Ducted fan efficiency if engine is taken out of the equation (pure electric case)

HomeBuiltAirplanes.com

Help Support HomeBuiltAirplanes.com:

karoliina.t.salminen

Well-Known Member
Joined
Oct 7, 2008
Messages
407
Location
Finland
Hi,

There has been some back and forth discussion about ducted fans in general in this forum and elsewhere. Usually the limitations come from the engine that turns the fan. In general it has been concluded that the efficiency of a ducted fan is lower than a prop, except I have seen Orion mentioning some sea plane he has been working with and that analysis concluded that the fans work good up to 200 kts, but I can't find this post anywhere anymore. Well there could be exceptions like the mentioned, but at least in RC aircraft world, the fans tend to produce relatively low thrust for the power they require. If someone would have link to the Orions post, could you please paste it here.

However, if we consider ducted fan from scratch (not placing a existing prop in a duct nor turning that with a combustion engine), would it be possible to design a fan that produces rivaling results to a prop and if so how it should be done? And also why then the EDFs on RC planes tend to have very low efficiency (not only is static thrust lower than on a prop, but it tends to be that also the dynamic thrust seems lower - I do not have accurate measured data, but I have flown one RC plane with both prop and fan and everything else equal, by eyeballing on what I see, I can tell it is going faster, longer and stronger with the prop hands down compared to the fan which was run with same battery and roughly similar amount of current and power)?

The fans interest me especially because it would be interesting to place these inside the boundary layer (partially buried engines near the trailing edge) of a wing and instead of a big prop, propel the aircraft with distributed propulsion and then solving the problem of how to implement suction for active boundary layer control at the same time. I am not aware though how the performance of electric ducted fan would be affected by non-clean air ingested from inside the boundary layer and what that could cause for the efficiency. Luckily EDF is not a jet engine and all the problems associated with jets in boundary layer may not apply. Also it would be interesting to know if it would be possible to combat the potential loss of efficiency by ingesting the boundary layer and thus reducing the drag somewhat this way. Also this kind of distributed trailing edge propulsion could allow both yaw and roll control of the aircraft with adjusting the thrust of the fans through the wing, a bit similar automated control could be developed than the quadrocopters have and this kind of configuration could allow also rather interesting maneuvers for the aircraft potentially.

I have been trying to find information about fan design, but haven't found anything conclusive because either these are experiments by homebuilders without so much science behind (fan not optimal) or they are sort of replicas of jet engines (cold jets) (and again fan not optimal for low speed flight).

I was thinking about a quantity of ballbark 10-20 fans on trailing edge (not talking about single or twin or such relatively conventional configuration), this makes the individual fan side relatively small. Would someone have some facts and opinions regarding this kind of crazy idea. I was thinking this kind of fan configuration would especially fit on a flying wing as these fans could cover the whole trailing edge without interruption with the fuselage as there would be none. How the fans should be designed to meet these goals (to be efficient (not lose too much energy for nothing useful), be optimized for low speed (not transsonic like most studies I can find, rather the GA aircraft speed range <=200 kts and would produce meaningful static thrust on takeoff allowing the plane to get off the ground too).

And there is one more thing: I have seen some papers about using ducted fan to reduce noise. THen why are the ducted fans on EDF jets very extremely noisy then? Like super loud hairdryers in the air. What makes the EDF of a EDF jet both inefficient and loud whereas there are claims that it could be possible to have noise reductions done by using a fan? What could be more cool than a flying wing? A quiet flying wing that makes almost no noise at all.

Best Regards,
Karoliina
 

addaon

Well-Known Member
Joined
Feb 24, 2008
Messages
1,686
Location
San Jose, CA
Efficiency limitations on ducted fans are entirely related to the fan itself, not the engine. Engines and electric motors both have efficiencies that are generally independent of speed (assuming they're running at their design speed), while weight of both tends to favor higher speeds than we customarily use for propellers.
 

Aircar

Banned
Joined
Feb 20, 2010
Messages
3,567
Location
Melbourne Australia
One embodiment that integrates the ducted fan and flying wing is the multiple ducts AS wing --think of the Aerovironment pathfinder or Helios with ducted fans spaced out along the span -- even a spanwise series of Custer channel wings could integrate the advantages of enhanced lift and hence reduced wing area and weight (the small strut needed to support an electric motor at the centre of the 'half duct' together with another tangent to the lower ducts would make a sort of truss with the curved ducts as diagonals -- with the tiny electric motors the duct blockage and profile drag of the classic IC engine Custer would be avoided.

