Rapidly deployable personal aerodynamic decelerator

[JaredJames]

And I would really like to talk to you real-time if possible. Please PM if you have any faster ways to communicate, even by phone.I'm sure you know a lot more than I do, but we're just having some communication problems.

I've PM'd my email to you twice.

I will be signing off soon though as I have lectures at 9am in tomorrow. I am on every night for a few hours to catch up with PF.

 

[jgeating]

By parachute I mean anything that produces a substantial amount of drag, in this case the wing. I'm not sure what the formal definition for flare is, but for the landing portion of the flight, the angle of attack will increase so that the vertical forces are always in static equilibrium and the pilot is skimming across the ground, decelerating horizontally.

Also, I think it is rare, if not dangerous for any hang glider to have a negative angle of attack. If the net force on the wing surface were ever negative, the wing surface would fail because it's just fabric.

 

[JaredJames]

By parachute I mean anything that produces a substantial amount of drag, in this case the wing.

Once the wing stalls, it won't be producing that much drag. If you're not on the ground soon you'll be in trouble. Wing "as a parachute" or not.

I'm not sure what the formal definition for flare is, but for the landing portion of the flight, the angle of attack will increase so that the vertical forces are always in static equilibrium and the pilot is skimming across the ground, decelerating horizontally.

Landing =/= flare.

Watch the video, you'll see that as he comes into land he gently increases the AoA. The flare is the last little bit where the AoA increases to maximum rapidly and he lifts slightly. After that he descends gently a very short distance to the ground.

It is that motion shown for the last 30 seconds at the end of the video that you need to replicate in order to land safely.

Also, I think it is rare, if not dangerous for any hang glider to have a negative angle of attack. If the net force on the wing surface were ever negative, the wing surface would fail because it's just fabric.

Watch the video, it's negative just before landing.

If there's no headwind, the "net force" on the top of the wing will be very little if anything at all.

 

[mugaliens]

I am not looking for anything supernatural, just something that one could jump out of a truck moving say 40mph, or off the top of a 20 ft building, and land safely with.

A normal parachute or ram-air foil would do the job off a truck in the Moon's gravity (assuming 1 ATM pressure), but not here on Earth. It wouldn't work in 20 ft, even here on Earth, as you'd never reach enough velocity to properly open the chute. For that you'd need something more exotic, such as a parachute with some sort of explosive opening device, such as that used to inflate an airbag. Perhaps a sqib-fired expanding gas system and a series of tubes to achieve full parachute inflation in 0.25 seconds or less.

Even so, falling from 6' at 40 mph horizontal, you'd probably still be doing at least 30 mph by the time you dropped 6' to the pavement. Quite a road rash.

 

[RandomGuy88]

Again, my original question is still what concerns me most. Can you give me an answer:
1) What is a good scalable fabric to manufacture an airfoil with to test in a small tunnel?
2) What is a good method to manufacture a controllable airfoil on a small scale, preferably using aluminum pipes, fabric, string, and other easily accessible materials? I have access to a machine shop. I would prefer not to have to CNC.

An easy way to make an airfoil quickly and accurately would be with a template, a hotwire cutter and some foam. This works really well assuming you can get a decent template. I use a water jet cutter, a CAD file with the airfoil and some plywood. You could also try making some ribs out of wood, running a spar though them and then covering them with a material like mylar. I don't really know how you would make an airfoil with fabric and string. How do you plan on giving it any thickness?

Also, if you plan to have something that is rapidly deployable and easily carried I don't think an airfoil with any thickness is really what you are looking for. I believe it was mentioned earlier but at these speeds a flat plate of some stretched out fabric would probably be just as effective. For example the kitewing that you referenced earlier does not have an airfoil shape. There are some flexible rods in there to give it camber but there is no thickness. What kind of airfoil profiles are you looking at.

 

[jgeating]

A normal parachute or ram-air foil would do the job off a truck in the Moon's gravity (assuming 1 ATM pressure), but not here on Earth. It wouldn't work in 20 ft, even here on Earth, as you'd never reach enough velocity to properly open the chute. For that you'd need something more exotic, such as a parachute with some sort of explosive opening device, such as that used to inflate an airbag. Perhaps a sqib-fired expanding gas system and a series of tubes to achieve full parachute inflation in 0.25 seconds or less.

Even so, falling from 6' at 40 mph horizontal, you'd probably still be doing at least 30 mph by the time you dropped 6' to the pavement. Quite a road rash.

That's exactly why we're not using a parachute design, but rather a glider. The purpose of this project is essentially to find a parachute alternative. Again, we're not focusing on the deployment, but it shouldn't take more than 1 full second.

Also, for the truck it is the air that would be lifting you out once you deploy. There would be no reason to jump then deploy if there is enough clearance to deploy prior to jumping.

Same goes for the building, so long as there is deployment clearance, the wing opens up first, then you jump, or you do them simultaneously.

The main reasons we believe this is feasible is because it is NOT like a parachute, meaning the requirements for a safe parachute deployment and landing do not apply.

 

[JaredJames]

Im on my phone so I'll be brief.

