Designing wings for a high performance micro-sub

[SubmarinerTom]

I have researched for close to a year a very special project and for the life of

me can't get my head around what wing shape to use for my micro-

submarine. Also known as control surfaces, they need to be low drag.

Throw everything you think of conventional submarines out the window.

Most of it does not apply in my design. It is unconventional to say the least.

Think of it as a submarine you wear. A manned torpedo if you will. I have

extensive marine experience in a multitude of disciplines. Jack of all trades,

master of none. What I am looking for is fun, nothing more. No resale, no

drug smuggling, no bombing, no commercial gain, no advice on how to build

or safety issues I need to be concerned about. I need advice on wing shape

AKA control surface design. For this I will be grateful beyond your wildest

imaginings.

The vessel is a NACA 0017 revolved, a tear-drop. ~20 inches at it's widest,

~10 feet long. 500 Newtons of drag at ~15 knots (8m/s). 20"X20" 2-blade

contra-rotating propellers driven by two, 4kW PMSM motors (hollow centre

for "solid shaft in tube" counter rotating drive shafts). One hour dive time

with one hour reserve. Service depth of 30 feet sea water. Delayed ambient

cabin pressure from +/- 1 atmosphere. 42 sealed lead acid batteries.

Displacement is ~600 pounds, buoyancy is -3 to +3 pounds of neutral (the

variable weight of an 80 cubic foot scuba tank of compressed air).

My objective is to fly in the water, in a spirited manner. The trailer is the only

way in and out of the water except in emergency of course, but this would

mean losing the vehicle (vessel, same thing).

I have considered and researched every wing shape possible and can't come

to any definitive conclusion. I would like to employ independent bow planes

1/3 aft of the bow (at the widest point), a full rudder and no elevators.

You may think this thread more appropriate in the marine section but I have

exhausted my marine references and have found them to be, like me, over-

challenged by this design endeavour. I am hoping some aeronautical know-

how will prove fortuitous.

Tom

Victoria BC Canada

 

[MikeHalloran]

I could tell you, but then I'd have to ....
... no, I'd have to kill you before telling you.


Seriously, you'll probably end up with a relatively thick foil, just because of the loading that's possible on a control surface when you really, really want it to produce some lift. You have the option of using a thinner solid foil, but that adds weight on the trailer.

I'm not sure how you're going to induce any pitching moment if you put the only horizontal planes anywhere near the center of pressure of the hull. You might at least want some fixed horizontal planes near the propeller.

You might want to look at the Albacore's story:

http://ussalbacore.org/html/albacore_story.html

Mike Halloran
Pembroke Pines, FL, USA

 

[GregLocock]

Underwater profiles are a lot less finicky than aero ones. i can't really explain why (for example, the gains in going to true teardrop shape instead of a torpedo shape are very slight), or perhaps more relevantly, the shape of the blades (both in planform and cross section) on a screw have remarkably little effect on its performance or efficiency.



Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[insideman]

Adding to what Greg locock wrote, one reason design for under water is less finicky than design for air is that water is far denser than air, making Reynolds numbers lots bigger.

 

[rb1957]

but speeds lots slower, so that'd be something of a wash.

2c ... thick = dragy, thin = small LE radius (separation)

thought ... why control with foils (if this is so unconventional) ? why not put a universal joint on the prop shaft (steer by vectoring thrust, like a sea-doo)??

 

[SubmarinerTom]

Mike,

You make me laugh, thank you.

Weight on or off the trailer is not an issue.
I actually need the weight to make buoyancy so I'll use steel framed control surfaces with molded contouring and Fiberglas finish.

I'd like to combine elevators and forward dive planes but control is a real challenge without fly-by-wire. Achieving pitch and roll response from the same levers is difficult. This project relies on simplicity.

The Albacore site is a good one isn't it. I discovered it long ago.


Greg,

I read through your SIG (what ever that means). My appologies for posting a DIY. I didn't realize it was against the rules but I understand, sorry.


insideman,

Interesting handle. ~800 times more dense in fact, thanks.


rb1957,

Small leading edge radius minimizing seperation. Interesting.

Vectored thrust is attractive but too complicated with "shaft in tube" contra-rotating screws. I'd have to gimbal the motors and they are close to a foot In diameter.


