Alternative to pusher design

[foamie]

I was considering building a UAV with dual motors mounted on the wing? Are there any drawbacks to this design when compared with the pusher?

I know the reason for going with a pusher (so you can have an unobstructed view of the front), but are there other reasons why the pusher design seems to be the preferred method for UAV? Twin prop would also give me unobstructed view and I can avoid the inverted V-TAIL issue.

[kd7ost]

I was considering building a UAV with dual motors mounted on the wing? Are there any drawbacks to this design when compared with the pusher?

I know the reason for going with a pusher (so you can have an unobstructed view of the front), but are there other reasons why the pusher design seems to be the preferred method for UAV? Twin prop would also give me unobstructed view and I can avoid the inverted V-TAIL issue.

There are a few reasons to go with a pusher. One is as you described. Un-obstructed view. Another is that if you use a liquid fuel type engine, any residue from the exhaust or carb throat goes backwards and doesn't get on any camera or electronics. There is a third more important reason for me though. Ease of access to your cargo bay and Camera, GPS whatever. This may or may not be a concern to you in your setup. If the plane is big and you have a lot of heavy gear to put in it. A conventional design is pretty limited. You just can't keep loading everything under the wings. You have to move it ahead and behind the center of lift. You don't typically have much room to go to the nose area. This can mean a lot of gear getting shoved back into the fuselage and making access to it hard as well as setting up tail heavy conditions.

If you're just running a small camera or a couple of things, no problems. If you're using a small electric the fuel etc isn't an issue anymore. The pusher UAV's are designed the way they are so there is a big cargo space up front with no wing in the way etc. I'll put a picture of my plane here so you can kinda see the hatch thing.

There is nothing that says it need to be a vee or A or cathedral tail. I found it easier in the big gasser here to use two tails and give a servo to each rudder and one for the nose wheel.

So concider your design, your cargo and access that you will need. You might not have to use a pusher design. Two motors on wings are fine but you have to structure for it. A single larger motor or engine on a pod above the wing in a conventional setup is also a good choice. Lots of guys are playing with the Engine in the rear like a pradator design. Thats cool too and has design properties of the pusher as well as ease of fuselage structure. You don't have to build and line up those booms. It adds inside cargo space.

Good luck

Dan

[AeroRon]

Here are some other things to consider if you have a twin engine a/c:

-How do you know the RPM on each engine is the same?

If you have a gas plane you can get a tach to measure the RPM and adjust the mixture accordingly. I'm not sure what you'd do if you have an electric plane.

-What happens if an engine fails in flight?

I assume you're thinking about putting this design on a UAV. How is your autopilot going to handle this situation? Also, you need to made sure you have enough rudder authority to handle the yaw from only one engine running.

-Where is the power coming from?

Obviously, you're going to need more fuel or batteries with two engines than one.

-Have you thought about the structure?

You are going to need a stronger wing to support two engines. The way you attach the wing to the fuselage is going to need beefing up to transfer the loads.

Not to sound completely negative, there are some positive aspects.

-You can carry a larger payload.
-If one engine does die, you could possibly bring it home on the second one.
-As you mentioned, an unobstructed view.

I'm sure I've missed serveral pros and cons of a twin engine design. Anyone feel free to add on or correct.

-Ron

[AeroRon]

Here's a UAV that the military is currrently using in Iraq and Afganistan that has two engines (well, motors in this case).

http://www.baiaerosystems.com/evolution.html

-Ron

http://www.baiaerosystems.com/images/evolution1.gif

[alexcmag]

A twin-engine aproach with fuel powered engines is really a trouble, it will be very hard to make the throtle response equal on both.

However, electric motors haven't this trouble. I made several twin-motor electric models with no problem. If you use two brushed motors on the same ESC or two identical brushless motors on two identical ESCs with identical setups they will run aproximately on the same RPM. Any small difference on power is not significative and will be less influent to the flight then the propeller-torque of a single greater motor.

Actually it is a good aproach if low-profile and ground clearance is needed (for belly landing or very small landing gear for example).

The power needed increases too, but if smaller motors are used you can have very near power and thrust values.

