Experimental EDF Jets and Other Ideas

[Houndpup Rc]

95A isn’t too bad either. (Then again, it’s the only one I’ve used 🤣)

I was hoping to be able to use it for the tires on the Rc car I wanna make...But as the video showed, the softer you get the more traction but then as you go faster centrifugal force with make it expand and pop off the hub...

 

[telnar1236]

Even 85A is honestly too hard for tires for anything less than about 2 lbs - it works well for rc plane tires on concrete and dirt since they dont need that much traction and the lower friction reduces wear, but back when I was at a club where we flew from some kind of plastic/fabric like what Flite Test uses, anyone using TPU tires was skidding all over the place. I think they make filaments embedded with pieces of actual rubber for friction but I hear they tend to be pretty awful to print.

85A is definitely a bit tough to print - you need a direct drive extruder and a filament path that doesn't put the filament in compression for any real distance.

 

[Mr Man]

Ordered an ender 3 s1, so I can print out my vortex!

 

[Houndpup Rc]

Ordered an ender 3 s1, so I can print out my vortex!

Cool!😁👍

 

[Houndpup Rc]

Even 85A is honestly too hard for tires for anything less than about 2 lbs - it works well for rc plane tires on concrete and dirt since they dont need that much traction and the lower friction reduces wear, but back when I was at a club where we flew from some kind of plastic/fabric like what Flite Test uses, anyone using TPU tires was skidding all over the place. I think they make filaments embedded with pieces of actual rubber for friction but I hear they tend to be pretty awful to print.

85A is definitely a bit tough to print - you need a direct drive extruder and a filament path that doesn't put the filament in compression for any real distance.

Yeah, if I do a car, I should probably just buy real rc car tires....

 

[Houndpup Rc]

85A is definitely a bit tough to print - you need a direct drive extruder and a filament path that doesn't put the filament in compression for any real distance.

Yup, mine is direct drive and honestly has printed any filament I have tried really good!! (Working on a review video!)

 

[telnar1236]

One of the challenges of designing this plane is trying to design it so that it is both mechanically and structurally simple enough to work, hopefully with minimum of fiddling about with it, but also so that it maintains the streamlining I originally designed in so that it can actually come close to my projected speeds. The gear doors and control surfaces are two big places where this will be especially important.

My original goal was to have fully internal linkages, but with the ailerons that didn't turn out to be practical (as seen in previous posts). More interesting though, from CFD, it seems like it's actually slightly better not to fair in the aileron linkages, at least using a conventional design. I need to look at airfoil shaped fairings as those may be better.

I think the cross-sectional area of the fairing is just so much greater than the control arms alone that it adds more drag than it removes.
The gear doors are also a bit challenging. Originally, I was going to try and design two-piece main gear doors with a hinge in the middle, so they could fit flush with the wing, but there just wasn't room for the hinges and spring/rubber band that would be required to keep them from folding the wrong way and jamming, so instead I went with a simpler design which leaves a bit of a gap while retracted. I may revisit this after I get it flying, but it's probably a difference of half a mile an hour in speed so I don't think it's worth the time to get it functioning properly.

A lesson I learned with my 90mm F-106 is that the gear doors should have enough clearance not to hit the ground even when the suspension is at its full travel since the F-106 gear doors get scratched up on harder landings, so on this design, there is plenty of space for that.

 

[Mr Man]

Interested to see how it flies!

 

[Piotrsko]

One of the challenges of designing this plane is trying to design it so that it is both mechanically and structurally simple enough to work, hopefully with minimum of fiddling about with it, but also so that it maintains the streamlining I originally designed in so that it can actually come close to my projected speeds. The gear doors and control surfaces are two big places where this will be especially important.

My original goal was to have fully internal linkages, but with the ailerons that didn't turn out to be practical (as seen in previous posts). More interesting though, from CFD, it seems like it's actually slightly better not to fair in the aileron linkages, at least using a conventional design. I need to look at airfoil shaped fairings as those may be better.
View attachment 255371
View attachment 255372
I think the cross-sectional area of the fairing is just so much greater than the control arms alone that it adds more drag than it removes.
The gear doors are also a bit challenging. Originally, I was going to try and design two-piece main gear doors with a hinge in the middle, so they could fit flush with the wing, but there just wasn't room for the hinges and spring/rubber band that would be required to keep them from folding the wrong way and jamming, so instead I went with a simpler design which leaves a bit of a gap while retracted. I may revisit this after I get it flying, but it's probably a difference of half a mile an hour in speed so I don't think it's worth the time to get it functioning properly.
View attachment 255373
View attachment 255374
A lesson I learned with my 90mm F-106 is that the gear doors should have enough clearance not to hit the ground even when the suspension is at its full travel since the F-106 gear doors get scratched up on harder landings, so on this design, there is plenty of space for that.
View attachment 255375

Thats why on full sized there is a separate door just for the wheel and axle assembly. Typically operated with springs to open and uses the wheel to hold it shut. You probably know that but you could go WW2 vintage and add a wheel cover wide spot.