Unless an aircraft is of huge size it looks to be impossible to integrate ducted propulsion into a flying wing (as some of the blended wing body mega airliners do ) --various schemes to add 'flappers' or 'translating' propelling blades running along the wing trailing edge --and using the re energizing of the wing wake to efficiently create thrust ,have been proposed as 'brainstorming' concepts and searching for patents will turn up many schemes (Junkers was one of the first with his propelling wing ) Ducted fans or shrouded propellers seem best suited to be tail surfaces as well as exemplified by the MSU Marvel so defraying the drag penalty of the duct -- a Britten Norman Islander was flown with wing mounted Dowty ducted props and looked awful even if it was quiet --maybe some homebuilder will build a scaled down 767 or other airliner with large turbofans using electric motors (as RC modellers already do --it's only a matter of time )
 

hogheadv2

Well-Known Member
Joined
Feb 7, 2010
Messages
127
Location
Booger finger of Michigan
I think of Duct Fans in much the same way as venturi's in "Old Time" now carburators. With the inside as an airfoil, many carbs had a Booster jet [very small venturi] above the main. This was used for high speed fuel delivery. Some well pumps use this to maximize water volume.
My pondering has been to put a Savage Ducted Fan just prior to a second venturi. The flow/speed will increase I am sure, but enough gain for the added weight, drag, cost, structure???
 

highspeed

Well-Known Member
Joined
Apr 10, 2011
Messages
275
Location
Spring, TX
I think it is important to point out a distinction that Orion has mentioned before. There is a difference between a ducted fan and a shrouded propeller. The difference lying in the aspect ratio of the duct. IIRC, a duct length of .5 propeller radius or less is considered a shrouded propeller. Longer than that and the drag of the duct overcomes the benefits you would expect..
 

kenkad

Well-Known Member
Joined
Nov 25, 2011
Messages
53
Location
union grove, al 35175
Look at my Jan 2 thread, there were no replies. I recommend that you look through the paper referenced in that thread. I will be doing this testbed in the near future. pronbably when the Fall Semester starts at the local school. The length of the duct is important, but this has be looked at relative to having two counter rotating four bladed propellers. The separation between these two props is important. The referenced paper tells you that there is s definite performance increase. There is more literature, but this literature has not been and may never be released because it was done for the US Army UAV program and may be classified. I do not recommend putting the ducted fans on the trailing edge of a wing. AoA is going to have a serious effect. They need to be out in the open.

When we do the testbed analysis, many variables will be sorted out. The testbed uses 16 inch propellers. The first four bladed prop may be running at 10-11kRPM and the back propeller ar 11.5-12.5kRPM. The electric motors are not in the center of the duct. There is only one electric motor and it drives both ducted fans with timing belts. I wish there was more information, but there simply is not at the moment. I am also trying to get the US Army to help do a CFD analysis (requires a rotating mesh geometry capable CFD). Good luck on your efforts.
kenkad
 

Voyeurger

Well-Known Member
Joined
Oct 9, 2010
Messages
612
Location
Northern Phoenix, Arizona, U.S.A.
Hello Karolina,
You should also check in with a member here that goes by "Lucrum". He has done a whole bunch of research on ducted fans and is building a plane around one.
Regards,
Gary
 

DeepStall

Well-Known Member
Joined
Dec 26, 2010
Messages
136
Location
Colorado
Ducted fans introduce many complexities not present in normal propellers. Carefully designed and built, they can be as efficient, if not better, than an open tip propeller. The main challenges are:
  • Maintaining tip gap to better than 0.5% r/R is crucial to efficiency and low noise. This leads to requirements for stiff ducts and shaft mounts.
  • But stiff ducts and shaft mounts tend to be heavy...
  • Interference between components in the duct are a major noise source. But those internal parts are needed for stiffness...
  • The other major noise source is from the blades alone, and scales as roughly the 5th power of tip speed. This is likely a major reason why the R/C units are so loud
  • The duct geometry must be tailored for a particular flight condition. Key parameters are the ratios of throat to inlet and exit areas, and the inlet geometry. Suction on the leading edge of the duct can produce a significant portion of the thrust; conversely spill drag from a lip operating off design can eliminate any benefits.

There is a paper with a little more of the math here:
Designing aircraft

A GPL ducted fan design code by Prof. Drela at MIT is available here for numerical experiments:
DFDC

Regarding your concept of boundary layer ingesting distributed propulsion, I would be concerned about:
  • Small diameter fans operating at low Reynolds numbers, with resulting high blade profile drag
  • The unsteady boundary layer being a noise source, and leading to excessive vibration if sufficiently turbulent, as in a stalled flight condition.
 