However, I love the pusher design on UAVs. My favorite is the Aerosonde, a very practical design:
http://www.aerosonde.com/images/upload/299486595_gallery004_250.jpg

[senor_paco]

How about twin engines on a pylon above the wing? One engine goes out, ok, maybe some torque issues, but nothing says they have to counter-rotate. If you are any kind of R/C pilot, or your UAV can't cope with that, you have other issues. I mean, it works for a lot of seaplanes/amphibious. Then if you had a camera within a dome, you could even look behind if you needed too. Plus less of a chance of broken props. you'd have to consider your flight envelope, but that shouldn't really make any appreciable difference. Just my .02

Electric motors, especially brushed, are the way to go, as per Alexcmag's response. You can always add more capacity, as long as your wing loading/speed allows it.

[Hovertime]

One motor is less costly, less complicated, and prop in front is more efficient than prop in the back, as air is not smooth anymore after fuselage/wing disturbs it. One large prop is also much more efficient than 2 smaller ones, meaning longer flight times.

[alexcmag]

One motor is less costly, less complicated, and prop in front is more efficient than prop in the back, as air is not smooth anymore after fuselage/wing disturbs it. One large prop is also much more efficient than 2 smaller ones, meaning longer flight times.

You're right.

Single engine with large prop on the front of a slim fuselage is the most efficient. With a 4-stroke gas engine or a geared brushless motor it is even better.

However, when a camera on front is needed, or ground clearance is required due to belly landing or a less precise automatic landing, pusher or twin design can be a good choice. Of course they will need more fuel or battery, but for short flights it can be usefull.

[kd7ost]

One motor is less costly, less complicated, and prop in front is more efficient than prop in the back, as air is not smooth anymore after fuselage/wing disturbs it. One large prop is also much more efficient than 2 smaller ones, meaning longer flight times.

All true except prop in front is more efficient than prop in the back, as air is not smooth anymore after fuselage/wing disturbs it.

A properly designed and built fuselage with wings extended out has a fairly laminar airflow around it in conventional flight. The turbulent air is only in a small layer next to the fuselage and only increases in length down the fuselage where aerodynamic care has been left out. The prop is most effective at the outer 3/4 to 2/3 of its length. In cases where proper design has been accomplished, the effective part of the propeller is outside of the air that has been disturbed by the fuselage and wing roots.

Conversely, A big propeller in the front is certainly in clean air. But once it does it's job it has created swirling turbulent air to send down the fuselage and around the wing roots. Depending on prop size, RPM, airspeed etc, the amount of efficiency lost by creating this dirty air, and its resultant loss of lift, is greater than the efficiency lost by putting the prop in the back. This is irrespective of how clean the lines are in the design.

We simply have such good power to weight ratio's that it's not generally important.

Dan

[Myron]

Hey guys,

I think alot of it really depends on your mission parameters. I think one of the most desireable atributes of a small or micro UAV is loiter time... Thats probabaly why most are single power plant. I want a unit that will carry a heavy load and I have been entertaining the idea of building one that is similar to an OV-10 Bronco with a couple AXI 2826-12's for power... the motor will fly a 13 lb aircraft on one large 11.1 volt pack... motor only weighs 6.0 oz and the pack is 11 oz.... Do the math on that one and you could haul a hefty payload with a very small power system
For all of our AP sorties that just need digital stills we use flying wings and they are all in sub 4 lbs category... It just really depends on what you want to do....

Myron

[alexcmag]

All true except

A properly designed and built fuselage with wings extended out has a fairly laminar airflow around it in conventional flight. The turbulent air is only in a small layer next to the fuselage and only increases in length down the fuselage where aerodynamic care has been left out. The prop is most effective at the outer 3/4 to 2/3 of its length. In cases where proper design has been accomplished, the effective part of the propeller is outside of the air that has been disturbed by the fuselage and wing roots.

Conversely, A big propeller in the front is certainly in clean air. But once it does it's job it has created swirling turbulent air to send down the fuselage and around the wing roots. Depending on prop size, RPM, airspeed etc, the amount of efficiency lost by creating this dirty air, and its resultant loss of lift, is greater than the efficiency lost by putting the prop in the back. This is irrespective of how clean the lines are in the design.

We simply have such good power to weight ratio's that it's not generally important.

Dan

Dan,

Since you have a great acknowledge of turbulence on propellers and after then, there is a question that always beat me...

