FLOTILLA: Hovercraft Hydrodynamics

[Leo Lazauskas]

Some boatdesign.net boffins might like to play with the attached demo
program that calculates the resistance and wave patterns of ACV.

INSTRUCTIONS
============
To install the program, uncompress the zip file into a suitable
directory. To uninstall, just delete the entire sub-directory.

Batch files are used to run the examples. These batch files copy the
appropriate input files from the "inp" directory to the "tmp"
directory so that the executables in the "bin" directory can find them.
Output is written to the "out" sub-directory.


EXAMPLE 1: Model ACV (Length = 2.07m, Width=1.372m)
===================================================
This example can be used to verify predictions against Everest and
Hogben's towing-tank experiments with a small model ACV.

EXAMPLE 1a: Wave Resistance in Deep Water
-----------------------------------------
Run the batch file "acv_ex1a.bat".
(The program takes about 10 seconds on a 3.2GHz PC.)

The "acv_cnp.csv" file contains two columns headed "F" and "C_NP":
F is the length-based Froude number; C_NP is the (Newman-Poole)
wave resistance coefficient. Plotting the results should produce
a curve identical to that in the top plot of Fig. 1 in the attached
note.

EXAMPLE 1b: Wave Resistance, Finite Depth (h/L=0.215)
----------------------------------------------------
Run the batch file "acv_ex1b.bat".
Plotting the "acv_cnp.csv" file should produce the curve in the
bottom plot.

MORE TO FOLLOW...
=================
Some further examples will be posted below and I will try to add more
when I get time. Suggestions for "test cases" are most welcome, but
please post them here.

Leo.

[Leo Lazauskas]

EXAMPLE 2: A small 3.5m long, 2.0m wide ACV, 400kg all-up weight.
=================================================================
Above-water quantities (in "abovewater.flo") are used to estimate
the air resistance: the "superstructure" is just a rectangular block
used to simulate the skirt; the "deck-housing" is another block that
has roughly the same dimensions as two people sitting one behind the
other. The combined frontal area of the two blocks is used as input
to the air resistance calculation. "Rough as guts" would be an
apposite description of this technique.

EXAMPLE 2a: Resistance in deep water
------------------------------------
Run the batch file "acv_ex2a.bat".

The "acv_drag.csv" file contains six columns:
U is the speed in m/sec,
R_W is the wave resistance, R_A is air resistance,
R_M is the momentum drag, R_L is the "equivalent lift", and
R_T is the total resistance.
For an explanation of these components, see:
"Hydrodynamics of advanced high-speed sealift vessels",
http://digital.library.adelaide.edu....dle/2440/37729

The resistance components are plotted against speed in knots in the
attached note. Note that nominal values have been used for the skirt
clearance and above-water portions of the vessel. Changing these values
in the input files could results in quite different equivalent
lift, momentum resistance and air drag.

EXAMPLE 2b: Wave Pattern
------------------------
Run the batch file "acv_ex2b.bat".
The program takes about 70 seconds on a 3.2GHz i7 PC. (Only the port
half of the wave pattern needs to be calculated because it is
laterally symmetric.)

Ship speed is 3.3 m/sec, which is close to the hump speed.
After the wave pattern has been calculated, the "acv_waves.csv"
file is displayed using the program "acv_plot".

Press 2 to see the pattern with a different colour palette.
Press 1 if you want to return to the blue wave pattern.
Press the Esc key to exit.
The pattern is saved to the standard output file "acv_wp.px".
The batch file copies this file to "acv_wp_ex2b0.pcx".

The wave pattern is now displayed in another format.
This graphic (which has axes and a legend) is copied to
"acv_wp_ex2b1.pcx".
Press Esc to exit.

The patterns should look the same as the attached graphics.
You can change them to a different graphic format using one of the
many freely available programs on the internet.

Leo.

[Leo Lazauskas]

EXAMPLE 3: Non-Uniform Cushion Pressures
========================================
The non-uniform cushion pressures in this example are the same as in
Fig. 8 of:
"Free-surface pressure distributions with minimum wave resistance",
ANZIAM Journal, Vol. 43, 2001.
http://www.cyberiad.net/library/pdf/tl01.pdf

EXAMPLE 3a: Resistance in deep water
------------------------------------
Run the batch file "acv_ex3a.bat".
(The program takes about 45 seconds on a 3.2GHz PC.)

The wave resistance coefficient at F=0.71 is less than half that
of the uniform pressure case, however, it is not known how such a
pressure distribution can actually be achieved in practice.
At other speeds, the non-uniform distribution does very poorly.

You can see the pressure distribution by plotting the data in the file
"acv_pressures.csv".