 

[telnar1236]

Thats why on full sized there is a separate door just for the wheel and axle assembly. Typically operated with springs to open and uses the wheel to hold it shut. You probably know that but you could go WW2 vintage and add a wheel cover wide spot.

I'll definitely have separate doors over the wheel wells, probably spring loaded like you mentioned, or using rubber bands more likely, though I also do have a gear door sequencer from a crashed plane from years ago which I might try to use (if it even still works). Those are pretty significant from a drag perspective - with how clean this design is I might lose 10+ mph without them. The gaps I'm deciding to be ok with are here, but the wheels are fully in the fuselage which gives me plenty of space (I learned my lesson on this with my F-106 where the main wheels are partly in the wing which caused a whole host of issues.

One of my favorite anecdotes from real aviation, though, is about the Boeing 737 - they realized that including a gear door added enough weight that it was more fuel efficient to just leave the wheel well open - though of course the parasite drag isn't what's limiting the stop speed of an airliner like that.

 

[telnar1236]

This is quite the long one, so I'll start with the summary - having less than perfect hinge designs on rc planes seems to basically not matter. The difference is so small as to be negligible between a carefully designed faired in hinge and something a lot sharper like what you see on most foam planes. Making nice looking hinges is mostly an aesthetic thing it turns out.

With my Ghost design nearing being ready to fly, it's time to start shifting this project back into high gear. So long as the Ghost performs the way I expect, I'll have as much testing and background on this design as I'm likely to get - the ghost uses a similar inlet and duct design, identical main gear, and the same laminar flow airfoil in a plane that should be heavier than the super duper sabre but with less thrust.
One thing I hadn't investigated until now is flap/control surface hinge design. In most of my earlier designs I used a very simple design where the wing had a sharp back. This works quite well and I haven't really had any issues. The one possible exception is with my Saab 105 where the flaps felt like they didn't really do much, but I didn't know if that was due to the hinges. My working theory was that the slots might actually slightly increase the effectiveness.

And in my newer designs, I've used a design where the hinge is intended to be smoother and disrupt the airflow less. I didn't have any data to support this being better.

So, I ran a simulation to get the data, for the smooth and sharp hinges with a clark Y airfoil, and for the smooth hinge for the laminar flow airfoil in this project (the Clark Y gave me enough data to know I wouldn't be using a sharp hinge on this plane since it is very slightly draggier but it's still good to know for the future since it's easier to model the sharp hinge). The clark Y flap is hinged at 22% of the chord from the back and the laminar flow airfoil at 20% of the chord from the back.

The answer turns out to be that the sharp hinge is slightly draggier and produces a tiny bit less lift, especially at higher angles of attack. Cy is the Clark Y, and IW is the laminar flow airfoil. S is the sharp hinge, M is the smooth hinge, and no letter is the base airfoil. If it has 45 at the end, the flap is deflected 45 degrees. The Y axis is the coefficient of lift or drag and the X axis the angle of attack in degrees.

What seems to be happening is that while the sharp edge does cause flow separation, the control surfaces are far enough back that the airflow is already pretty turbulent anyway. A tiny bit of lift is lost and drag added, but it's so small as to be unimportant. You also get a tiny reduction in nose down pitching moment with the sharp hinge which is an improvement in most use cases but again the change is minimal. You can see the boundary layer is already turbulent and slow by the arrow on the Clark Y airfoil.

So less than perfect hinges have minimal impact for most planes. The broader conclusion from this is that it's not really worth the time except for possibly in terms of looks.


This analysis also incidentally further confirms my choice of the laminar flow airfoil for the wing - both on the super duper sabre and the Ghost - lift is 19% lower with full flaps but drag is 30% lower which means that for a given stall speed, a wing with the laminar flow airfoil will be faster taking flaps into account.

 

[Mr Man]

This is quite the long one, so I'll start with the summary - having less than perfect hinge designs on rc planes seems to basically not matter. The difference is so small as to be negligible between a carefully designed faired in hinge and something a lot sharper like what you see on most foam planes. Making nice looking hinges is mostly an aesthetic thing it turns out.

With my Ghost design nearing being ready to fly, it's time to start shifting this project back into high gear. So long as the Ghost performs the way I expect, I'll have as much testing and background on this design as I'm likely to get - the ghost uses a similar inlet and duct design, identical main gear, and the same laminar flow airfoil in a plane that should be heavier than the super duper sabre but with less thrust.
One thing I hadn't investigated until now is flap/control surface hinge design. In most of my earlier designs I used a very simple design where the wing had a sharp back. This works quite well and I haven't really had any issues. The one possible exception is with my Saab 105 where the flaps felt like they didn't really do much, but I didn't know if that was due to the hinges. My working theory was that the slots might actually slightly increase the effectiveness.