The input airstream (before the propeller) is slower, but the volume of air before and
after the propeller must be the same, so the airstream coming to the propeller must be a large cone.

The output stream of the propeller is faster, so its diameter must be smaller then input stream.

If I have a large fuselage (perhaps half of propeller disc area), will be a tractor propeller better (even blowing on cowl, windshild, landing gear, etc.) or worst then a pusher propeller (receiving a low pressure turbulent airflow but blowing to the open air)?

[kd7ost]

For all of our AP sorties that just need digital stills we use flying wings and they are all in sub 4 lbs category... It just really depends on what you want to do.... Myron

Flying wings. There is a clasic example of a pusher design, with ultra efficient airframe, and not much to dirty the airflow to the pusher prop. It has pro's and con's like any design but in the area of being effecient, it's pretty much at the top of the charts. The whole front of that design is LE (Leading Edge) with nothing standing in the way of generating lift in the center. Put some tips on that to keep the vortices off the wing tips down to a minimum and you have the single design that will carry more weight per platform size than anything else. It's pure lift. If you put the prop in front, you would lose a fairly hefty amount of clean lift by the prop wash disturbing the laminar nature of the air flow needed to create the lift. The whole center section, at least as wide as the prop (And actually a little more) would be disturbed.

And uh, ahem, just for the record I fly a twin boom pusher. ;-)

Dan

[Hovertime]

Front mounted prop is better. Don't forget that much larger portion of thrust is created by forming a vacuum in front of prop blades than by pushing air back or deflection of air from the back of prop blades..

[kd7ost]

Dan,Since you have a great acknowledge of turbulence on propellers and after then, there is a question that always beat me...

You are too kind. I am not an expert but have read much on the subject and like others here am simply regurgitating it as fact.

The input airstream (before the propeller) is slower, but the volume of air before and
after the propeller must be the same, so the airstream coming to the propeller must be a large cone....

Yes, but it’s not a constant. When your plane is stationary and you give it full throttle, this effect you describe is greatest. Also high pressure behind the prop prevents it from reaching full RPM. The pressure behind the prop, the size and RPM of the prop all determine how much this effect is there. As the plane goes into motion, The higher pressure behind the prop gets less, the cone in front shrinks, the RPM increases as it ”unloads”. Soon you are moving fast enough that the cone in front is smaller. (I also want to stress that for the purpose of this discussion, we are talking conventional UAV type flight. Not helicopters, no 3D etc) Just smooth efficient, works fine and lasts a long time type of gentle UAV flying. It would be best to not muddle things with discussions of other type of aircraft, even though the principles are all the same.

The output stream of the propeller is faster, so its diameter must be smaller then input stream....

The propeller is actually a flat bottom high lift air foil section. So it works the same as a lifting airfoil. Since the distance across the top (front) of the airfoil shaped prop blade is longer than the flat underside (rear) of the prop, its motion creates low pressure in front and high pressure behind. This is where the propeller design makes it work. Changing angle of attack (pitch) allows the propeller to work at high air speeds. (Need more horsepower) Your description partially holds, but the systems aren’t working in a vacuum. Also, your description isn’t what makes it work. It’s a bi-product of it working. Think about swinging a high lift wing in an arc. Once you get enough speed, it’s the lower pressure on top because the volume of air on top has a longer way to go than the short distance traveling air underneath that causes it to move upward. As we add power we can increase the pitch angle and get motion to deflect further upward. In your statement above though, if there was nothing in front of or behind the prop, there would still be cone shapes in front of and behind the prop. (We’re still pretending a vacuum) Think of fluid dynamics. It’s quite similar. There will be a wake from the rear high pressure area that gradually spreads its cone shape to the ambient air and decrease in intensity as it spreads out. Like the wake behind a moving boat but in 3D.

If I have a large fuselage (perhaps half of propeller disc area), will be a tractor propeller better (even blowing on cowl, windshild, landing gear, etc.) or worst then a pusher propeller (receiving a low pressure turbulent airflow but blowing to the open air)?

