EDF Ducting and EDF Efficiency

[Elbee]

I like to gain some knowledge of EDF fans and ducting. My FW A-10 has no ductwork to speak of and of course, my F-4 does. I will be purchasing the FA-18 when offered. The differences in my observation are A-10 has no ducts so efficient is as good as is possible, F-4-has bifurcated ducting so a smooth duct interior would be relevant, FA-18 presumably has bifurcated ducting and an additional vent (cheater) if I am reading the forum correctly. So my question: Is there a formula used in design to determine duct opening sizes and lengths related to fan size? Does anyone mod these as a rule? I get it, in that, if it ain't broke don't fix it, though I would like to know how it all works and if there are things I can do to increase efficiency. Inquisitively, LB

[Alpha]

Hi Elbee! Great question, and one that we could discuss ad nauseum. I'll try to summarize as best I can.

First, it's important to establish a few parameters:

1) We're dealing with models here. Our models scale down, but air doesn't.
2) It's easiest to envision air behaving like a fluid.
3) As with a fluid, turbulence (think of a white water rapids) is generally slower than laminar flow.
4) As with a fluid, viscosity matters (water flows more readily than peanut butter)
5) Temperature matters too, even if on a smaller scale in RC (warm peanut butter flows more readily than cold peanut butter)
6) Velocity and viscosity directly influence how easily the air flows into, through, and out of a duct. We could say the same for wings, fuselages, etc, but in the context of your question let's just focus on air interacting with a model aircraft's duct and the EDF inside that duct.


With those points established, we can state that the more efficiently airflow can enter and transit through your model's 'ducting and then be accelerated out of it, the better. You can also begin to envision what factors negatively impact that entrance, transit, and exit. These negative factors include:

1) Anything that interrupts the airflow's external, pre-entrance state. This can be a physical barrier, or even the surface tension of the ducting material (we'll talk about boundary layers later).
2) Anything that interrupts or in this case alters the airflow's external, pre-entrance state such that the air is compressed transiting through a smaller area and then expanded transiting through a larger area. These transitions cause turbulence which slows the air down. Imagine a straight straw of consistent inner diameter, and then imagine a straw with an alternating inner diameter from large to small to large to small. It's easy to imagine that a liquid will pass through the first straw more easily than the second. Resistance begets turbulence which begets slower exiting airspeed for a given amount of watts consumed in an EDF system.
3) Anything that constricts the air molecules into a more tight space at a rate slower than the air can exit the duct. For example, trying to sneeze throw a coffee stirring straw. This creates pressure between the EDF and the exhaust outlet because the fan can't get the air out fast enough.
4) Anything that allows more air to enter in front of the EDF impeller than can be expelled behind the EDF impeller (imagine a 500mm intake on your 80mm A-10). This creates drag as turbulent air spins in front of the fan because the fan can't ingest it fast enough.

With these negative factors identified, you probably can fill in the blanks for positive factors that avoid these negative factors, such as:

1) Smooth inlets
2) Smooth ducting surface
3) Consistent total area of the duct
4) Minimal if any changes to the direction of the airflow within the duct
5) Sensible amount of airflow brought in and sensible sized outlet to let that air out with the least resistance and with the targeted amount of velocity for a given watts consumed.

This 5th factor above is the basis for what you might see often on RC EDF forum discussions regarding "FSA", or "Fan Swept Area". This is the area of the inner diameter of the EDF's shroud, minus the area of the EDF's impeller's nose cose. The remaining donut is the area swept by the impeller's blades, hence the term Fan Swept Area. I'm generalizing here, but most EDFs fall around 110% FSA for their intakes and ~90% for their outlets. With what we know from the three checklists I've given, you can see how the intake FSA % and the outlet FSA % affect an EDF's performance.

1) If the intake FSA is too much larger than the EDF's FSA, then we can expect that too much air will get stuck in front of the EDF, at a rate higher than the EDF can suck it through. Bad.
2) Conversely, if the intake FSA is too small, then the EDF will "starve", because it won't be fed enough air that it wants to be ingesting at that given RPM. Bad.
3) If the outlet FSA is too much larger than the EDF's FSA, then the air exits at a diffuse pressure (imagine trying to out a candle with your mouth wide open versus with your mouth forming a whistling shape). Bad.
4) Conversely, If the outlet FSA is too small, then the EDF cannot expel the air fast enough and your amps will spike. Bad.

