Turbofan/prop/UDF efficiency
Thread Starter
Join Date: Oct 2012
Location: Australia
Posts: 4
Likes: 0
Received 0 Likes
on
0 Posts
Turbofan/prop/UDF efficiency
Hi everyone,
Nice to meet you all. A couple of questions that I have had reading about aircraft propulsion on mighty Wikipedia have remained unanswered throughout the years. I finally decided that it would probably be a good idea to rock up here and ask for some professional opinion, otherwise I'd go insane thinking about it all day.
1. I understand that props have problems with wave drag at the tips as the tips approach the speed of sound. Why is it that the fan inside a turbofan does not experience the same problem? Is it because it has so many blades?
2. Carrying from the above question, I've also read that adding more blades to a prop makes it harder to balance, such that we don't usually see more than 8 blades on a single prop. Why is it that again, fans are granted some kind of exception from the rule?
3. An unducted fan or propfan looks almost exactly like a turboprop and works on pretty much the same principle. Neither Wikipedia nor Google have come up with a satisfactory answer on how they are different and why UDF/propfans can work at greater speeds (smaller blade diameter/more blades? But can't you just do that to a turboprop?)
Thanks in advance.
Nice to meet you all. A couple of questions that I have had reading about aircraft propulsion on mighty Wikipedia have remained unanswered throughout the years. I finally decided that it would probably be a good idea to rock up here and ask for some professional opinion, otherwise I'd go insane thinking about it all day.
1. I understand that props have problems with wave drag at the tips as the tips approach the speed of sound. Why is it that the fan inside a turbofan does not experience the same problem? Is it because it has so many blades?
2. Carrying from the above question, I've also read that adding more blades to a prop makes it harder to balance, such that we don't usually see more than 8 blades on a single prop. Why is it that again, fans are granted some kind of exception from the rule?
3. An unducted fan or propfan looks almost exactly like a turboprop and works on pretty much the same principle. Neither Wikipedia nor Google have come up with a satisfactory answer on how they are different and why UDF/propfans can work at greater speeds (smaller blade diameter/more blades? But can't you just do that to a turboprop?)
Thanks in advance.
Join Date: Jan 2012
Location: Ontario
Age: 74
Posts: 82
Likes: 0
Received 0 Likes
on
0 Posts
1. I understand that props have problems with wave drag at the tips as the tips approach the speed of sound. Why is it that the fan inside a turbofan does not experience the same problem? Is it because it has so many blades?
I think you are on the right track with the many blades, ie the prop blade tips are out there on their own, whereas the fan blades can help each other out with their shared passages between blades.
Fan blades (as well as the transonic compressor stages which you cannot see inside the engine) operate with blade relative tip speeds well beyond what propellers see so the same basic supersonic flow phenomena are there.
The wave drag equivalent term in transonic compressors is shock losses.
The reasons why fans can operate with rotor tip inlet relative mach numbers up around 1.5 these days is due in part to thin blade sections but also being able to control the supersonic diffusion within the blade passages in ways that minimise the losses, eg with multiple shock systems.
2. Carrying from the above question, I've also read that adding more blades to a prop makes it harder to balance, such that we don't usually see more than 8 blades on a single prop. Why is it that again, fans are granted some kind of exception from the rule?
There is no max blade number rule as regards balancing a fan. The fan is what it is based on many difficult design decisions and being able to balance it is a foregone conclusion, as long as you haven't forgotten to design in a rim to put the balance weights on.
Having said that, balancing it is not necessarily easy because it isn't balanced until it meets a required specification which could be very demanding. The effects of small amounts of rotor unbalance can easily find their way into passenger cabins (noise) especially from fuselage mounted engines.
I'll try and compose something for Q3.
PK
Last edited by peter kent; 2nd Nov 2012 at 02:09. Reason: clarification
Quite a few years ago I was fortunate to have an extensive tour of a turbine engine repair facility, not propjet.
Each turbine blade was weighed to some really fine degree (.001 gram?) and the spools were assembled in a particular pattern to ensure the best balance.
I don't remember if any further fine-tuning was done after final assembly, but was impressed by the cleanliness and precision of the whole procedure.
f
PS
Peter,
I don't think adding weights to a "rim" in a turbine is a normal procedure, I think that would be a bit crude.
f
Each turbine blade was weighed to some really fine degree (.001 gram?) and the spools were assembled in a particular pattern to ensure the best balance.
I don't remember if any further fine-tuning was done after final assembly, but was impressed by the cleanliness and precision of the whole procedure.
f
PS
Peter,
I don't think adding weights to a "rim" in a turbine is a normal procedure, I think that would be a bit crude.
f
Last edited by fleigle; 2nd Nov 2012 at 01:57. Reason: adding PS
Join Date: Jan 2012
Location: Ontario
Age: 74
Posts: 82
Likes: 0
Received 0 Likes
on
0 Posts
I don't think adding weights to a "rim" in a turbine is a normal procedure, I think that would be a bit crude.