This isn’t a real easy answer. You might have to wind tunnel test to know. It depends on the efficiency of the design. If you have a flat fire wall up front and simply move the engine to the rear while that fire wall is left up there like that, you will suffer. If you have a real long thin aerodynamic shaped nose, with a smooth fuselage, gussets around the roots where they attach to the fuselage etc. then you will get better results or more efficiency from a pusher. It might have to be cowled in with a spinner on it too. You don’t want anywhere behind any moving surface where the air slams shut with a bang. That creates low pressure (drag) to a moving object. The effects of all these things are dynamic too. They change in concert with speed, angle of attack etc.

Please, any aerodynamic engineers feel free to ring in on this. I might not be correct on all points or have explained it well.

Dan

[Magician]

Aircraft configuration design is an interesting mix of compromises and many people have their own "favorite" for a variety of reasons.

As for the tractor vs pusher debate, it is a fact that on the Cessna 337 Skymaster, push-me-pull-me in-line twin, is more efficient on the aft motor (pusher) than the front motor (tractor). This was a favorite trick of observation pilots for better fuel economy, they stopped and feathered the front motor and cruised on the aft motor. The reason, the high energy air from the tractor engine got into all the nooks and crannies (wing/landing gear/fuse junctures) and had a significant drag rise. Conversely the air into the pusher prop, although slightly turbulent from the fuselage boundary layer and wing wake, had less drag rise overall.

However, it's all a tradeoff depending on the mission. The Dragon Eye/Evolution electric twin pictured earlier was a twin not only to allow for a removable modular payload nose but to cancel out the torque during a full power handlaunch by a non-modeler to avoid a nasty torque roll into terra firma.

The pylon mounted engines must be approached cautiously due to the potential trim changes with power caused by the high thrust lines. Modern radios can help this with proper mixing scheduling.

Hope this helps the fun discussion!

Chris

[kd7ost]

thrust is created by forming a vacuum in front of prop blades than by pushing air back or deflection of air from the back of prop blades..

This is largely a true statement. There is some deflection but it is minor when compared to lift of the prop cross section.

Some deflection concepts, like a helicopter with fully simetrical blades defy the above explanation. It ends up being able to deflect and cause pressure under the drooping trailing edge. (Again though, this is not the right place for that conversation) Have power? Will fly.

I do disagree that the prop up front is the most effecient way to fly. I do agree that the prop up front is the most efficient place to put the prop. Thing is, the fuselage and wing are also in the best place if they are up front. So who goes to the back? The prop disturbs a lot of air for the plane. If it is built clean, a fuselage and wing disturb very little airflow for the prop. It is therefore more efficient to put the prop in the back of a clean plane if you're looking at the whole plane, prop, engine combination to determine your effeciency rating for a given flight platform.

Dan

[kd7ost]

Here's just some of the Masters work.

http://www.airventuremuseum.org/collection/aircraft/Rutan%20VariEze%20Prototype.asp

Dan

[alexcmag]

Also on the pusher design...

I made some tests on tail design vs pusher propeller.

I tried cross-tail, V-tail and inverted V-tail with twin boons (Aerosonde Style) and even with the same fuselage and wing, the inverted V-tail flies much better. I think it's because the drag of tail on the high speed - high pressure airstream. My tractor models didn't have this huge difference with different tail designs.

Strange reactions to controls when full-power are other trouble with the tail on the output stream.

My conclusion... It is better to avoid surface controls on the direct airflow of a propeller.

I tried 3 models with the thrust line over the CG: SeaWind (motor mounted on the vertical fin), Savoya-Marchetti S-55 (motor on a support over the wing, over the leading edge) and Frog (motor on a pod on the trailing edge of wing).

All have a really bad flight... The SeaWind needed 5-degree incidence on the horizontal fin (too much drag), the S-55 needed a 10-degree upthrust (blowing over the wing, more drag) and the Frog tries to dive when full-power is applied, needing a lot of correction on the elevator to fly.

The Motor of SeaWind was later replaced by twin motors on wings, the tail incidence was set to 0 and the old motor support removed, saving weight. The twin-motor version flies much better and with less power.

The S-55 was dead after some crashes.

The frog has his geared-motor replaced by a direct-drive with a smaller propeller, lowering the thrust line. The power needed to fly is a little greater then before (with the gearbox), but the flight is much better now.


My conclusion... I will never build something with the thrust line high over the CG again...