Joined
Nov 28, 2009
Messages
11
Location
Sacramento CA
I would add to what has been stated here already, that it is important to notice the difference between work that has been done with DUCTED FANS (meaning High Bypass compressor Sections turned by Auto Cycle) vs the Shrouded Props or Shrouded Fans.

I have seen many people attempt to call what they were considering a Ducted fan, when what they meant was a Ducted prop.... The Orion is a misinterpretation based on this point.


Next, Using High Bypass compressor sections with High Output (HP & RPM) & Light weight ratio (lbs) Engines can actually be a winning combination for a Jet like thrust generator..... HOWEVER-- you will be limited by your forward movement at Higher speeds due to FAN SPILL OVER.... this is the point at which the Air Entering the Duct cannot be processed by the ROTOR and begins to PILE UP against the ROTOR causing Drag---- at this point, the rotor become a barn door limiting your forward movement.


There will not be a good solution for a Fast flying Scale sized DUCTED FAN EXPERIMENTAL PLANE---- UNTIL SOMEONE DESIGNS THE PROPER VARIABLE INLET to the Duct -----------that can allow less air MASS to enter the INLET when at faster speeds.


I refer to this issue as ROTOR PROCESSING SPEED----- the faster we move forward-- The MORE AIR MASS works against your Thrust ROTOR.

I myself am looking at a few possibilities to solve this issue------ But it is all Models in Cyberspace waiting to go to the FAB for prototype.

Someday------- I will have it put together to run test...

Cheers
 

orion

Well-Known Member
Joined
Mar 2, 2003
Messages
5,800
Location
Western Washington
Hi,

There has been some back and forth discussion about ducted fans in general in this forum and elsewhere. Usually the limitations come from the engine that turns the fan. In general it has been concluded that the efficiency of a ducted fan is lower than a prop, except I have seen Orion mentioning some sea plane he has been working with and that analysis concluded that the fans work good up to 200 kts, but I can't find this post anywhere anymore. Well there could be exceptions like the mentioned, but at least in RC aircraft world, the fans tend to produce relatively low thrust for the power they require. If someone would have link to the Orions post, could you please paste it here.
We really have to differentiate between what is conventionally called a ducted fan and our work with the Privateeer, which is essentially a prop in a shroud. Historically, the primary purpose of a ducted fan is to allow one to configure a near jet like airplane yet power it with a conventional engine that most commonly drives a small diameter multiple bladed fan. We've discussed this here numerous times so we really don't need to rehash old threads, and some of the subject matter was covered above.

The system that the Privateer uses takes a relatively common prop and places a shroud around the blades so as to seal the tips and reduce the losses seen at the low end of the speed spectrum in more convectional configurations. This allows the prop to operate more efficiently, especially at low speeds where prop efficiency tends to be very low. Then there is also a second function of this structure, which essentially takes the energized air and allows it to act on the shroud creating a net thrust benefit, thus increasing the static and low end performance of the airframe. This of course is nothing new and has been verified in numerous tests multiple times. I don't have the references (in case someone asks) but the folks that designed the systems have a high degree of confidence in the results.

The unique aspect of the design is that it is optimized for both phenomena: Increasing prop efficiency and generating additional thrust. The design on which this system is based was developed in line with the US Army's and the US Marine Corps' goal of increasing the performance of the LACV line of utility vessels - in those applications the new system allowed for an increase in static thrust in excess of 100%, allowing them for the first time to carry their actual rated weight.

The Privateer design changes the configuration a bit from that work in that it also works at maintaing a thrust benefit to a higher airspeed. Historically shrouded props were thought to be useful up to speeds in the neighborhood of 120 to 150 mph. After that point it was common for the benefits to disappear due to the added wetted area of the shroud. The trick then was to design the shroud so that it allowed an increase in propulsive efficiency to a higher value, which in our case is just in excess of 220 mph.

But again, this work cannot be equated to ducted fans, especially the grossly overpowered models as seen in model aircraft. That's a whole another animal and is in no way equivalent to this work here.
 

Voyeurger

Well-Known Member
Joined
Oct 9, 2010
Messages
612
Location
Northern Phoenix, Arizona, U.S.A.
Hey Orion,
Let me see if I have this correctly. The shroud is essentially an airfoil made round. An enclosure so precise that the prop tips are not exposed to free air? Do they run in a slotted track of some sort?
BTW, when's the proto going to get airborne?
As always,
Thanks
Gary
 

kenkad

Well-Known Member
Joined
Nov 25, 2011
Messages
53
Location
union grove, al 35175
Anyone interested in this topic should put this PDF into their database. It covers a very great deal of the previous work with a decent discussion of the results.

drum.lib.umd.edu/bitstream/1903/8752/1/umi-umd-5771.pdf

kenkad
 
2
Top