As you can see, there is the need to balance inflow and outflow for optimal performance. This is one of the reasons why many modern full size jets utilize variable inlets and outlets, to adjust the air coming in (which is amounts to drag at higher speeds), and the air exiting out. The challenge in designing an electric EDF model aircraft is that

1) We cannot vary outlets, let alone inlets, cost effectively.
2) The advantage of doing so would be outweighed by the components, and rendered unnecessary considering the finer aerodynamic points beyond this discussion
3) The pitch of an EDF's blades are fixed

Accordingly, the only variable we really have to play with are sensible, optimized ducting from the designer, and the RPM at which the EDF's motor can be spun. This informs a motor's KV, its amp draw, input voltage, which in turn affects the model's weight, wing loading, and operating cost. But at different stages of a model aircraft's flight, the EDF will benefit from higher inlet FSA % (more air to grab and push behind it to accelerate faster from a stop) and benefit from a lower inlet FSA % (less drag to overcome since the EDF is already ingesting as much as it can expel at maximum throttle).

In summation, I can say that this is just the surface of what makes good EDF design,,, well, good. It isn't about any single "formula", but rather the application of experience with many factors. Balancing them will optimize an aircraft's ducting for a certain "sweet spot", but there will always be trade-offs. I could wander into some of these similar topics such as how blade count/pitch/design affects static thrust versus dynamic thrust, or why much of the "whoosh" so many people love to hear is actually turbulence/inefficiency.... perhaps for another discussion.

Suffice to say, our team understands these intricacies, which is why our EDFs fly as they do. Gone are the days where every EDF owner should expect to have to rebuild their model's ducting to get it to fly right. Your F-4 was, aerodynamically, a triumph, if I do say so myself. The hurdles we went through to get that bulky bird to fly as nicely as it does were largely exercises in what we've discussed here. Plus the wings.

As I said, another discussion.

:)

[Elbee]

Alpha, Thank you for taking the time to reply. I am familiar with some fluid dynamic principles and what you've written affirms what I understand. I would not change anything you and your group have designed, I am inquisitive, I like to tinker (carved a prop or two in my day, designed and hot wired shaped foam wings just because) and just like to know what makes things tick. It is the F-4 that got me to thinking about this again. I owned the JHH version and was intrigued by the propulsion design and not in my wildest imagination could I have thought about the offerings you all bring to the table today. I thank you for that. Yours, LB

[rifleman_btx]

How about when changing the shape of the ducting from circular to rectangular. Would that also lead to turbulence? Say for instance the F22

[air-jon]

Great explanation Alpha! Explaining a ducted fan is about as easy as explaining cold fusion..

Freewing and yourself have really stepped it up lately with the ducting design and I appreciate that so much!

I will just throw in what I have learned reading so of the scholastic papers on ducted fans available on the interweb.

-There are many different kinds of "ducted fans". electric powered, fuel powered, nuclear stem powered, etc...

-The most efficient ducted fans are always ones with short shrouds and large diameter with nicely airfoiled inlets (like the A-10 or airliners). The most efficient ducted fans also have a large diameter constrained only by desired efflux speed, weight and powerplant. A decent percentage of the Ducted fan's thrust at lower speeds comes from the air passing over these proper intake lips. if you don't have these proper intake lips, you loose that part of the ducted fan thrust puzzle and what is known as "pressure recovery" which I still don't fully understand.

-The properties of a properly designed large ducted fan like this are so good, they are more efficient than a propeller at any altitude at subsonic speeds...even with the extra weight of the nacelle. This is why airliners use ducted fans.

-For most fighter jet scale models, we can't use an ideal ducted fan. instead we have many compromises and these installations are referred to as "deeply ducted fans” The fans are restricted in diameter by the size of the fuselage “packaging restrictions”, intake size, aesthetics, and also ultimately the maximum size of a model the manufacturer wishes to produce. it is difficult to fit a fan that produces enough thrust for a high-performance model/ Many early supersonic jets have small inlets with sharp edges…worst case for turbulence forming near the inlet on a ducted fan. Advance in aerodynamics and propulsion gave us newer fighters that featured slightly larger ducting (because they featured low-bypass turbo-fans and needed more air) and this helps a little with the F-18 for instance with slightly rounded and larger intakes than say a F-101 Voodoo. even a small bit of rounding makes a difference in ducting performance, but it also has a pretty big impact on aesthetics that many would not accept.

