PDA

View Full Version : Large turbofan noise


Seloco
10th Jun 2003, 21:07
With apologies in advance if this has been raised before........

As a frequent pax/TAP/sad git (choose your label) living under the LHR outbound flightpath for 09R (and the occasional 09L) I can't help but notice the significant difference in t/o noise between BA's "old" 777s with GE90s and the newer, Trent-powered version. To put it succinctly, the former ones "whistle" whilst the latter ones "growl". The question is, why such a difference in tone from engines of similar size and power? Is it the Trent's extra shaft, perhaps, or that it has a snubberless fan, that makes its sound so much lower in frequency (and, I hate to say, somewhat noisier)? Can any engine techies out there enlighten please?

Intruder
11th Jun 2003, 00:35
Differences in the speed and shape of the fan blades are most likely the source of the difference in sound.

Seloco
11th Jun 2003, 00:56
I agree, Intruder, that that is the most likely explanation, but the noise difference is so marked that the fan speed/shape must also be significantly different between the two. I wonder therefore if one characteristic of the Trent's three-shaft design is to produce a much slower rotating fan and therefore a lower pitched-noise? Hopefully an engineer will read this thread and comment. Incidentally I have yet to be close to an A340-600 at take-off but I would expect there to be a similarly marked difference between it and a CFM-56 powered -200 or -300.

747FOCAL
11th Jun 2003, 02:07
It is the shape of the fan and the speed that causes some change in the sound. There are also other differences that make the engines sound different.

:cool:

cirrus01
11th Jun 2003, 04:36
Two distinctive engine noise profiles...... CF6-50s in the DC-10 at idle and The rasping RB211-535s as fitted in the 757 at high power.

lomapaseo
11th Jun 2003, 06:04
Yes to all answers above.

HotDog
11th Jun 2003, 12:14
I doubt if the fan speed makes much difference to the noise print.

100% N1 RB211-524D4=3,900rpm, fan diameter=95.00"
JT9D-70A =3,600rpm -:- =95.56"
CF6-50 =3,433rpm

The fan bypass ratio of 5:1 on the RB211 is the highest of all three engines but there is really not much difference in the basics between them except the RB211 of course, which is a three spool engine.

747FOCAL
12th Jun 2003, 00:18
Hotdog,

Unless I missunderstood what you said..........Fan Speed makes all the difference when it comes to engine noise. In fact, lowering RPM by 10 will change the noise output of an aircraft by a few tenths of a dB. Not that the noise source has changed, the increase in noise is mostly due to being lower over the mic.

Example: Adding the N1 monitor to CF650 engines lowered climb out RPM by about 9 RPM this was a 0.2 EPNdB change.

I could spend a few hours taking those that care through Noise 101, but I don't think anybody would really care.

As an FYI - the GE powered 777s are certified quieter than the Trents. The growl you hear is what is called "Buzzsaw" noise and will be fixed by a new inlet that will be available come July this year. I don't understand the whistle noise you are claiming to hear. It could be turbo machinery noise that is louder than the engines, but if that is true the noise certification of those aircraft is bogus.

:{ :{ :mad:

Intruder
12th Jun 2003, 00:42
The primary/loudest noise from the front engine will almost always be from the fan. Not only the fan speed, but the number of blades will determine the basic noise frequency. Since noise is propagated in air, the movement of more air (i.e., higher thrust settings) will also directly affect the volume/level of the noise.

The character of the perceived noise will be affected by local airflow through the inlets, the frequency and level of noise that emits from the compressor section, exhaust noise that can be heard from in front of the engine, vibrations in the nacelle and airframe caused by the main engine noise, and interactions between all those noise sources, vibrations, and structures.
Noise that we perceive as whines, growls, hoots, and shrieks are the products of those interactions.

The 2-spool vs 3-spool configuration undoubtedly makes SOME difference in perceived noise, but it would be in the secondary interactions more than in the primary fan noise.

phnuff
12th Jun 2003, 03:36
Hmm interesting thread this one.