You say it was many years ago, but today the allowable unbalance levels on rotors are so low that what you saw (weighing and distributing in best pattern) is today only the first step because it is no longer good enough to meet the low levels of unbalance required to prevent, for example, unacceptable noise levels in the aircraft. The next step is to actually measure the unbalance by spinning the rotor on a balancing machine. The machine then says what balancing weights are required in what positions on the rims.
So, the initial assembly, although done with the utmost precision and dedication, has to be fine-tuned with balance weights, since as is the general way of the world we are always demanding something today that's better than yesterday.
pk
Thread Starter
Join Date: Oct 2012
Location: Australia
Posts: 4
Likes: 0
Received 0 Likes
on
0 Posts
So it would appear that the inlet duct of a turbofan engine works a lot of magic in reducing shock losses, namely by slowing the flow to subsonic speeds? Ignoring problems with blade balance (since it seems we have come to the conclusion that we do have the technology to get it crazy precise) then, in the case of the UDF, why would we choose to have all the blades hanging outside the duct for what is basically a turboprop wanting to go fast?
I guess the better question is, why can't we use an existing turbofan design and just have all the air exiting the combustion core drive a conventional (ducted) fan?
I guess the better question is, why can't we use an existing turbofan design and just have all the air exiting the combustion core drive a conventional (ducted) fan?
Join Date: Sep 2002
Location: Pfffft
Posts: 218
Likes: 0
Received 0 Likes
on
0 Posts
Originally Posted by R1no
2. Carrying from the above question, I've also read that adding more blades to a prop makes it harder to balance, such that we don't usually see more than 8 blades on a single prop. Why is it that again, fans are granted some kind of exception from the rule?
The solidity of a propeller system is effectively a ratio of the frontal area of the propeller blades compared to the total propeller disc area. For example, if the propeller was simply a solid circle (and thus completely ‘blocked’ any airflow), it would have a solidity ratio of 1.
In general, as you increase the solidity ratio of a propeller, you can absorb more power from the engine (increasing the thrust you can produce). The trade-off is that you loose efficiency as the solidity ratio is increased beyond a certain point.
Anecdotally, as the tips of a propeller reach ~Mach 0.91, a large decrease in efficiency is observed. This can be remedied by increasing the width of the blades, but this can then have repercussions in other operating regimes. I believe the Open Rotor and UDF blades spin quite quickly (the tips therefore have a high mach number), but they’re also short and stubby (regaining efficiency).
All in all, propeller design is a tricky art where large compromises are made in the design – Typically you can optimise it for a specific condition (i.e. cruise flight) at the expense of other regimes.
The balance and noise problem is a lot more of a challenge when you consider
dynamic balance
You can start out with something as simple as "knife edge" balance where you place the fan shaft on a frictionless knife edge and watch which blade goes to the bottom. Only to find that it still shakes the bejeebers in a dynamic balance machine.
Then you find out that you have opposing equal masses but a foot apart or more in the axial direction. For that common occurrence in gas turbine rotors you need a means to balance out these forces with two balance planes after the fancy balance machine tells you where to put the weights. Thus the designer needs to add these balance planes to the rotor design (having a balance ring with rivet holes for these weights is common).
I think that race car drivers are faced with similar challenges on the engines as well as the wheels.
Then for fan engines with those fancy looking twists you find that the mass is never uniformly centered along a constant radial line so just a change in twist in a single blade can affect the imbalance felt. Thus you can have4e a bird strike and not even lose any metal but yet the fan will sing out that it's vibrating like hell.
I've heard of mechanics spending days across multiple flights trying to on-wing cure a fan vibration problem with balance weights when it ultimately turned out that one blade had untwisted slightly after a bird strike the week before. One can imagine how difficult it is to see an abnormal twist in one of these scimitar looking blades by looking in an inlet
dynamic balance
You can start out with something as simple as "knife edge" balance where you place the fan shaft on a frictionless knife edge and watch which blade goes to the bottom. Only to find that it still shakes the bejeebers in a dynamic balance machine.
Then you find out that you have opposing equal masses but a foot apart or more in the axial direction. For that common occurrence in gas turbine rotors you need a means to balance out these forces with two balance planes after the fancy balance machine tells you where to put the weights. Thus the designer needs to add these balance planes to the rotor design (having a balance ring with rivet holes for these weights is common).
I think that race car drivers are faced with similar challenges on the engines as well as the wheels.
Then for fan engines with those fancy looking twists you find that the mass is never uniformly centered along a constant radial line so just a change in twist in a single blade can affect the imbalance felt. Thus you can have4e a bird strike and not even lose any metal but yet the fan will sing out that it's vibrating like hell.