-Cheater intakes (holes) are sometimes a necessary evil for scale models with deeply ducted fans and small intakes. These cheater intakes allow proper air volume into the fan, but have drawbacks such as inducing turbulence into the fans air supply and also creating some off-thrust line forces. All cheaters are not created equal. A poorly designed cheater can be too large, or cause a great deal of turbulence near the fan and reduce potential performance. A properly designed cheater intake can promote laminar flow and be sized correctly as to maximize the air coming in through the primary intakes and maximize pressure recovery, and minimize moving air-mass in directions you don’t want to (upward motion of air through the cheater intake causes a down-ward reaction of the jet per Newton). All cheater intakes unfortunately increase the chance of FOD. It all a bunch of trade-offs and requires a lot of time to maximize.
With all this considered, it is quite impressive how good the EDFs are these days considering the price point, because a lot of thought, designing, and testing goes into these models to get what we have available today at the click of a mouse button.

[fhhuber]

A bare fan is less efficient than one on a well designed duct.

Length of duct is not as important as getting the inside smooth and having the cross section area change slowly. Abrupt changes are bad.
Taper of a common styro drink cup is about as rapid as you want to try.
If length does not allow the styro cup taper to get the outlet down to 85% FSA you're probably better off with the wider outlet.

Inlet having a funnel effect and having a rounded leading edge is better then a straight tube.

Outlet being 85% of FSA (Fan Swept Area) is important for best results.
Pi * R^2 (Radius of fan) - Pi * R^2 (Radius of fan spinner) = FSA.
SQRT((.85 * FSA) / Pi) = desired R of outlet
For bifurcated, SQRT((0.5 * .85 * FSA) / Pi) = R of each outlet

I'll assume the ducting molded into the EPO foam is designed for the OEM "default" fan installation. Different fans of the same nominal diameter have different FSA.
Its easier to put in a cone to squeeze down an outlet that is too large than it is to open the outlet if its too small.

The most common issue I find with foam EDF models is inadequate inlet until the aircraft has built up speed to give a "ram air inlet" effect. (really bad lack of inlet area by some other brands)

I haven't done the math on the Freewing F-14 but I feel it needs some more cheater hole area. Once its up to speed it desn't need it. This one will be easy to alter just by changing the cheater hole grates.

Its easier to cut more opening than to close it off... But it can also be very hard to find a good place for the cheater hole, so you accept the longer take off run.

[air-jon]

The length of ducting is fundamentally very important. every mm increases the thickness of the boundary layer and increases intake/efflux drag. the ideal fan is one with a short nacelle. Google

"Airbus E-Fan" to see what type of fan is created by lots of engineers, computers, and many 10s of millions of Euros. I assure you this is the most efficient EDF in existence.

There is no" ram air" with EDFs in level flight at constant throttle setting with a properly designed fan. In a properly designed ducted fan, the intake and exhaust velocity will always exceed the local external airspeed for a given fixed thrust setting, or an EDF would not accelerate. this is the most common misconception in ducted fan design. for instance, the efflux velocity of a typical 90mm EDF is around 180 to 200 MPH. Based on the delta between the intake and exhaust area, the intake velocity will be proportional to the delta. so intake velocity will be about 5 to 20% less than efflux velocity in the typical EDF (based on type of fan). For typical 90mm foamies, an intake velocity of 150 MPH would be expected, which is well in excess of the 100-120MPH max level speed. High performance composite EDF like a BVM Electra have higher efflux and intake velocities nearing 300 MPH and they still never experience "ram air" even when flying 220 MPH in level flight. The only way you could experience "ram air" is if the intake is vastly oversized, with very low intake velocity as Alpha pointed out earlier. This will kill the intake efficiency and reduce the top speed of the aircraft.

What does happen with speed is a reduction in intake turbulence, which effectively makes the intake larger when it goes laminar. depending on how sharp the intake is, this will happen at different speeds for different models. a stalled intake lip can make the intake lip function like one half its size. They do sound cool though!!!

Intake ducts have been compared by scholars to cylindrical wings with a fan inside (which is just a bunch more little wings) and all the standard wing principals still apply. Deeply ducted fans cause all the same problems as very low aspect ratio wings with higher drag and a thick boundary layer. this is why gliders, airliners have very thin wings (and short fan nacelles). Fan ducting requires a pressure differential as well with lower pressure at the intake and slightly higher pressure in the exhaust. This gets into pressure recovery and i'm not smart enough to talk on that..