Leo.

[Leo Lazauskas]

Attached is a short note comparing Flotilla wave resistance
predictions in deep and shallow water with experiments and
some very recent CFD computations.
(See Examples 1a and 1b above).

Flotilla produces similar results to the CFD code, but about
4 million times faster. To be fair, the (non-linear) CFD code is
capable of much more than our simple linear code but those
enhancements might be more than required during the
preliminary design phase.
Moral: Use the appropriate tool for the job.

Leo.

[DCockey]

Thanks for the posts. I missed this thread when you first posted. Very interesting and nice work. I agree that simple analysis tools are often the most appropriate tool for preliminary design. Seems that frequently folks want to use a multi-axis, multi-head CNC mill for tasks where a simple hand drill would be better suited.

Any thoughts on using a similar approach for "shallow" boats with flat or draft much less than length? Can the amplitude be determined directly from the submerged shape or would iteration be needed?

[Leo Lazauskas]

Originally Posted by DCockey

Thanks for the posts. I missed this thread when you first posted. Very interesting and nice work. I agree that simple analysis tools are often the most appropriate tool for preliminary design. Seems that frequently folks want to use a multi-axis, multi-head CNC mill for tasks where a simple hand drill would be better suited.

Any thoughts on using a similar approach for "shallow" boats with flat or draft much less than length? Can the amplitude be determined directly from the submerged shape or would iteration be needed?

Thanks, David.
Yes, it is possible (but very difficult) to find a planing hull that creates the same shape and, yes, it requires iteration and some other considerations.
I am working towards it, but it took me 6 months to correct errors in some older code by my colleagues. I ended up rewriting it from scratch, so don't expect a full planing code any time soon

The next step is to release a program for Surface Effect Ships (SES) and then, if I am on the dole after the new year with lots of free time, to move onto the planing problem.

Leo.

[CWTeebs]

This is most interesting, accomplished work Mr. Lazauskas (I'm trying to digest "Free-surface pressure distributions with minimum wave resistance")

[Leo Lazauskas]

Originally Posted by CWTeebs

This is most interesting, accomplished work Mr. Lazauskas (I'm trying to digest "Free-surface pressure distributions with minimum wave resistance")

Very kind, Mr Teebs!
There is a little more on that topic in:
http://www.cyberiad.net/library/pdf/tsl02b.pdf

Leo.

[CWTeebs]

Thanks Dr. Lazauskas, I love this work. I've become fascinated with the different hydrodynamics methods out there since I interned for a Naval Architect (I did programming on the GUI for the planing hull program POWERSEA).

Wish I had the brain required to be an applied mathematician *bows*.

EDIT:
You may find this thesis interesting "A DRAG ESTIMATE FOR CONCEPT-STAGE SHIP DESIGN OPTIMIZATION."
http://library.umaine.edu/theses/pdf/ReadD2009.pdf

[Leo Lazauskas]

Originally Posted by CWTeebs

Thanks Dr. Lazauskas, I love this work. I've become fascinated with the different hydrodynamics methods out there since I interned for a Naval Architect (I did programming on the GUI for the planing hull program POWERSEA).

Wish I had the brain required to be an applied mathematician *bows*.

I have more fun applying applied mathematics and writing programs than
scribbling squiggles.
I was very lucky to have had E.O. Tuck as a colleague and mentor: he did
the tough maths and I sat around coding while watching B-grade scifi at home.

[CWTeebs]

What coding language do you use? I've been trying to figure out how to add a free surface condition to an old aerospace panel code.

[Leo Lazauskas]

Originally Posted by CWTeebs

What coding language do you use? I've been trying to figure out how to add a free surface condition to an old aerospace panel code.

C/C++, Fortran and very occasionally some assembly code for graphics.

I'm still using Turbo Pascal (2.4?) for Vertical Axis Wind (and Water) Turbines but
I need to rewrite that soon because it doesn't work with 64 bit Windows 7.

[Leo Lazauskas]

Originally Posted by CWTeebs

EDIT:
You may find this thesis interesting "A DRAG ESTIMATE FOR CONCEPT-STAGE SHIP DESIGN OPTIMIZATION."
http://library.umaine.edu/theses/pdf/ReadD2009.pdf

Thanks. I read this a while ago.
Douglas (and Bath Iron Works) were one of the first to donate money
to Michlet well over 10 years ago.
Like Read, I thought about using the modification to Michell's work by Amromin et al but Tuck convinced me that it wasn't consistent and that other avenues (e.g. including squat and paying attention to transom stern effects) would be better.

Leo.

[CWTeebs]

Small world, he was one of my professors