1. It takes time to get to required airspeed (re jumping off buildings) so there are height restrictions.

2. A glider needs a significantly longer landing run - approach and landing - a parachute can land fairly accurate in a small area. The hang glider must maintain velocity otherwise it crashes, so again there will be a restriction based on required land space required.

 

[jgeating]

An easy way to make an airfoil quickly and accurately would be with a template, a hotwire cutter and some foam. This works really well assuming you can get a decent template. I use a water jet cutter, a CAD file with the airfoil and some plywood. You could also try making some ribs out of wood, running a spar though them and then covering them with a material like mylar. I don't really know how you would make an airfoil with fabric and string. How do you plan on giving it any thickness?

Also, if you plan to have something that is rapidly deployable and easily carried I don't think an airfoil with any thickness is really what you are looking for. I believe it was mentioned earlier but at these speeds a flat plate of some stretched out fabric would probably be just as effective. For example the kitewing that you referenced earlier does not have an airfoil shape. There are some flexible rods in there to give it camber but there is no thickness. What kind of airfoil profiles are you looking at.

Yeah, we can't do anything with thickness here due to both deployability and weight. From my understanding, kitewings (and hang gliders for that matter) do have airfoils. The initial air that starts flowing under the wing is in turbulence, which acts as somewhat of a wall for the incoming air. If kind of "fills in" the empty space underneath, to some extent at least. Furthermore, the pressure of the air underneath gives it its shape.

However, it is these "single surface airfoils" that I cannot find any practical information on. Kitewings are made out of dacron, so I'm thinking that's what we'll use, but then the dacron thickness won't be proportional, let alone react to the scaled wind appropriately in terms of flexibility/stretch, fluid flow, etc.

In terms of manufacturing, I was thinking to wrap the fabric around the leading edge pole, possibly fasten it back 25% or so of the chord (terminology?), to give a slight double surface. They do this in hang gliders nowadays. To adjust the tension, and therefore many other properties, the trailing edge could be pulled tighter or looser by string. battens would be nice, but we'll probably forego them for now.

Thanks for your help. Please let me know any other information you are aware of or come across.

 

[jgeating]

Im on my phone so I'll be brief.

1. It takes time to get to required airspeed (re jumping off buildings) so there are height restrictions.

2. A glider needs a significantly longer landing run - approach and landing - a parachute can land fairly accurate in a small area. The hang glider must maintain velocity otherwise it crashes, so again there will be a restriction based on required land space required.

This was actually a big discussion we had; whether or not there will end up being a "dead range" or height interval in which the wing cannot be used. Although airspeed is required for flight, that does not mean the wing cannot be used as a parachute. e.g. a stalled hang glider can still descend to the ground stably several feet, if not more (thankfully I haven't figured out the upper limit to this). For now, our main focus has shifted to whether it can land from an "infinite height" i.e. terminal gliding velocity, so we're not really worrying about the dead space. Definitely will show up in the future work section though.

Let me know what you think about our current plan, and why it won't work:
Once we build the airfoil, we will generate a n*y*2 matrix containing all the data to simulate a landing. N corresponds to all the AoA we will be testing. Y corresponds all the velocities we will be testing. 2 corresponds to the X and Y forces. We might also be able to get Z moment, but probably won't implement it. We will do a simple step integration/approximation using MATLAB via force balance to get acceleration and therefore derive velocity and position.

 

[RandomGuy88]

However, it is these "single surface airfoils" that I cannot find any practical information on. Kitewings are made out of dacron, so I'm thinking that's what we'll use, but then the dacron thickness won't be proportional, let alone react to the scaled wind appropriately in terms of flexibility/stretch, fluid flow, etc.

What kind of thickness are you talking about here? I am assuming the thickness of this stuff is the same order of magnitude as paper. You aren't going to be able to scale that down and there is no reason to, it won't affect the aerodynamics.

 

[RandomGuy88]

Let me know what you think about our current plan, and why it won't work:
Once we build the airfoil, we will generate a n*y*2 matrix containing all the data to simulate a landing. N corresponds to all the AoA we will be testing. Y corresponds all the velocities we will be testing. 2 corresponds to the X and Y forces. We might also be able to get Z moment, but probably won't implement it. We will do a simple step integration/approximation using MATLAB via force balance to get acceleration and therefore derive velocity and position.

You may not be able to apply your force balance data directly to this problem because of the unsteady aerodynamics. For example, look up dynamic stall. Your static airfoil in the tunnel will not have the same stall characteristics as your wing undergoing a rapid pitch up maneuver.

 

[jgeating]

Thickness: I had it in my head that dacron was more of a thick plastic-like material,a good couple millimeters thick. Not sure why,maybe the name, or because there are transparent variants. It look's like it's actually negligible like you said, so we'll probably just ignore it.

Good point about the dynamic stall. Could you advise on these options?

1. Ignore completely
2. Some sort of equation to approximate dynamic stall effect as a function of our known data (wing properties/shape, forces at given AoA and Vel)
3. Test experimentally (flare in wind tunnel and record forces at each dt, although for scaling,I imagine our flare would have to be super fast)

Lastly, do you know how to do proper scaling using dimensional analysis? or is that navier stokes to determine scaling properties. I'm pretty sure that our velocity wouldn't just be the same ratio as our length differences. What is the proper way to calculate this?