You are correct about a slowly reducing diameter from intake lip to fan. this keeps the boundary layer minimal and creates the minimum drag with the air reaching maximum velocity just prior to the fan plane.

I think intakes are cool, others think they suck...;)

[fhhuber]

the fan can not pull in air fast enough to prevent stalling the fan blades at full throttle if the model is not moving forward. Thus there is a ram air effect of a sort.

You can also stall the blades of conventional 2 blade prop without the duct. This depends on prop pitch and rpm. A high pitch prop at high rpm is easier to stall.

The EDF is a very high pitch prop at very high rpm.

Tie a fish scale to the back of the airplane and slowly advance throttle. You will find a point where the pull of the EDF drops off with increased throttle. You stalled the fan. More throttle may continue to decrease the pull n the fish scale. Its not likely to increase without somehow improving air intake. this is ne reason we have cheater holes. Better intake means the fan is at a higher rpm before it stalls.
A good intake shape can be a major improvement, especially at low speed. (trying to gain speed for takeoff)

The airplane moving forward has the air speed going into the duct inlet higher, so it takes a higher rpm to stall the blades. When the airplane is moving fast enough, the air is coming in fast enough that full throttle does not stall the blades.

[air-jon]

Fans are much more difficult to stall than a prop. this is another advantage to ducted fans, (and fan stall another common misconception) a fan will not be fully stalled in a a stationary position...unless the intake is greatly restricted, or there is significant turbulence. a non-ideal intake lip will certainly be stalled at takeoff. This is easily tested on the bench by running a fan (with a proper bell intake, like the ones provided by WeMoTec) and observe the thrust and current consumption. Now, if you devise a slightly longer intake bell you can add a remotely actuated shutter...and you start to restrict the intake airflow, you can induce a fan stall (this can be done by reducing the exhaust as well). you will initially see a rise in power consumption, and then a sudden drop in consumption and increase when the fan stalls and is no longer moving air through the system with a closed intake...its just swirling like a blender (warning!...fan motors overheat very quickly in this state..ask me how I know...and this is a potently dangerous experiment..dont try at home ) This is why your home vacuum cleaner actually increases rpm when you block off the airflow.

Open propellers operate in a free air mass and the effect of the low pressure in front of the blade is quickly dissipated in all directions. there is also lots of reversion at the tips of an open propeller which promotes the tips to stall even sooner. there is very little or no reversion at the tips of a properly designed ducted fan. With an ducted fan, the parcel of air the fan disk sees is uniformly accelerated onto the fan disc. most of the pressure differential in front of the fan blades is distributed forward all the way to the intake lip. this creates a very different aero environment for the fan to live in tan an open propeller and is they keystone of ducted fan efficiency.

I only know what I do about ducted fan theory from reading academic papers from the interweb. there are hundreds dating back to the 1930s...written by guys and gals much smarter than me! you just have to dig a bit to mine the good ducted fan papers. once you lear the academic lingo,...you can search the terms like "deeply ducted fans" yielded this: also try fan tip reversion, fan pressure recovery..etc… good stuff to read when you cant sleep on a long flight...on a ducted fan powered airliner;)I promise you those fans are not stalled when an airliner releases the brakes at full throttle from Orange County Airport, ballistic noise abatement procedure...(they actually warn you before takeoff!)

https://www.eucass.eu/doi/EUCASS2017-651.pdf

[fhhuber]

Sorry, but EDFs are really easy to stall.

Just try the fish scale test... It is almost guaranteed that 100% of the EDFs you own will be proven to stall

Also it takes 150% as much power to get the same performance from an EDF as from an open prop... if the rest f the airplane (including weight) is the same. But the EDF model will weigh more...

Don't confuse a shrouded fan with an EDF

A shrouded fan the prop is in a ring that acts to reduce tip losses. The prop is still turning in the same rpm range as a conventional prop without the ring.
The shroud is typically less than 1 radius in length. Most of the length is for structural rigidity. Only the portion equal to 2 times blade chord is really needed to get the benefits of the shroud.

[air-jon]

All kinds of Nope! Ducted fan/shrouded propeller....all the same thing. I promise you if that were true, we would all be flying on turboprops across the oceans. Turbofan is a ducted fan, powered by a gas-turbine. Ducted fans can be much more efficient than open props in certain circumstances.