95" diameter at 3900rpm. Unless I have done my sums wrong, that equates to a blade tip speed of somewhere over 1100mph, which to me as the guy sitting back in the cabin with his G&T is quite staggering (even without the G&T). I know at normal T&P's, that would be way over the speed of sound, however of course, a jet engine is far from a place of normal T&P. Assuming my calculation is correct (and I am no way suggesting the sonic shock wave contributes to the sound heard outside), how is the transition to supersonic blade speed which presumably takes only a few seconds handled within the engine ? It must put unbelievable stresses on the components.

If I am talking cr*p or asking cr*p questions, please dont shoot me down in too many flames. Just take it as an example of a little knowledge being a dangerous thing.

747FOCAL
12th Jun 2003, 04:49
Yes fan tip speed is supersonic. Same as on prop aircraft. When You see a prop aircraft in the sky that is not the engine you are hearing. That is the tip of the blade smacking the air as it dips in and out of sonic.


If you have the chance to get near where the pilots are pushing the TOGA button you will hear a "WUMP" as the blades go sonic.

The amount of pressure on the engine is quite large. It can be -60 out and the blades will be amazingly hot and that is why they are hollow.



:cool:

phnuff
12th Jun 2003, 05:22
I supose its obvious that it would be supersonic I guess - I think the tip of the PA28 that I normally fly is a meager 330 mph at 2500. Ho hum rather a lot less energy


I guess this opens a number of other questions - at least to me

Is there presumably a max rpm for blades after which they stop being able to 'grip' the air

Are there any variable pitch blades ?

what does -60 out mean ??

What are the fan speeds on the -400 747 ?

I could go on all night - this is facinating

HotDog
12th Jun 2003, 08:02
Fan blade angle is fixed, -60 is OAT in degrees C, -400 blade speed depends on engine fit but similar to previous figures quoted.

747 Focal, what I meant was that the comparison of noise between the P&W, GE and RR engines would not be generated by the difference in fan speeds alone as the speeds do not differ that much. I take your point though.

phnuff
12th Jun 2003, 17:14
Thanks for the answers to my questions guys.

It just makes me realise just how clever mankind has got to be able to contain and work with these extreems (and to make them spin in those conditions) I could go on asking questions on the subject for ever, but I think people would get bored with me.

Is there any laymans guide to this technology kicking around - a sort of 'Engine design for dummys' (as per the series more common in my IT world) ?

HotDog
12th Jun 2003, 17:45
Rolls Royce published an excellent book titled The Jet Engine. You might still be able to get hold of a copy , write to Rolls-Royce Ltd. Derby Engine Division, Derby England, or Rolls- Royce Aero Engine Inc. 551 Fifth Ave New York, NY10017 USA. I don't know where you are situated. Cheers, HD.

Mark 1
12th Jun 2003, 22:00
Perhaps more relevant is another R-R book - "Aircraft Noise" by the late Mike Smith.

As this used to be my specialized subject, I'll offer a few more comments.

Even the very high bypass engines of late are still dominated by jet noise (i.e. exhaust mixing noise) at take off settings, but the fan noise isn't too far behind. The fan tips are typically at about 1.4 relative Mach number, so the shock waves from the tips propagate non-linearly to produce what is often called "buzz-saw" noise. Small differences in the fan, length of duct and acoustic lining of the duct can significantly alter the character of this noise. It is characterised as a set of tones at harmonics of the LP shaft frequency peaking in level at about 1 to 1.5 times the blade passing frequency (i.e. 1 - 2 KHz). Subsonically you only see harmonics of the blade passing frequency.

The jet noise is mainly characterised by the velocity of the hot exhaust and the diameter of the hot nozzle. Lots of techniques are used to encourage mixing with the surrounding bypass flow, especially with long cowl engines with buried hot nozzles.

I don't have much specific knowledge of the engines that you mention, but I hope this better explains it to you. Its a very big and expensive subject for the manufacturers.

747FOCAL
12th Jun 2003, 22:28
Hotdog,

I got you now. A lot of people confuse "engine" noise with "aircraft" noise. Different airplanes with the exact same engines will most likely not have the same certified noise values.

When I see magazines that say things like "this engine is "Stage 3" or "Chapter 3" I laugh. No engine is noise certified, it is the entire aircraft configuration that gets noise certification. Different aircraft perform differently as to thrust required at cutback, speed at the sideline point and thrust required on approach.

Engines with a larger bypass ratio will undoubtably be spinning a lot slower to produce the same or greater thrust as a lower bypass engine. This is the killer on approach. Cutback is mostly jet noise with very little fan component in the spectrum.