View attachment 255803

And in my newer designs, I've used a design where the hinge is intended to be smoother and disrupt the airflow less. I didn't have any data to support this being better.


View attachment 255804

So, I ran a simulation to get the data, for the smooth and sharp hinges with a clark Y airfoil, and for the smooth hinge for the laminar flow airfoil in this project (the Clark Y gave me enough data to know I wouldn't be using a sharp hinge on this plane since it is very slightly draggier but it's still good to know for the future since it's easier to model the sharp hinge). The clark Y flap is hinged at 22% of the chord from the back and the laminar flow airfoil at 20% of the chord from the back.

The answer turns out to be that the sharp hinge is slightly draggier and produces a tiny bit less lift, especially at higher angles of attack. Cy is the Clark Y, and IW is the laminar flow airfoil. S is the sharp hinge, M is the smooth hinge, and no letter is the base airfoil. If it has 45 at the end, the flap is deflected 45 degrees. The Y axis is the coefficient of lift or drag and the X axis the angle of attack in degrees.
View attachment 255825
View attachment 255826

What seems to be happening is that while the sharp edge does cause flow separation, the control surfaces are far enough back that the airflow is already pretty turbulent anyway. A tiny bit of lift is lost and drag added, but it's so small as to be unimportant. You also get a tiny reduction in nose down pitching moment with the sharp hinge which is an improvement in most use cases but again the change is minimal. You can see the boundary layer is already turbulent and slow by the arrow on the Clark Y airfoil.
View attachment 255830
View attachment 255827 So less than perfect hinges have minimal impact for most planes. The broader conclusion from this is that it's not really worth the time except for possibly in terms of looks.


This analysis also incidentally further confirms my choice of the laminar flow airfoil for the wing - both on the super duper sabre and the Ghost - lift is 19% lower with full flaps but drag is 30% lower which means that for a given stall speed, a wing with the laminar flow airfoil will be faster taking flaps into account.

Very interesting!

 

[LitterBug]

This is quite the long one, so I'll start with the summary - having less than perfect hinge designs on rc planes seems to basically not matter. The difference is so small as to be negligible between a carefully designed faired in hinge and something a lot sharper like what you see on most foam planes. Making nice looking hinges is mostly an aesthetic thing it turns out.

With my Ghost design nearing being ready to fly, it's time to start shifting this project back into high gear. So long as the Ghost performs the way I expect, I'll have as much testing and background on this design as I'm likely to get - the ghost uses a similar inlet and duct design, identical main gear, and the same laminar flow airfoil in a plane that should be heavier than the super duper sabre but with less thrust.
One thing I hadn't investigated until now is flap/control surface hinge design. In most of my earlier designs I used a very simple design where the wing had a sharp back. This works quite well and I haven't really had any issues. The one possible exception is with my Saab 105 where the flaps felt like they didn't really do much, but I didn't know if that was due to the hinges. My working theory was that the slots might actually slightly increase the effectiveness.

View attachment 255803

And in my newer designs, I've used a design where the hinge is intended to be smoother and disrupt the airflow less. I didn't have any data to support this being better.


View attachment 255804

So, I ran a simulation to get the data, for the smooth and sharp hinges with a clark Y airfoil, and for the smooth hinge for the laminar flow airfoil in this project (the Clark Y gave me enough data to know I wouldn't be using a sharp hinge on this plane since it is very slightly draggier but it's still good to know for the future since it's easier to model the sharp hinge). The clark Y flap is hinged at 22% of the chord from the back and the laminar flow airfoil at 20% of the chord from the back.

The answer turns out to be that the sharp hinge is slightly draggier and produces a tiny bit less lift, especially at higher angles of attack. Cy is the Clark Y, and IW is the laminar flow airfoil. S is the sharp hinge, M is the smooth hinge, and no letter is the base airfoil. If it has 45 at the end, the flap is deflected 45 degrees. The Y axis is the coefficient of lift or drag and the X axis the angle of attack in degrees.
View attachment 255825
View attachment 255826

What seems to be happening is that while the sharp edge does cause flow separation, the control surfaces are far enough back that the airflow is already pretty turbulent anyway. A tiny bit of lift is lost and drag added, but it's so small as to be unimportant. You also get a tiny reduction in nose down pitching moment with the sharp hinge which is an improvement in most use cases but again the change is minimal. You can see the boundary layer is already turbulent and slow by the arrow on the Clark Y airfoil.
View attachment 255830
View attachment 255827 So less than perfect hinges have minimal impact for most planes. The broader conclusion from this is that it's not really worth the time except for possibly in terms of looks.