So, what I'm hearing is it doesn't really matter what shape I give my wings as long as the aspect ratio is somewhat favorable.

I'm not sure I'm willing to accept that. Perhaps I need to make interchangable wings so I can test a few.

Where would YOU start?

Thanks again,

Tom

 

[MikeHalloran]

The wings have to be replaceable anyway, because you're sure to hit something with them. The boat is not human- powered, so drag is really just a range problem. The wings are the least of your worries; just eyeball 'em and move on.




Mike Halloran
Pembroke Pines, FL, USA

 

[rb1957]

actually i meant a small LE radius leads to separation (which is a bad thing). i was trying to make a point that a thick section is draggy but to have a generous LE radius (for good control of separation) then you need a thick section ... typical design lose-lose situation.

making a set of interchangable foils is definately the way to go. i'd start with ol' reliable NACA 0012. i don't thin k you need to worry about aspect ratio, most sub wings are pretty squat.

one control to do all is ok ... consider a single side stick on A320, F16, etc. they're "easy" to use, but complicated to design, particularly if you've got a mechanical (fly-by-rods) system. i don't think two controls would be unmanagable for most operators ... one of pitch and speed (trigger), one for roll.

 

[KENAT]

Are you talking hydrofoil section or planform?

Regarding section, I'd take a look at what torpedo's and the like use. Of course they may well compromise because of needing to tube launch/air carried, cost etc. but it will give a start point. I'm guessing you've already looked at regular submarines for inspiration. It's been a while but they were fairly thick as I recall. Then again, their size/drag in proportion to the overall sub size is small, where as for your vehicle it's more significant. Push comes to shove a flat plate generates lift at the correct angle of dangle, it's things like control of separation and minimizing drag that leads to the typical modified tear drop section.

Regarding planform, theoretically ellipse is best from purely aero (and I assum hydro) dynamic reasons. However, it has a significant structural penalty which is why it's rarely used on aircraft. (The Spitfire supposedly used it primarily for reasons other than aerodynamics, and was relatively a pig to manufacture.) The planform of a typical wing is a compromise between a number of factors including aerodynamics, structural, ground handling, things that have to be fitted in a wing (landing gear, cabling, hydraulics, weapons, engines, fuel...) and probably some I've forgotten.

However, for you mass may be less of an issue so you could play with an ellipse, but my guess is because of the ratio of wingspan to body that interference effects will be more significant than the theoretical performance of the wing.

So, if it was I, I'd start with simple rectangle just to make it easy to manufacture.

If that doesn't suit, I'd go for tapered with straight leading and trailing edge. If you make 0° sweep at approx 25% chord then you can maybe use a single spar directly going into your pivot, the guided weapons I've dealt with did similar.

Or, you could take a leaf from the book of Russion Missile designers and use a 'lattice' or 'grid' plan form

http://en.wikipedia.org/wiki/Grid_fin.

This may have some advantages in reducing the force required to angle them, and if you can come up with a decent manufacturing technique, might have structural advantages.

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[SubmarinerTom]

Mike,

Thanks. Drag is a max velocity, motor heat and battery loading issue for me so I would like to minimize it.

Wings / controll surfaces... "Eyeball 'em"? I was really hoping for a more calculated approach.


rb,

Thanks for clarifying. Copying convensional subs is easy, again I was hoping for a little more educated approach.

Two sticks works for me. I was hoping to tie the ailerons in to elevator control to enhance pitch response, but it's just too complicated.


KENAT,

Planform.

Wow, hadn't seen "lattice/grid" before, thanks.

Lots of interesting points, thank you, Ellipse should fit my needs. I'll need to be fixed until ~25% out and then fully pivot the remaining section perhaps minimizing the interference effects you referenced. On the aft planes 100% will pivot. It's a point-of-purchase issue on the forward wings as there is no support inside the 20 inch fuselage. I'd love to post an image but can't...

Tom

 

[rb1957]

without being a SA ... do you have access to a swimming pool or somewhere to test scale models ?

the foils on a sub are pretty small (compared to wings on a plane) ... how significant are they to the overall drag ? speeds are pretty low so induced drag effects are small (again compared to a plane).

ailerons would have very little benefit in pitch control. if you really want have fore planes and aft planes (working in opposite directions of course).

why counter-rotating props ? yes, i know they counter-act each other's torque. but you lose the possiblity of vectored thrust and you have a pretty complicated mechanical set-up. you could react the torque by setting trim angles on the foils. why a prop ? why not a water-jet ??

