Student needs help with TANK TESTING

[jamesflett]

Im a BE Mech student doing a 2course project on planing hull velocities and planing efficiency (time takes to plane).

At this stage we were thinking of two options -
i) using a pulley system in a swimming pool (most likely due to time
constraints)
ii) constructing a circular tank with a rotating arm on
connected to the model

- The issues that I was hoping someone could help me with are
i)using correct scaling laws

ii)whether the size of the models will be substantial enough to measure
and how to go about measuring it.

iii)measuring of the planing characteristics:two options
1) I have thought about either visually determining when the
hulls are planing but thought this may be difficult if there is
only small differences given the size of the models.

Another option is to attach strain
gauges to the inside of the model hull and record the forces
and pressures that are acting on the model and thus determine at
which point it will be planing graphically.
Our scale is 1:10, so the model lengths will be around 600-700mm
waterline length.

Test Conditions

[jamesflett]

Test Conditions will be:
in calm water
using 2 different hulls, variable deadrise (20-33 degrees) and constant v (20dgrees)
hulls will be made from a 3D printer which will be hardened using epoxy

[Tim B]

The normal method to determine whether a boat is planing is to measure pitch and heave. It is usually obvious to determine whether the boat is planing or not. However, I haven't heard of much testing on the accelleration properties of planing hulls.

You might have to run a camera down the "tank" at the same speed as the model, and exactly at 90 degrees to the direction of travel, then pick out the points from each video frame

Hope This Helps,

Tim B.

[jehardiman]

Tim B is correct. The method you proposed will only measure drag and forces and will not allow the hull to naturaly step up on to plane (i.e. the rise and rotation of the hull to the point at which the hull is being primarily supported by hydrodynamice forces vice hydrostatic forces). To measure the inception of planing you will need a "cricket" with linear/rotary indication as well as strain gages.

[jamesflett]

Thanks guys for the prompt input. Very much appreciated.

Could you both please elaborate a little more on the suggested options (Jehardiman on the "cricket" with linear/rotary indication) and Tim B on the pitching and heaving and how to measure it. Our university is relatively new and I am studying just Mechanical Eng. and we have no courses associating with marine engineering and none of lecturers have had experience in the marine industry domain. I am however looking at a career in Naval Architecture after my degree hence the reason for the project which I derived for myself....!

Thanks

James

[Willallison]

you might like to take a look at a couple of articles published in the last 12 months or so in Proffesional Boatbuilder magazine. ( www. proboat.com )
There have been a number regarding tank testing, but one method that may be appropriate for what you are looking at would be that published (I think) by Donald Blount where he proposed using tow testing from a boom (off a real boat...).
This tends to be cheaper and somewhat simpler than true tank testing. You could mount a video camera on the side of the boat, directly opposite your model in order to obtain the point at which planing occurs.

[jehardiman]

Could you both please elaborate a little more on the suggested options (Jehardiman on the "cricket" with linear/rotary indication) and Tim B on the pitching and heaving and how to measure it.

In the simpilest of tow systems, the model is connected to the tow carriage towing arm by a simple swivel sufficiently forward in the hull to enable tracking and the model is allowed to freely roll, but is abnormally constrained in heave,sway, pitch and yaw, and totally fixed in surge. In better systems, the model is connected to the tow arm by a long towing rod which allows the model more freedom of movement, but requires proper placement of the tow point (at the expected CG for the run) and course stability for the model. Both of these systems only record resistance directly. All other behaviour must be qualitavily determined by inspection of model response.

A "cricket" (or grasshopper, or double diamond link) is a double link system that fixes the model in surge, yaw, and sway; while leaving it completely free in heave, pitch, and roll. Buy use of rotary/linear indicators on the links, not only can you get powering data, but pitch and heave data also. This is the common method for towing high speed models when directional stability is not a given.

[marshmat]

Some good ideas posted so far as to the tank setup. As to the model-on-boom-beside-real-boat, this would be great in calm water but keep in mind that waves will mess up your results.
As for scaling, any fluid dynamics textbook that covers dimensional analysis will describe the basic math. For a surface vessel, the Froude number is usually the most critical, followed by the Reynolds number. There are many others too, that must be either matched or deemed irrelevant. Generally, the bigger the model, the closer the test results will be.

[MikeJohns]

James
Auckland or Cantab.?

Unless you have access to a towing tank it's going to take you longer to build the test rig than your time will allow.
For your project IMHO you are best advised to power a model and radio control it. From vidoe frames processed on a PC you can calculate parameters sufficient for your requirements.

Hope this helps

[jamesflett]

Thanks Mike, Im in Auckland.

Radio control is an option we have thought of as it would help produce a more realistic propulsion due to trim, however had initailly not considered due to budget costs. We have now been informed that we have a floating budget of around $400NZ so it could be a viable option. Who are IMHO?

Cheers

James

[marshmat]

IMHO = shorthand for "In my humble opinion...."

[jamesflett]

Haha. My apologies. Thanks for the advice again

[Willallison]

some of us spend way too much time around here...:D