SCALE MODEL , Size?

I plan on making a model to see how a "box boat" (built to ship in a std hi cube shipping container) will run.39ft loa x7.5ft beam.

It will on powering it with a small DC electric motor and with an ammeter & voltmeter see if the claimed efficiency of the Atkin Seabright box keel + Reverse Deadrise was all hype or is real.

The finished boat is to be 38ft , a 1 inch to the ft model would be easy to transport , but if a larger size is more useful , 2 inches?

Advice sought .

The model will power alongside a powered dink with a pole providing directional stability. Weight will be added and the results measured.

FF

 

Model testing is a science in itself and there is a lot of material available. You may want to read the ITTC recommendations carefully since it is easy to get fooled with models. Section 7.5 relates to modeling at:
http://ittc.sname.org/documents.htm

If you want meaningful results from this sort of resistance testing then the bigger the better, I would suggest that 6 feet OA is a good minimum length (1.8 m) for this sort of testing. You also need to simulate turbulence with scale models (there is a lot in this) .

cheers

 

Fred,

I'm not sure what you can learn from this testing. You can measure the power in, assuming you have telemetry link to see the results, but can you relate that to thrust and that to the thrust required to drive a "normal" model of the same hull measure and displacement? Real tank testing is way beyond us amateurs and even that produces poor results sometimes.

I see why you can't get any good results from a towing model of the Atkin tunnel. That is, if the "recaptured energy" theories expressed on the forum have any merit. Still, my thought is that the best result would be from having two models alike in all respects except the bottom shape. Both would be driven by the same propulsion unit. This would approximate the A/B comparison tests done on two models towed from a balance beam or individually with a drag measuring instrument.

Even this test would only give the result for the particular "normal" bottom shape chosen. There are a lot of normal shapes out there. It's probably easier to build a full scale boat.

Generally, I see most experimenters using towing tests for drag measurements and powered tests for handling characteristics.

 

Fred,

I would agree that with boat models size does matter, and bigger is better
At the molecular level water is not scaleable, easy to see that in those old movies with waves breaking on ships hulls and water drops the size of Buicks flying in the air.

Your idea is an interesting one, but to get any kind of meaningful results I think you would need to make two models. One with a Atkin/Seabright underbody, and one with a minimum wetted surface canoe type hull.

My gut feeling is that at low speeds (S/L<1.3) the canoe type would propably be superior. As speeds increase into the 1.5 to 2.0 range however the tunnel Seabright could show some advantage. What I'm thinking is that because B/L ratio has such a profound effect on waveform resistance, the fact that a large amount of the immersed hull section is very narrow could have a significant influence in wake reduction. Again, just guessing, but I'm thinking that as speeds continue to increase (S/L>3), the lower wetted surface of the simpler hull shape would again maybe be better.

As to the ability of the A/S hull to "surf on it's own wake" as Robb White said, well maybe :?: But I'm from Missouri on that one until I see some evidence. (maybe from your tests?)

moT

 

Scale models are good for making observations that can be applied to the full scale.

If you want to get hard data then controlling the test conditions can be challenging. Having another boat in close proximity could interfere with results. The depth of water could alter performance. It does not take much to set up radio controlled steering and this eliminates one source of test noise from your proposed method of steering with a tether.

Of all the methods I have used to predict boat performance I find Michlet miles in front.

If you have drawings of what you want to compare I could do comparitive runs at full scale and your chosen model scale to predict what will happen. This should identify the likely differences between the models and full scale over a range of speed.

For the models, I would at least have the battery pack interchangeable and do a series of timed runs over a set course by swapping the battery pack from one boat to the other. Ideally you would interchange the entire drive system but this might have to be different for each hull. Swapping the drives still does not eliminate error because the flow past the prop could be altered by the shape of the hull. Probably want to check that the battery conditions are identical for each run using the same instruments.

If Michlet is used to predict comparitive performance then you build one model and it aligns with Michlet you would have confidence in what Michlet predicts at full scale. Would save building the second model.

Rick W.

 

How do you know when to go metric scale?

FYI:
Hovercraft model in tow tank video, second to the bottom item on this page:
ACV Model Seakeeping test video
http://www.foils.org/mtgpapers.htm

 

You will probably need to build at least a two inch scale model in order to get to scale weight. It is difficult to build a small model that is light enough to simulate full sized weight.

A home spun way to estimate scale weight is to use the cube of the scale as a divisor for the full scale weight . At 2" to the foot, it is a 1/6 scale model. 6^3= 216 so if the big boat is, for example, 2160 pounds then the model can weigh 10 pounds all up. Try that with one inch to the foot and you have 12^3= 1728 and that model must weigh only 2160/1728 = 1.25 pounds. Little models will work, in this regard, only if your full sized boat is really heavy. You can select any scale you like but first you need to estimate the total weight of the model and go from there. Bigger is better for that reason too.