 

[KENAT]

While we aren't really meant to help DIYers this is the kind of post that gets my creative juices flowing, although trust me that I'm no expert in this area so take anything I say with a pinch or bag of salt.

So a couple of more thoughts.

God/Mother Nature/Evolution whatever you choose to think of it has come up with a number of underwater vehicles of approximately the same size and been optimizing them for a long time. So, I'd perhaps look for inspiration toward aquatic/marine life especially mammals. Again they may be compromised by other requirements but you could at least think about it.

As regards thrust vectoring, you don't necessarily have to vector the propellors. You could place control fins directly behind the propellor. This is effectively what aircraft such as the F22 and late model Sukhoi's do. It's also how hovercraft are steered and how some torpedos, including the manned variety are steered

http://en.wikipedia.org/wiki/Human_torpedo.

I assume you've looked at ROV's for inspiration as well as some of the existing one man submersibles etc.

http://www.homebuiltrovs.com/

is probably far too basic and you've probably seen it before but shows there is stuff out there.

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[SubmarinerTom]

Okay, you gotta see my drawings. When I said ailerons, I meant forward dive planes. They'd work great in unison with elevators for crisp pitch control.

Perhaps I should make a web-site for this project because there seems to be genuine, knowledgable interest in it but I just seem to spin my wheels on every forum I visit and the conflicting opinions frustrate me to no end.

Yes, I have access to a test tank/pool. What scale would you suggest.

My contra-rotating set-up is amazingly simple. With hollow centre, flat PMSM motors, a solid shaft turns inside a hollow, counter-rotating, tube.
Redundancy is a huge reason as well as cancelling torque and dirty wash.
Vectored thrust from 500 Newtons isn't much. From a jet drive it would be 40% less due to their inefficiency. There are a couple of other reasons which are difficult to explain here.

Yes, I've seen all those web-sites, thanks. I really think I've missed something in my research. I have about 22 pages of design evolution and have come full circle a number of times in a number of different ways.

I should have studied Aeronautical Engineering. Who knew...

Here's a web-site that is bang-on what I'd do if I had unlimited funds:
aaarrrrggg, I can't find it. Google "Cliff Redus" R-300 submarine,
you wont be disapointed.

Sorry, no time to correct my spelling, dinner's ready...

 

[KENAT]

http://www.psubs.org/pic/dr_redus.html

what you're talking about? It's easy to make a pretty CAD model, doesn't always translate to anything that works worth a damn for a reasonable price. Not saying that's the case here but beware of it.

Based on my limited experience with home made inventions, I'd go for KISS.

The immense satisfaction you get from getting a working prototype, even if it is incredibly crude and only lasts a short time etc. before failing, is significant. It also allows you to discover at least some of the flaws before coming up with the gold plated version.

Plus often times that last 10% of performance or so takes 50% or the effort or something like that. So I'd say when you're paying the bills/giving up time initially go for a 75% solution, then refine it time & resources permitting.

Otherwise, this may be one of those great ideas that never gets finished.

My guess is that in the performance range you're talking about, the difference in planform will be barely noticeable, kind of like the difference between true tear drop and approximated tear drop Greg mentioned. Even the aerofoil section may not make much difference.

You say the thrust from vectoring isn't much, what are you tring to achieve with your dive planes? L = Cl * A * .5 * r * V^2

http://wright.nasa.gov/airplane/lifteq.html

should allow you to very roughly size them.

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[waross]

How about ducted props with two sets of rudders at right angles?

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[MikeHalloran]

Stop focusing on perfecting the machine on paper; you can't possibly foresee all the problems you'll find, so go find 'em.

Build the machine, launch it, test it, measure it, document it, put it in storage, and make the next one better.



Mike Halloran
Pembroke Pines, FL, USA

 

[CH5OH]

if you have well balanced counterprops,is there any need for bowfins (if battery weight/cabin air is in good ratio)? Why not mount props in steerable nozzle?