Things like exhaust mixers, acoustic liners and the newest tweek of the week.....performance robbing "chevrons" help in reducing the jet noise, well the liners in the exhaust only work for about a month and then they fill up with so much crap that they basically do nothing. Repairs to the liners also reduce their ability to reduce the source noise.

The best way to reduce noise on an already operational aircraft is to improve the performance of the aircraft(ie. improve L/D) Anything that can be done to reduce required thrust for any aircraft will have a large effect on the noise numbers. For the most part we in the noise world have hit a huge wall as far as technology that we can throw on aircraft to reduce noise. We used to get huge reductions and now we scrape for even tenths of a dB.

FYI - The A380 design increased fuel burn in cruise by 1% by tweaking the wing design towards lower thrust required on approach so as to make their intended QC catagory at Heathrow. Nice of the Bus to design a plane that is cheaper for the operator to operate but costs the rest of us in cash(increased ticket price) and environmentally.

:ok:

HotDog
12th Jun 2003, 22:52
Thank you Mark1 and Focal. I attended an Aviation Club luncheon in Sydney last week. The speaker was an Airbus rep. who gave a presentation on the A380. Extremely impressive and he touched on the noise characteristics as well, which concurs with your comments Focal.

Seloco
12th Jun 2003, 23:08
My thanks to all of you out there who have provided such erudite responses to my original posting. Whilst I still don't understand quite why there should be such a difference in noise between a GE90 and a Trent, I have certainly learnt a whole lot about the variables and complexities of large turbofan design.

I should point out incidentally that the observations which prompted my original post were based on experiencing 777s coming towards me in a RH turn soon after t/o from LHR 09R. Once the aircraft have passed (almost overhead) there is much less difference in noise between the two, since the exhause roar has taken over from frontal fan noise as the dominant element.

I must say that I am intrigued by the expression "buzz-saw" noise, since it certainly accurately describes that of the Trent (and its RB211 predecessors) but not so much the GE90. It is interesting that a new nacelle intake design is being considered to reduce this effect - I'm intrigued to know how this will work.

If I interpret the various detailed postings aright, this noise difference is unlikely to be due simply to 2 v. 3 spools, or fan speed per se, but rather a number of factors that combine to make a different noise profile. There was a suggestion that the number of fan blades could be relevant, but I have not been able to check on how many blades the Trent has versus the GE90 (actually I did try counting last night as I passed the BA engineering base at LHR last night but it was too dark - sad!!); anyone have any more information in this regard?

gas path
13th Jun 2003, 17:36
Seloco
To save you getting eye strain:p
GE90
Fan Diameter 123.0 inches
22 wide chord blades (carbon fibre with titanium leading and trailing edges)
100% N1 2,261 rpm
100% N2 9,332 rpm
Thrust 76k to 85k (90k for some operators)
New bigger fan and intake takes it to 115k (swept chord blades)

RR Trent
Fan Diameter 110.3 inches
26 wide chord blades (vacuum formed hollow titanium)
100% N1 3,300 rpm
100% N2 7,000 rpm
100% N3 10.611 rpm
Thrust 95k

BTW one can get the GE90 fan to growl quite nicely:E

I never understood why Boeing found it necessary with the Trent to flatten off the lower part of the nose cowl similar to the cfm 56 on the 737 whereas the '90 which is a bigger diameter has a 'round' nose cowl:confused:

Seloco
13th Jun 2003, 18:18
Thanks Gaspath for saving my already dodgy eyesight! Your information has, I suspect, also provided the answer to my original post. As I am not an engineer I may just have interpreted your numbers wrong, but assuming that N1 represents the speed of the inner, ie fan, shaft, then at 100% the Trent's fan is rotating a whole 50% faster than the fan on the GE90! If this is true then it is no surprise that it makes a very different noise. Since the difference in diameter is small (about 12%), the Trent's tip speed must be considerably greater than the GE's. I don't have the patience to do the calculation, but this alone, when converted to noise frequencies, would surely explain the difference? Over to the physicists/engineers......