I've heard of mechanics spending days across multiple flights trying to on-wing cure a fan vibration problem with balance weights when it ultimately turned out that one blade had untwisted slightly after a bird strike the week before. One can imagine how difficult it is to see an abnormal twist in one of these scimitar looking blades by looking in an inlet
Join Date: Jan 2012
Location: Ontario
Age: 74
Posts: 82
Likes: 0
Received 0 Likes
on
0 Posts
addition of balance weights
Anywhere I can see a pic of balance weight addition???
I just can't get my head around the physical "how its done".
I just can't get my head around the physical "how its done".
Patent US6893222 - Turbine balancing - Google Patents
Hope that helps.
pk
Join Date: Jan 2012
Location: Ontario
Age: 74
Posts: 82
Likes: 0
Received 0 Likes
on
0 Posts
So it would appear that the inlet duct of a turbofan engine works a lot of magic in reducing shock losses, namely by slowing the flow to subsonic speeds?
Incidentally, the inlet doesn't determine the speed of the air at the engine face. The engine takes what it needs and slows down or speeds up the air as appropriate for a given thrust setting and flight velocity.
Shock losses are caused by the fan design, ie tip speed, and reduced by the fan design, ie thin blade sections and profiles to control the diffusion to subsonic speeds within the blade passages.
why can't we use an existing turbofan design and just have all the air exiting the combustion core drive a conventional (ducted) fan?
It's to allow flight speeds of about Mach 0.8 with better SFC than today.
These concepts attack the propulsive efficiency part of the SFC.
This has been done continually since the advent of the turbojet by increasing BPR until today the ever increasing weight and drag of the big fan engines intake and cowling are negating the benefits of jet speed reduction.
That's why you don't want a ducted fan.
Since propeller efficiency is very good due to its high BPR up to about M0.65 the focus is on how to redesign the prop for M0.8 with low compressibility losses and much higher power loading (because you have to go faster). The first requirement leads to thin, swept blades, and the second means maybe twice as many blades of smaller dia and spinning faster (which reduces the reduction gearbox size at the prop moves towards the power turbine speed).
This achieves the goal stated above.
The propfan is a turboprop with an advanced propeller, with todays examples used on the A400M and AN-70.
why would we choose to have all the blades hanging outside the duct for what is basically a turboprop wanting to go fast?
Join Date: Feb 2005
Location: flyover country USA
Age: 82
Posts: 4,579
Likes: 0
Received 0 Likes
on
0 Posts
While I don't recommend Wikipedia as a competent text on the subject, this article provides a good starting point for both the technology and history of the unducted fan.
Thread Starter
Join Date: Oct 2012
Location: Australia
Posts: 4
Likes: 0
Received 0 Likes
on
0 Posts
Thanks everyone for the insight.
So Peter, are you saying that there is basically no difference between a turboprop and a propfan then? That's what I suspected all along. Because the only difference I see between a modern turboprop and the GE36 is that GE decided to put the props at the back and have the high velocity exhaust drive them directly via a rear turbine, while modern turboprops use a reduction gearbox. The Progress D27 is basically a contra rotating turboprop?? But there has got to be a reason why they gave these things a different name 
. I thought MAYBE the propfans/UDF are "fans" and the turboprops are "props"? (again, no satisfactory source on the internet that explains where you draw the line between a prop and a fan).
It appears we would agree that:
The propfan is a turboprop with an advanced propeller, with todays examples used on the A400M and AN-70.
. I thought MAYBE the propfans/UDF are "fans" and the turboprops are "props"? (again, no satisfactory source on the internet that explains where you draw the line between a prop and a fan).It appears we would agree that:
- Balancing a prop with many blades is not really an issue
- We have become so good at designing fancy swept prop blades that the cruise speeds they can achieve at a reasonable SFC are approaching those of subsonic turbofans, and the A400M is a living example.
- Ducting is not worth the weight and drag at high BPR because the fan will be really large.
Join Date: Jan 2012
Location: Ontario
Age: 74
Posts: 82
Likes: 0
Received 0 Likes
on
0 Posts
But there has got to be a reason why they gave these things a different name 
. I thought MAYBE the propfans/UDF are "fans" and the turboprops are "props"? (again, no satisfactory source on the internet that explains where you draw the line between a prop and a fan).
. I thought MAYBE the propfans/UDF are "fans" and the turboprops are "props"? (again, no satisfactory source on the internet that explains where you draw the line between a prop and a fan).
An Airbus official back in 1985, when high fuel prices had raised the hysteria level to such heights that some of these machines were actually made, was attributed the description 'a load of bananas whirling around'.
We have become so good at designing fancy swept prop blades that the cruise speeds they can achieve at a reasonable SFC are approaching those of subsonic turbofans, and the A400M is a living example.
30 years have gone by since the last foray which says something.
