Comparison testing of an Atkin tunnel-stern v-bottom Seabright skiff and another boat

Ken,

Optimizing the hull/powerplant for each boat does not necessarily mean that individual testing needs to be done to make that possible. We would need to assume that Atkin has already done that for his tunnel hull based on his considerable experience. What I meant by "optimization" for the other hull is to use a power system based on the equivalent body of past experience for that particular type of hull. In other words, each hull would use a power system that it could reasonably be expected to use in the "real" world and these would not be expected to be the same.

I think most of us could agree that there a very few of truly optimized packages out there. Our mantra is, after all, compromise all design elements to get the desired over-all effect we want from a particular design.

 

For this we need to build and test full sized boats -- easily an order of magnitude more costly and time consuming than the tests I proposed. But hey, I'm game to test any size hull if that's what the sponsors want to fund. How much shall I put you down for?



Actually I think lots of people would love such figures ASM, but testing full-sized boats is never going to happen without a grant from Bill Gates -- at least not with new hulls in a side-by-side comparison. In the meantime we can do something else ... or not?

 

Wow, this thread is zooming along faster than I can read, it seems.
I will try to answer a couple questions about the shaft angle comment I posted earlier.
The fellows who built the boats and did the side by side comparisons are no longer living so I can not clarify details. They built a lot of traditional Columbia River bowpickers (salmon gillnet boats) and when I was told about their shaft angle experiments I assumed that was the type of boat they experimented on. But they also built a lot of other boats that were stern pickers and seiners, etc. If my bowpicker assumption is correct then the boats would most likely have been powered by Chrysler Crown engines. They would have had a 2:1 reduction gear and a pretty square wheel, most likely about 22x22. These guys were no dumbies so they would not have claimed to have done comparisons of boats that were not built from the same molds, had different engines, reductions or props or varied in any other significant way. Their 'research', I am sure was for their own satisfaction, or perhaps because of customer's questions. Typical bowpicker dimensions were; length 28', beam about 8', draft about two and a half to three feet. They mostly ran in the 15 to 20 mph range we are talking about here.
The point I wish to leave with is that these results were not the theory of some engineer about what the shaft angle should be for maximum efficiency, it is actual real world results that indicate you do not need to worry much about not having a zero degree shaft angle.

 

Got it, thanks Tom. This brings up my next question: How do we do this with boat models?

Maybe it's possible with small "real" boats because there are small outboards and inboards available, but I've never heard of an outboard that's small enough to be used on a 4 foot model. And how do we measure the fuel efficiency of two dissimilar hulls *AND* two dissimilar engines?

Agreed, but we could have a software efficiency expert like Rick Willoughby take a couple of my FreeShip designs and run them through Michlet/Godzilla in an effort to optimize them for 15 knots with the constraints of equal waterline lengths and breadths. Then at least the hulls will be optimized even if the propulsion systems are not.

 

Gilbert,
Weren't most bowpickers powered by outdrives? My experiences with the Pacific salmon fleets were that bowpickers were small dayboats built for relatively high speed and small loads. Almost all that I have ever seen had outdrives with adjustable shaft angles. I have seen a couple bowpickers with straight shafts. One was a 50'er that seemed an over enthusiatic example of naval architecture, the other was a stripped out old plywood hulk rotting away on the beach.

Stern pickers built after the early 60's were modeled on planing rather than the displacement hulls. Using straight shafts and depending on the power used they would lumber along at semiplaning speeds for the most part. Trim angles change with load, heading out bow heavy and coming back to port dragging the *** end (hopefully)

Seiners are for the most part full displacement hulls running straight shafts. They run bow heavy out and trim to thier lines when loaded. They haul far greater loads than the gillnet boats having first being designed for volume rather than value fisheries and often traveling longer distances carrying these more substantial loads.

I must say most of my experiences were on sternpickers. They were not in my opinion optimised for economy or performance but instead were grand comprimises designed for widely varying conditions of load, speed, and net tending abilitities.

 

Atkin Heron Model

Wooden model
Made a wooden model according to Atkin Heron, size 10:1, weight 0.6kg. This model is suitable for towing tests too, as the weight reflects to 600kg in full size. Will see? A reference hull form is requested.
This hull is actually like a skinny L/B=5 pointed stern hull with a more wide, low draft dinghy added over it! Very similar to that "Displacement Glider":
http://www.boatdesign.net/forums/showthread.php?t=2545
May be that prop tunnel is not essential in economy business of these boats at all? Instead, the skinny "keel-hull" is?

It would be unfair to compare this against a standard runabout hull. Runabout hull is designed only for high speeds, while the Displacement Glider is best in speed range of froude 0.4 to 1.0 :
http://www.alsphere.at/download/DG_Factsheet_E_1.3.pdf

redu

 

redu, we won't learn about propulsion efficiency by towing models, we need propulsion tests for this. Unfortunately there are no model-sized outboards in existence (none that I know of anyways) and producing one would cost more than simply abandoning the model scenario and going with half-sized "real boats" and off-the-shelf outboards.

Maybe, but without the tunnel section we would be testing an original Seabright skiff, not a tunnel-stern version -- and it's the tunnel-stern boats most people have an interest in.

Exactly, which is why I suggested that the boats be optimized in Michlet / Godzilla for 15 knots. This will give us two boats with the same waterline length and breadth but dramatically different bottoms. Since both hulls will be optimized for their particular shapes at 15 kt, our comparisons might actually mean something!

 

Ken, are you looking at the hullform I sent (did you get it) I need an idea of the monetary contribution necessary? One version was optimised by Rick for 15knots.