PAXboy
14th Jun 2003, 00:42
One of the most interesting points about hearing aircraft 'noise' with the human ear, is how different people hear the same noise. Those that have to handle noise reports from the citizenry (airports and councils) know that:

1) One person sitting in their garden will hardly notice the aircraft going overhead - whilst their neighbour will be going mental and on the phone to the local group. One man's buzz saw is another's mental note that a RR, rather than a GE product is at work.

2) When you plot reports of intrusive a/c noise onto a map, a village/town/city will be speckled with reports, rather than banded. One might expect bands to build up as you get towards the airport or centre of flight path but the plots are almost always speckled.

It goes without saying that some of us in here [peeks out of anorak to look around for fellow beings] are less likely to complain about a/c noise than others. :8

Intruder
14th Jun 2003, 11:10
Using gas path's numbers, and assuming approximate STP at sea level, the GE fan tips are traveling at 1213 feet/second, or about 1.055 Mach, at a primary acoustic frequency of 829 Hz at 100% n1. The RR is 1588 ft/sec, or 1.38 Mach, at 1430 Hz.

I estimate Mach at 1150 ft/sec, and define the primary acoustic frequency as the number of times per second a fan blade tip passes the 12 o'clock (or any other) position.

Local air conditions will change the Mach number, but not the other data. If the engines are synchronized, the primary frequency should be a primary factor in perceived "tone."

Seloco
16th Jun 2003, 17:41
Thanks Intruder - some interesting numbers there! The GE90 fan blades are barely supersonic - simple maths says only about the last 3 inches spanwise have supersonic flow. With the Trent on the other hand the last 15 inches or so of blade span are supersonic at 100% power.

Now, presumably the blades act like any other aerofoil travelling supersonic, ie: the leading edge and mid-chord shockwaves make a sonic boom. So the Trent's 15 inches of each blade are each creating a sonic boom as they rotate. The effect must be similar to a football fan's rotating rattle with its sounding board contacting a toothed wheel. The "buzzsaw" noise is therefore generated by the booms occurring at the "past 12 o'clock" frequency of the rotating blades.

On this basis, it is small wonder that the Trent sounds so different!

Mark 1
16th Jun 2003, 21:36
Unfortunately its not that simple for noise due to shock-wave propagation. The velocity of a shock wave is non-linear i.e. it varies with amplitude. Tiny blade to blade differences get exaggerated to large variations by the time that the shocks have propagated to the end of the duct, whereupon linear sound propagation takes over.

At the end of the duct, instead of a uniform set of shocks passing at the blade passage frequency, you get a very unevenly spaced set of shocks (in both spacing and amplitude), sometimes one shockwave will overtake another and the pressure patterns merge. The whole pressure pattern thus only recurs after a complete rotation of the fan. That is why the 'buzz-saw' noise has a fundamental frequency at the shaft rpm rather than at the blade passing frequency for the subsonically generated component.

When calculating the Mach no. you must also vector sum the axial velocity component in the intake (~0.5Mach) acting at 90 degrees to the tangential component (tip speed).

747FOCAL
17th Jun 2003, 03:59
Current thinking on the cause of "Bad" Buzzsaw is variations in surface near the fan, thus the hardwalling in front of newer inlets. Something like a rivet head protruding, probes, etc. cause an uneven airflow to the fan thus "stressing" it more in different areas. :)

BahrainLad
17th Jun 2003, 08:33
I can remember a very good quote that can be used to amaze the general public on the topic of large jet engines........

"Consider an engine that is designed for one part to be subjected to supersonic pressure, another part to be bathed in hot gas, for 20 hours a day, for 25 years. And with proper maintenance, to not fail during that time."

Puts a Ford Focus into perspective.

Anthony Carn
18th Jun 2003, 06:16
Been some time since I was involved with high by-pass engines.

Bearing that in mind, I recall that a major cause of cold stream noise was the interaction between the air leaving the fan and impinging upon the stators behind it, in the duct. The then characteristic "buzz-saw" noise was a major consequence of this.

The Americans are more into variable stators (OK, mainly in the core engine compressors). Do these apply in the fan duct, giving better matching off design point ? Just a wild guess.

I repeat, it's been a while since I've been involved, but there may be something useful in my ramble.

Be interesting to know the answer.

