago) and read 8276 times:
Quoting faro (Reply 36): Why do engine manufacturers implement active compressor blade tip clearance instead of having some sort of thin, flexible tip shroud that would attach to and encircle the blades tips and afford zero tip clearance like with turbine blades?
That just moves the problem...now you have to seal between the tip shroud and the case. This is exactly why you implent active turbine clearance control.
faro From Egypt, joined Aug 2007, 1327 posts, RR: 0 Reply 38, posted (1 year 7 months 2 weeks 5 days 10 hours ago) and read 8131 times:
Quoting tdscanuck (Reply 37): Quoting faro (Reply 36):
Why do engine manufacturers implement active compressor blade tip clearance instead of having some sort of thin, flexible tip shroud that would attach to and encircle the blades tips and afford zero tip clearance like with turbine blades?
That just moves the problem...now you have to seal between the tip shroud and the case. This is exactly why you implent active turbine clearance control.
Tom
Forgot about that! Thanx.
I guess for the blades in the turbine leakage is not a big issue because you are very close to the outlet nozzle in any case? Or do you also have to hermetically seal the stages between turbine blades and stator vanes?
tdscanuck From Canada, joined Jan 2006, 11032 posts, RR: 72 Reply 39, posted (1 year 7 months 2 weeks 5 days 3 hours ago) and read 8097 times:
Quoting faro (Reply 38): I guess for the blades in the turbine leakage is not a big issue because you are very close to the outlet nozzle in any case?
It's still a big issue...i'ts actually a bigger issue because the pressure differential across a turbine stage is considerably larger than the differential across a compressor stage. This is why you see shrouds and active clearance control on most turbines, but few compressors. Thee shrouds do help, but they don't eliminate the problem and they come at a cost.
Quoting faro (Reply 38): Or do you also have to hermetically seal the stages between turbine blades and stator vanes?
I don't think anybody knows how to hermetically seal the stages...if they do, they should patent that right now. The most common thing I've seen is a labyrinth seal for the turbines (both sides) and the inside of the compressor gas-path, and a clearance seal for the outside of the compressor gas path.
jetlife2 From United States of America, joined Jul 2006, 187 posts, RR: 22 Reply 40, posted (1 year 7 months 2 weeks 5 days 1 hour ago) and read 8082 times:
Quoting aircellist (Reply 35): I feel there is something musical in this...
Yes you are right
We speak about modes, harmonics, orders, resonance, damping, excitation
Much of the language and all of the math is the same.
Think of a cello string...it has infinite modes (notes)...but you have no sound (response) without excitation (bowing)....bowing imparts energy that is multi-frequency in content and excites the fundamental modes...the mode you will hear the most is the 1st open string mode, but also lesser content of the 2nd, 3rd, etc; plus response of the system (cello structure, air in the cavity...) which makes the unique sound of that combination. Now modify the length of the string (fingering)...and thereby change the modes of both the string and the structure...
in the engine the analogy is the fan blade is the string, the rotor structure is the cello, the bow is the airflow...
aircellist From Canada, joined Oct 2004, 1302 posts, RR: 9 Reply 41, posted (1 year 7 months 2 weeks 4 days 20 hours ago) and read 8044 times:
Quoting jetlife2 (Reply 40): Think of a cello string...it has infinite modes (notes)...but you have no sound (response) without excitation (bowing)....bowing imparts energy that is multi-frequency in content and excites the fundamental modes...the mode you will hear the most is the 1st open string mode, but also lesser content of the 2nd, 3rd, etc; plus response of the system (cello structure, air in the cavity...) which makes the unique sound of that combination. Now modify the length of the string (fingering)...and thereby change the modes of both the string and the structure...
Would what you call "modes" be what in my musician's language I would call "fundamental tone" and "harmonics"?
I believe I would gladly spend an afternoon discussing those matters! Only thing, it is much easier to bring a cello to the discussion!
"Mort aux cons!" (anonyme) "Vaste programme..." (Charles de Gaulle)
Sponsor Message: 
From Egypt, joined Aug 2007, 1327 posts, 
From United States of America, joined May 2004, 1554 posts, 
From Canada, joined Jan 2006, 11032 posts, 
We loved it! Boom! A 15' diameter fireball at ignition out the fan provided 'proof of light.' It did no harm to the engine, but for some reason the executives made us 'remove that feature.' 
Except in the diffuser, which is a custom component designed to convert rho*u^2/2 back into static pressure, there is a static pressure drop through the compressor. Fuel flow rate (on and off) is controlled to prevent a reverse pressure gradient of the type faro was asking about.
So they are gone! For commercial engines that is.
From Germany, joined Jul 2005, 4627 posts, 
Thank you very much!
From Norway, joined Sep 2007, 608 posts, 
Sponsor Message: 
Printer friendly format
){kind=link}