[Tristar500]

Unless It was already addressed, another good reason to have the motor (engine) in the rear or at least not on the nose is that is isn't the first line of defense in the event of a crash. Especially for first test flights. I fly the plane empty except for essentials and a dummy payload in the nose if I need it to make weight and balance while testing a new airframe.

Most all of my UAV designs are pushers for several reasons the above being just one of these. Pushers in general collect some stability from the pusher configuration and are in general more efficient due to the fact that the aircraft isn't flying in it's own propeller slipstream. The others brought up a lot of good points as well.

Our Sapphire UAV will be featured in the next issue of FlyRC magazine. We hope to have an all carbon fiber turbine model ready for testing soon.

Here’s a pic of the prototype.

Lawrence

Innovation Robotics
Tea, Leader/Space Glider Project

[Hovertime]

Well whatever rocks your boat, but large prop is much more efficient than small one, and in most cases you can mount larger prop in front than in the back.

So use large props gents;)

Yeah, Burt Rutan is certainly a guy to ask these sort of questions;)

BTW , went to Oshkosh , got bad sunburn.... it hurts... But it was sure fun trip otherwise. (LOTS of VariEze'stoo :))

[kd7ost]

Well whatever rocks your boat, but large prop is much more efficient than small one

A large prop turning slow is definately better than a small prop turning fast. I read some time back that the most efficient prop design of all is a single bladed very long prop that is counter balanced on the other side. Practicality of manufacturing prevents us from trying it out though.

and in most cases you can mount larger prop in front than in the back

I have a fuji BT32A. If I mount it up front I use a 18x8 prop. If I mount it in a pusher configuration like I have it now, I still use a 18x8. It's just a pusher prop is all. There is nothing I'm aware of that dictates props in the back are generally smaller than props up front. It comes down to loading and horspower ratings with gas, and how much wattage you want to expend with electric.

Dan

[kd7ost]

Man, Thats a nice looking wing Lawrence. Sweeeeeet.

Dan

[Hovertime]

I have a fuji BT32A. If I mount it up front I use a 18x8 prop. If I mount it in a pusher configuration like I have it now, I still use a 18x8. It's just a pusher prop is all. There is nothing I'm aware of that dictates props in the back are generally smaller than props up front. It comes down to loading and horspower ratings with gas, and how much wattage you want to expend with electric.

Dan

Well in case of delta wing or v tail space is limited for a big pusher prop.

[alexcmag]

Unless It was already addressed, another good reason to have the motor (engine) in the rear or at least not on the nose is that is isn't the first line of defense in the event of a crash. Especially for first test flights. I fly the plane empty except for essentials and a dummy payload in the nose if I need it to make weight and balance while testing a new airframe.


This is the same reason I started to build pusher RC models... My pushers saved me a lot of propellers :P


Here’s a pic of the prototype.


Nice design, don't need to say I love this design...

Look mine.

They are not UAVs, just JunkyPlanes to test flight concepts, but all are stable and have an excelent flight.

[kd7ost]

Very nice designs. You stay very active trying different things I see. I think thats fun.

Dan

[Tristar500]

Good looking stuff Alex, I'm guessing the polyhedral designs combined with some sweep should give you a great platform that is very stable.

Please keep us updated on your progress!

Best, Lawrence

Team Leader/ Space Glider Project/Sapphire UAV

[Tristar500]

Man, Thats a nice looking wing Lawrence. Sweeeeeet.

Dan

Thanks Dan, One of these will probably be flown with one of the Unav PDC-10 auto-piliots you sent me.

I'll post as we progress.
Best, Lawrence

Team Leader/ Space Glider Project/Sapphire UAV

[alexcmag]

Very nice designs. You stay very active trying different things I see. I think thats fun.

Dan

Thanks, this is the advantage of depron / EPS building...

I can make at least one model each month and try different designs.

[alexcmag]

Good looking stuff Alex, I'm guessing the polyhedral designs combined with some sweep should give you a great platform that is very stable.

Please keep us updated on your progress!

Best, Lawrence

Team Leader/ Space Glider Project/Sapphire UAV

Thanks, they are really very stable.

The second model (my avatar) can self-recover from most flight attitudes back to horizontal flight in less then 1 second.