Join Date: Sep 2006
Location: Los Angeles, USA
Age: 52
Posts: 1,631
Likes: 0
Received 0 Likes
on
0 Posts
In layman's terms:
Low to medium speed efficiency is greatly improved by enlarging the fan/prop. For instance, a very good example is a helicopter - the bigger the rotor diameter, the less HP it needs to fly. Same is true for aircraft. This is exactly the reason why all the new "fuel efficient" turbofans get bigger and bigger in diameter (look at the GENx and the Trent 1000's). The core is just as inefficient as it's always been (it's still a simple turbojet in the middle), but the bigger the fan is at the front, the more of that can be turned into thrust for the same amount of fuel. Now, obviously, at too large a prop/fan diameter the penalties of weight become a problem. But as mentioned - there is no real difference in performance between a turboprop and a fanjet of the same size. And the two will eventually grow even closer than they are today. Just witness the fanjet blade count steadily going down for every new generation...
The notion that props are inherently slower than jets is also false. The Piaggio Avanti flies a lot faster than a Citation burning 40% less fuel. Why? Because it moves more air.
Low to medium speed efficiency is greatly improved by enlarging the fan/prop. For instance, a very good example is a helicopter - the bigger the rotor diameter, the less HP it needs to fly. Same is true for aircraft. This is exactly the reason why all the new "fuel efficient" turbofans get bigger and bigger in diameter (look at the GENx and the Trent 1000's). The core is just as inefficient as it's always been (it's still a simple turbojet in the middle), but the bigger the fan is at the front, the more of that can be turned into thrust for the same amount of fuel. Now, obviously, at too large a prop/fan diameter the penalties of weight become a problem. But as mentioned - there is no real difference in performance between a turboprop and a fanjet of the same size. And the two will eventually grow even closer than they are today. Just witness the fanjet blade count steadily going down for every new generation...
The notion that props are inherently slower than jets is also false. The Piaggio Avanti flies a lot faster than a Citation burning 40% less fuel. Why? Because it moves more air.
Last edited by AdamFrisch; 4th Nov 2012 at 21:18.
Join Date: Jan 2012
Location: Ontario
Age: 74
Posts: 82
Likes: 0
Received 0 Likes
on
0 Posts
misunderstanding, see PS
The core is just as inefficient as it's always been (it's still a simple turbojet in the middle
The 2 efficiences go hand in hand as continual improvements in both have brought about an ever declining sfc since the day after the first jet entered service.
There's a lot of intense stuff going on in a GE90-style gas gen in terms of temperature and pressure at levels far beyond those in an inefficient machine of a bygone era. Those levels are not for nought. They are required to meet the power absorption demanded by that size of fan.
The pressure ratio and turbine temperature are only 2 of the things that go into defining the gas gen efficiency. So its improvements in all those factors which mean an old turbojet core just doesn't get a look in any more for any application.
PS On reflection I think I have just got Adam's meaning, ie not that the core is inefficient, but that todays core, if used in a turbojet at subsonic speeds would have an inefficient propulsive eff despite a much-improved core eff. Silly me and apologies to Adam.
I guess my post really deserves deleting but I'll leave it in case anyone else misread the original.
Last edited by peter kent; 7th Nov 2012 at 01:02. Reason: PS
Join Date: Feb 2005
Location: flyover country USA
Age: 82
Posts: 4,579
Likes: 0
Received 0 Likes
on
0 Posts
The GE36 UDF set a record for TSFC - but with substantial risks in terms of the counter-rotating LPT, variable-pitch fan blades, and noise issues. By the time the engine was ready to be marketed, fuel prices had retreated and airlines were reluctant to dip their toes into that risk pool. Better the devil you know...
Join Date: May 2003
Posts: 409
Likes: 0
Received 0 Likes
on
0 Posts
Barit wrote:
I would be really REALLY interested in seeing the details. Barit, do you have any specific numbers? Just how did the UDF compare with the present turbofans, and turboprops?
Hawk
The GE36 UDF set a record for TSFC
Hawk
Thread Starter
Join Date: Oct 2012
Location: Australia
Posts: 4
Likes: 0
Received 0 Likes
on
0 Posts
was attributed the description 'a load of bananas whirling around'.
but with substantial risks in terms of the counter-rotating LPT, variable-pitch fan blades
Join Date: Feb 2005
Location: flyover country USA
Age: 82
Posts: 4,579
Likes: 0
Received 0 Likes
on
0 Posts
but with substantial risks in terms of the counter-rotating LPT, variable-pitch fan blades
But the blades are each attached to this outer case system, with a complex flex-shaft system supposedly keeping them all synchronized. Every flex shaft system I've worked on has left me unimpressed regarding airline-standard reliability; and unless some sturdier wizardry is on the horizon, I doubt it could pass muster.
TSFC at cruise - IIRC - was under 0.25.