 

[RandomGuy88]

Thickness: I had it in my head that dacron was more of a thick plastic-like material,a good couple millimeters thick. Not sure why,maybe the name, or because there are transparent variants. It look's like it's actually negligible like you said, so we'll probably just ignore it.

Good point about the dynamic stall. Could you advise on these options?

1. Ignore completely
2. Some sort of equation to approximate dynamic stall effect as a function of our known data (wing properties/shape, forces at given AoA and Vel)
3. Test experimentally (flare in wind tunnel and record forces at each dt, although for scaling,I imagine our flare would have to be super fast)

Lastly, do you know how to do proper scaling using dimensional analysis? or is that navier stokes to determine scaling properties. I'm pretty sure that our velocity wouldn't just be the same ratio as our length differences. What is the proper way to calculate this?

Unfortunately I do not know much about dynamic stall other than a general description of how it works and what its effects are. I do know that it is a very non-linear process and is not easily analyzed and you probably won't be able to use data from a static test because the flowfields are completely different. Dynamic wind tunnel testing is pretty difficult compared to static testing and to be honest I don't think you will be able to do it. Equipping your wind tunnel for that would be a big project on its own. Then, even if you did design a rig to rapidly pitch your model you would still not be simulating the fact that your velocity will change dramatically during the flare.

As for scaling, if you are just doing a static test at these low speeds you will really only need to match the Reynolds number in order to properly simulate the aerodynamics.

Re = V*L*rho/mu

rho=density
mu=viscosity

The density and viscosity can be assumed constant here so you will just need to make sure the product V*L is the same for the full scale case and the wind tunnel test.

 

[JaredJames]

Let me know what you think about our current plan, and why it won't work:
Once we build the airfoil, we will generate a n*y*2 matrix containing all the data to simulate a landing. N corresponds to all the AoA we will be testing. Y corresponds all the velocities we will be testing. 2 corresponds to the X and Y forces. We might also be able to get Z moment, but probably won't implement it. We will do a simple step integration/approximation using MATLAB via force balance to get acceleration and therefore derive velocity and position.

Let me be perfectly blunt here, based on what I've read here your basic aerodynamics understanding is sh*t. You made some rather major errors previously (I hope you understand them now and have made amendments?) and yet you come out with something like this (the above) which is a step in the right direction.

I have to agree with RandomGuy88 on this, you can't really simplify things this much as this and expect to gain anything particularly useful outside of a very wide approximation. There's a reason wind tunnel testing is still used over CFD.

 

[jgeating]

This is for MEM 220, a undergraduate Fluid Mechanics class, and we aren't expected to put in more than 5 hours each. I have already put in a good portion of that just on this forum. Most groups talk about how pressure increases with depth or do a dumbed down version of Navier Stokes. The fact that this is even remotely practical or useful is leaps and bounds above what is required for the course. That being said, however, most of my team is now complaining about having to build an airfoil and test it because none of the other groups are even doing anything experimental.

As a substitute, I said we could either (A) use a NACA airfoil from an online creator or something, or (B) find lift/drag data for a similar hang glider wing as ours.

As of now, (A) seems more likely, but NACA wings need to have a thickness entered, which ours does not have. You can enter the thickness as 0, but I believe that the data they give does not really account for how a single surface airfoil really works. Can anyone confirm or deny this? Could you supply some examples/links to get some data for the sims?

 

[S12mon]

I think for anything like this to ever be practical, you have to think of the deceleration of the subject as only half of the whole concept. As you brought up Batman early on, you would have to assume that the wearer has significant physical prowess. People who excel in parkour can jump a few stories and land with little to no injury. Now imagine someone with that capability, now wearing a device that slows there descent. Next, to assist in their safe landing, they are wearing a leg brace system, a human body "landing gear" of sorts. Making just a glider, is like a plane landing withingout landing without landing gear, it won't necisarily kill you, but the bottom of the plane won't be in too good of shape, and when the bottom of that plane is your legs, well that's an issue.

-Simon

 

[jgeating]

Well, your legs essentially are the undercarriage, and that's how hang gliders have worked for years. However, if you assume that the subject can handle a higher impact (probably a better word to use than that) velocity, that would make the design specifications a lot easier. However, the idea of altering the undercarriage (i.e. the legs) brings on a slew of complications. Wheel based defeats the purpose altogether. Some sort of dampening or shock absorbing system may be a little more understandable, but would require training (ok, no big deal), and more importantly, (IMHO) would reduce the level of interaction and feedback between the subject and the ground. For example, we still can barely get humanoids to run, and by isolating the user's feet from the ground, it may dampen the impact, but also increase the difficulty of maintaining stability and not falling or crashing as the subject lands. OK, a little off topic, but w/e. I had really been hoping to look more into the aerodynamics and scalability of a hang glider/wing suit hybrid rather than altering the problem statement.

 

[martha75]

Hello everyone. I have a solution but looking for lab to test the idea. If anyone knows of any, please let me know.
Thank You