Mark 1
18th Jun 2003, 19:03
Hi Anthony,

You are right in thinking that stator-rotor interaction is a major component of fan noise, but it doesn't contribute to 'buzz-saw' noise, only to blade passing frequency harmonics.

It is caused by the fan wake interacting with the vanes, this wake decreases with distance, and the potential field (bow wave) of the stator can modulate the angle of attack of the fan and thus cause a BPF tone. Both of these mechanisms reduce significantly with bigger spacing, and by having more than twice the number of stator vanes to fan blades, the interaction will be rapidly attenuated in the duct (cut-off condition).

This mechanism is more important at subsonic fan speeds as in the approach certification condition.

I don't think variable as opposed to fixed guide vanes have much influence on the noise though.

Truckmasters
20th Jun 2004, 07:28
A very interesting read.
What was the original purpose that wide chord fan blades were introduced. Do they have any other effects other than what Mark 1 has described. I'm guessing that they would increase the pressure (increased bypass ratio) Any information much appreciated.

casual observer
20th Jun 2004, 17:46
Flight Safety:

Your information is completely incorrect. The CFM56 uses the F101 core. The GE90 core was based on a NASA-funded E3 development program(E3, pronouced as E cube, stands for Energy Efficient Engine).

Throughout this thread, there is some good, semi-good, and bad information. I have worked on engine acoustic predictions on a peripheral role. However, it isn't really my field of expertise. So, I won't add more half-truths to this thread. Just to reiterate some of the points that have already been made. The science of engine noise is very complicated. Fan size, tip speed, bypass ratio, rotor count, stator count, blade-row spacing, acoustic liner, jet speed, jet mixing, cowl length, chevron, pylon, tail plug geometry, etc., are just some of the things than can affect the noise characteristics of an engine. Unlike in aerodynamic design, aeroacoustic design still heavily rely on semi-empirical-based tools instead of numerical simulations.

One more thing:

Anthony:

The stators behind the fan rotor, or exit/outlet guide vanes, are there to straighten the flow, as the name suggested. So, they are not variable. P&W in the 90s had developed an ADP (Advanced Ducted Prop) that used variable pitch vanes behind the ducted prop fan. These vanes also served the purpose of a thrust reverser (thus, reduce engine weight, or rather compensate for the much larger and heavier prop fans). I believe this was the only commercial-type engine that used variable vane in the fan stream.

Flight Safety
21st Jun 2004, 14:08
CO is correct about the source of the GE90 core, so my guess about the whistle noise was incorrect. My apologies to Seloco and all Ppruners. I knew when I was creating that last post that I was in a hurry and didn't have the time to do the research, and I strive to be accurate when posting on this website. :(

I thought I'd be bitten when I wrote that, and I was. Thanks CO for setting me straight.

PS - I decided to delete the entire post.

Torquelink
24th Jun 2004, 13:44
I read somewhere that recent versions of the Trent and other engines - including both 7E7 offerings are to have contra (or is that counter?) rotating spools. Is this driven by efficiency or noise considerations or both i.e. what exactly will this technology contribute?

747FOCAL
24th Jun 2004, 14:10
casual observer,

The GE90 stators are variable.


Torquelink,

Not really, but counter rotating fans are. :E

TURIN
24th Jun 2004, 14:40
747FOCAL

The GE90 stators are variable.

Er, methinks your are getting mixed up with the compressor stators. The coldstream outlet guide vanes (N1 fan stators) are fixed. They are there to 'straighten' the airflow. This means that the airflow is more axial than radial, increasing efficiency of the engine.

The compressor stators of most big fan engines (yes even the 3 spool Trent) are variable.

Back to the topic.

My understanding was that the buzz saw noise was caused by the fan blade's being supersonic. A320/CFM56 being a classic case.

chock2chock
24th Jun 2004, 15:33
those GE90s sound rather gimmicky more than anything else:yuk: with their delicate little EEEEEEEEEEEEEEEEEEEE. God I miss those DC9s. I love the fact that when youre sitting in the front of the 9s at takeoff its soo quiet. It is quite easy to forget that your plane is actually making quite a raucus down there :cool: wish i cpold say JT8D forever!

747FOCAL
24th Jun 2004, 16:44
TURIN,

Thats what I meant thanks for correcting me.

chock2chock,

I think you would change your mind about that DC-9 if you were sitting in the back. ;)