I'm the case of "deeply ducted fans" as we deal with in fighter jet models....the efficiency goes way down due to limited diameter and long ducting losses.

I never said you can't stall a ducted fan, only that it much more stall resistant than a propeller if the fan has proper intake and exhaust area.

https://massflow.archivale.com/ductbook.htm

https://www.grc.nasa.gov/www/k-12/airplane/inleth.html

https://www.researchgate.net/publica...n_VTOL_Systems

https://www.researchgate.net/publica...FD_Simulations

https://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19700006092.pdf

[Alpha]

I love conversations like these.

You're right, Air-Jon, length absolutely matters, both before the EDF and behind the EDF. It's easy to imagine all of these factors operating in extremes; I think most of us can easily envision that a 90mm F-16 with a ten foot long intake and a five foot long outlet tube would be less efficient at ingesting, transiting, accelerating, and expelling air. Logically there must be a sweet spot. Refer to my checklists above about resistance and pressure and the conclusion is foregone.

A lot of our prototype testing involves flight simulation modeling and in-flight measurement tools, because moving airflow is what we're most interested in optimizing. With each model aircraft, the "sweet spot" differs, because different aircraft bring different numbers to the equation (drag coefficient, etc). Astute fliers will notice that in general most of our modern generation Freewing jets are configured such that they have good acceleration when near stall speed. This is to help them "power out" of bad spots, such as when low and slow immediately after a waved off landing. To achieve this, we sacrifice portions of pure static thrust (on the tarmac at a standstill) and max dynamic thrust (full throttle diving).

[Elbee]

Guys, I want to thank all of you who contributed to this topic, personally. So, I thank you! I hope this thread continues as is reasonable as the information is great and the discussion relevant to me, anyway. One can never have too much knowledge and I am woefully behind in the tacit. Gratefully, LB

[wolpertinger]

Wow, reading the conversations make me realize why I have 2 ears, 2 eyes and 1 mouth. Twice as much listening and reading before talking so keep the lessons coming.... :Cool:

[air-jon]

I read these dissertations on aerodynamics and fans and such and of course I can really only absorb/understand about 30% because I don't speak equations...but I love to glean what I can from the little bits of english in between the numbers. Then you find the references In each paper and go down those rabbit holes. It's amazing the amount of work that has gone into ducted fan design alone!!!

[xviper]

I don't have enough brain cells to take in what's been said in this thread. :Confused:

[PaulZ]

Thanks Alpha! Very helpful and as an engineer this discussions gets my "propellers spinning". Gives me an appreciation for the work that goes into making these models fly so amazingly. Thank you!

[doctormike]

I fly at 7000 ft. alt. After buying several EDF's that don't fly for squat, I realized it wasn't the plane, but thin air. Are fan blades available with a higher pitch and would this help the problem? Thanks, Doc

[xviper]

I fly at 7000 ft. alt. After buying several EDF's that don't fly for squat, I realized it wasn't the plane, but thin air. Are fan blades available with a higher pitch and would this help the problem? Thanks, Doc

HobbyKing, years ago, sold empty fans and blades (individual) that you could pick separately that had user selectable pitch. You then balanced and assembled them at home. I don't know if they or anyone else does this now.

[fhhuber]

You can get an assortment of EDF rotors to put in your fan. I don't know of adjustable blade rotors, but you can get many different rotors of any diameter. Different blade counts and different pitches.

You can "cheat" and use a higher power fan rotor and motor in a lower rated fan housing. I did that to the Hobby King BAE Hawk - Red Arrow 70mm EDF 990mm (discontinued), using the motor and rotor from a Dynam Me262. LOTS more power than the included rotor and recommended motor.

When you add power you may have to add more cheater hole to get adequate air to the fan.

You can try a simple test without buying or changing anything... struggle for altitude and get it high. Point the nose down 45 deg keeping power on. See if the plane having accelerated is letting enough air enter to keep up with what the fan can push out. Smooth WIDE turns so you don't lose speed. If that helps, then altering the inlets to improve airflow will help.

sometimes the molds don't have the right taper for the proper 85% FSA outlet area. You can alter the outlet (usually needs some squeezing) with a section of styro or thin wall plastic drink cup. Cut the bottom off for the proper outlet and fit the cup inside the duct. You may want to make a ring to keep the new outlet centered in the original.

It just happens that the taper of a common fast food drink cup is a good taper for EDF outlet cones.