 

Messabout is right about the size limitation on weight for a model. I have used the 2":1' scale and it worked out ok. That gave a 48" model for me and allowed displacement of the model to vary from -20% to +80% of design displacement for evaluation with different loading. Unless your boat is heavy, a 48" model is almost the smallest that you can reasonably use. It is also the smallest that experts recommend for getting decent measurement results. I was able to build with 3mm plywood and meet the weight restriction. You can get aircraft plywood for lighter weight but it can get kind of flexible in the thinner thicknesses. Don't forget that some weights may be required to trim the model to your design static level.

 

Just tow two identical models (except for bottom configuration) on a yoke, similar to what N. G. Herreshoff did. Insure enough separation so the wave trains don't interfere with each other. The bigger the better, 8' being a nice manageable size.

You can eyeball S/L by the wave train quite accurately, particularly if you use video or a camera. Tow at the respective S/L's you desire and the yoke will tell quickly which is more efficient at that S/L, which I assume is your goal.

Working in a pool will provide better results, compared to the full gale sea state conditions (proportionately), a near calm day on a protected pond will generate. I've done it with a fishing rod, but prefer a small motorized winch for reliable, adjustable, repeatable observations of different configurations and S/L's. Herreshoff did it in a pulling boat usually, near shore, very early in the morning in the dawn calms.

 

One of the advantages of being in Florida is the water is warm.
So it should be easy enough to follow Lord and get under the boat as its towed past..

2 boats, at 2 inches to the foot and a yoke is twice the work , but I guess the only way to get a sure comparison.
IF both were powered with the same watts (identical power plants & props)the boat in front would be lowest drag.

Designed all up displacement is 8800lbs (to get into simple Euro cert) so the scale model could be 30-40 lbs , pretty EZ to build with non high tech material.

Not worried that at low SL the std. boat might have less drag , the Goal is 18K at 5 mpg , with a seakindly ride and enough stability to endure a rough day at sea.

FF cruise

 

The count the boxes in a sheet of graph paper is driving me back to the eye doctor.

If anyone has a quickie computer that can figure , please,

I need to know the displacement of the box keel, which has a plumb stem and stern and widens to 3ft at the widest point with a slim fair curve, on the flat bottom .
The top of the box keel is similar , but 4 ft wide and the widest point is a few ft beyond the widest point of the bottom. The draft is 18 inches.

THANKS!!

FF

 

FF
Is the attached something like you had in mind??

It is a slender hull by most standards but I suspect easily driven.

This boat displaces 3.1 tonne.

I can fiddle with it if it does not fit with what you are trying to model.

Rick W.

 

Thanks , You have drawn an entire boat . The box keel is 30 ft from bow to stern , the whole boat is 39. If you look at Rescue Minor , or River Belle on the Atkin site perhaps the shape will be clear.

I am in the process of playing "catch up" we cruised 4400 miles this summer and today I fly north to get our motor home . It has the scanner aboard , so by next week I can post the shape .

Thanks all ,for all the assistance.

FF

 

FF
The proportions of what you describe seem to be about half the beam of the River Belle. Do mean the WL beam is 4ft or 8ft? I have drawn it at 4ft based on your previous description. It will take only a few minutes to take it out to 8ft if that is what you are looking for.

By the way, once the hull is drawn in Freeship it is a 5 minute exercise to estimate performance using Michlet.

What is your target cruising speed?

Rick W.

 

Fred;
Sometimes we design boats with our buddies while hanging out in the bar room. A cheap and dirty method for calculating displacement is as follows....We can use your dimensions to illustrate. Your box keel has a width of 3 feet on the bottom and 4 feet at the top. The box is 18 inches deep (one and a half foot), The box is 30 feet long. Average width of the box is .........(3+4)/2 or three and a half feet. Multiply 3.5 x 1.5 x 30 for a volume of 157.5 cubic feet. That is the volume of a box that has a constant section. Alas the box keel is not a constant section because it has pointy ends. Enter prismatic coefficient. A boat of the general shape and dimensions herin will have a Cp of somewhere in the region of 0.52 to 0.54. Now multiply once again...157.5 x 0.53 = 83.47 cubic feet. 84.47 x 64 = 5342 pounds. You can use one of the other numbers for Cp, 0.52 or 0.54 and you will see that your bar room estimate is within a hundred pounds of either of the other Cp estimates. The numer 64 is of course the weight of sea water.

When fooling with the models you will be dealing in inches rather than feet. Instead of laboriously converting inches to decimal feet just do it by using pouinds of water per cubic inch. For fresh water the number is 0.03611 pounds per cubic inch. That number obviously arrived at by 62.4/1728.

This scheme sometimes has various results that depend entirely on the number of beers or vodka martinis previously consumed. Anyway it is a quick way to make preliminary estimates. It works equally well with round sections if you can detrermine the area of the mid section.