This analysis also incidentally further confirms my choice of the laminar flow airfoil for the wing - both on the super duper sabre and the Ghost - lift is 19% lower with full flaps but drag is 30% lower which means that for a given stall speed, a wing with the laminar flow airfoil will be faster taking flaps into account.

Can you do a comparison of paper (tape) hinges on the top vs the bottom surface of the wing?

 

[skymaster]

I have a question since the topic is edf's. does the edf motor always have to go in the back of the plane? has anyone ever made an airplane with the EDF motor in the front?

 

[Piotrsko]

@telnar1236 : how are you getting laminar on model scale wings?

 

[telnar1236]

Can you do a comparison of paper (tape) hinges on the top vs the bottom surface of the wing?

Might as well next time I run some CFD for completeness sake. Based on the results from the ones I've run so far I'm guessing they'll be about the same with the one with the tape on the bottom being maybe a bit worse

 

[telnar1236]

I have a question since the topic is edf's. does the edf motor always have to go in the back of the plane? has anyone ever made an airplane with the EDF motor in the front?

I think typically it's easier to but it at the back, since that's where it tends to make a plane balance well and maybe a bit more efficient, but I've seen a couple builds with the EDF at the front. In terms of having the motor in the front of the EDF unit, I've never owned one, but I think there are a couple out there with a "pusher" configuration. I haven't really looked into why that is

 

[telnar1236]

@telnar1236 : how are you getting laminar on model scale wings?

Same as full scale more or less - careful shaping and construction of the wing. There have been a number of RC laminar flow airfoils floating around for a while - some of the MH airfoils for example - and mine isn't really an improvement on them - it's just tailored a bit more to exactly what I want for this plane - optimized for a bit lower a Reynolds number (smaller/slower rc plane) and a more gradual pressure recovery zone so I don't need to use turbulators. The penalty is that my drag bucket (AOA range where the flow is laminar) is a bit smaller.

Both on real planes and rc planes, a laminar flow airfoil doesn't actually mean fully laminar flow - just a delayed transition from laminar to turbulent. The purple line on the wing represents the transition and you can see how much later it happens on a laminar flow airfoil (top) vs. a different airfoil with a similar thickness (Prandtl D wing root airfoil) on the bottom. The front of the wing is down.

I also can't 3D print the wing perfectly, so those minor imperfections are probably causing a transition earlier on the wing, but you can still feel the difference between flying a plane with a laminar flow wing and one without. They hold on to speed a lot more and gain speed much more easily in a dive. You can also feel the transition from laminar to turbulent flow as you slow them down to land and get into slow flight - after a certain amount of elevator, the plane goes from feeling like it won't slow down to feeling a bit sluggish. The penalty for the improvement to drag is basically everything else - you get less lift and a lower stall angle of attack. If you're going for performance, or a wide speed range since you get more in terms of drag improvements than you lose in terms of lift, a laminar flow airfoil is great, but if you just want a nice to fly plane, it's probably not the best choice honestly.

 

[Piotrsko]

I think typically it's easier to but it at the back, since that's where it tends to make a plane balance well and maybe a bit more efficient, but I've seen a couple builds with the EDF at the front. In terms of having the motor in the front of the EDF unit, I've never owned one, but I think there are a couple out there with a "pusher" configuration. I haven't really looked into why that is

Possibly since there few full size craft set up that way so the modelling options are slim. Not a lot of avant guard builders any more, it seems. The Soviets were good at odd purpose built craft, but that era seems to be over.
Your experiments with laminar match my experiments, although I suspect mine were probably less efficient since until just now I thought my issues were with sharp leading edges.

 

[telnar1236]

Can you do a comparison of paper (tape) hinges on the top vs the bottom surface of the wing?

Ran the comparison between top and bottom tape and foam board hinges. I used the Clark Y airfoil since it's pretty representative of most RC plane airfoils. I'm referring to the top hinge as having the tape on top.

And the bottom hinge as having the tape on the bottom.

Overall, the top hinge is a bit better when the surface is not deflected, but it's not a huge difference - other considerations like aesthetics and strength are probably more important. Both are slightly worse than the gapless hinge design from my previous post in terms of drag, but the top hinge version is better than the sharp hinge version from that post. The bottom hinge version, on the other hand, produces more a bit lift with the flaps down. I think this because the hinge position means a larger part of the flap extends below the wing, effectively making the flap a bit bigger.

Based on these results, my recommendation would be to use top hinges on the ailerons for pretty much any design. Bottom hinges should be used on the flaps if present. For the flaps, the bevel can basically be 0 so it won't impact the aerodynamics at all when they're up. The elevator should use top hinges if large amounts of authority are needed and bottom hinges to reduce drag (the elevator is basically an upside-down wing so the conclusions are reversed).