 

Brian, I got your email and asked you to send me an exported .fef file but I never heard from you after that so I thought maybe you changed your mind. I just checked my SPAM box and cannot find your reply there either. I wonder what happened to it... ?

Why don't you just send it again, maybe I will get it this time. Since I'm focusing on 15 knots maybe you should send me that version, unless you prefer another one?

Thanks.

 

What hull style should we use for the second boat?

Hello all,

While our theoretical discussion is very interesting and I enjoy it tremendously, I think I have come to a conclusion here. Tom and I agree that my proposed testing scenario is a valid way to determine the relative efficiency of two test hulls when both are powered by inboard engines. I trust my own judgement here, but when it is confirmed by someone like Tom I am even more certain I'm on the right track -- thanks Tom!

Since I cannot build models or run tank tests I'm going to proceed with my original offer to do the tests I described earlier -- real world tests using real boats in real sea conditions -- but this time I'm hoping to limit the discussions in this thread to details of the following specific testing scenario, which I hereby post for your re-consideration:

1- I will agree to build the boats and run side-by-side, real-world performance tests of two small boats, the first a computer-optimized "Shoals Runner" tunnel-stern v-bottom Seabright skiff of William Atkin, and the second a simpler computer-optimized hull shape as yet to be determined.

2- Both hulls will be built to a scale that makes the boats about 10-12 feet long. They will both have the same waterline length and beam, and both hulls will be optimized in Michlet / Godzilla for maximum efficiency at 15 knots. During testing I will use sandbags and/or water jugs to make the displacements of both boats the same.

3- In "Phase 1" of these tests I will use cheap off-the-shelf inboard propulsion components typically used by local fishermen in the Philippines, including China-made 6.5 HP gas engines, stainless steel prop shafts, 6-inch diameter two-blade stainless steel propellers, etc.

4- The inboard drive systems will be direct drive. I will measure the fuel consumption of each boat in side-by-side tests to determine which one is more fuel efficient in various conditions. I will repeat each test after swapping engines and propellers to eliminate the possibility of one boat always using the most efficient propulsion system.

5- In a possible future "Phase 2" round of testing (should a second round of funding become available) I will remove the inboard engines and install outboards in both boats, then I will repeat the same tests I made with the inboard engines. I will swap the outboards like I did the inboards to equalize the power systems again.

6- By running Phase 1 tests with virtually identical inboard power systems, we will learn which inboard powered hull is more efficiently driven in a variety of different sea states. Then later, if we proceed with Phase 2, I will run the same tests with virtually identical outboard engines so we can learn which hull is more efficient with outboard propulsion.

7- In addition to efficiency comparisons noted above, I will also report on the handling and performance characteristics of both boats when launching, beaching, accelerating, decelerating, turning, driving hard into a chop, and more. Hopefully this information will be an interesting and useful addition to the efficiency data.

8- Since I forgot to include fuel in my last budget it was too low at only $700, so my new budget is $1000. This will cover the same expenses listed previously, and also the missing fuel cost, and extra funding for unforseen expenses should they come up.

9- Anyone willing to support this project is welcome to make their pledges in this thread. Please don't make a pledge unless you are serious about funding the project unconditionally. When you make a pledge, just keep your money until we confirm that 100% of the proposed budget will actually be funded by those who pledged -- and then send your funds promptly when we make this confirmation.

---------------------------

Don't worry, I won't be disappointed if no one wants to discuss this project with me any more. I understand that this kind of testing will not satisfy some of you ... and for these folks I suggest creating a new thread of your own so you can discuss your alternate testing philosophies and how to get the money for them. As far as I am concerned, I am satisfied that these tests will make it clear which of the two hulls is more efficient when propelled by identical inboard engines and propellers -- and these results are good enough to satisfy me in a Phase 1 testing program.

I know my testing scenario is not perfect, but I still think it's as good as it gets for such a low budget. I also think it is very important to keep the costs as low as possible. If we don't, the whole thing becomes little more than a theoretical discussion that will never get funded ... and I (for one) would prefer some new information about the performance of these hulls for a change, rather than yet another round of theories.

---------------------------

Assuming anyone is still with me here, I would like to suggest that the next topic in this thread be:

What hull style should we use for the second boat?

Does anyone have a reasonable suggestion? If so, please post a picture and explain why you think it's a good choice -- then we can all argue with you until you give up and run away with your tail between your legs ...

Remember, we will run the second hull through Michlet / Godzilla to optimize it for 15 knots, so it won't end up being the exact same shape as it starts out, but it should be optimized for efficiency if the software tools are as good as some say they are.

By the way, please suggest hulls that are fast and easy to build in plywood with as few panels as possible, just like the first hull, thanks.

 

Both emails were sent one to gfj. etc on 17 may and the other to your corporate name - same place just minutes ago.

Put mine in the comparison if you like for broader comparison - makes 3? in the challenge..

I presume the wet areas are only going to be built (no cabin tops etc - not a pretty sight but useful design test-beds)

 

Hi Brian, I got your .fef files this time, thanks.

A third boat makes a lot more work, that's for sure ... and a catamaran at that -- two more hulls to build, another engine (or two?), another skipper, etc. I have some special questions regarding a catamaran too, but we can discuss them privately since this part of the test will be yours and yours alone.

Yes, the builds will be minimal to insure hull shape, stiffness and strength, so no cabins or other extras -- in fact nothing other than what is absolutely necessary for the boats to hold their shape and not capsize, swamp, flex too much, or break apart at speed.

 

Ken,

The hull originally done by William Hand and modified a bit by Weston Farmer and later by Harry Bryan as the Handy Billy has a reputation as an easily driven hull in the speed range we are considering. Only slight modification is necessary to build it in sheet plywood form and I think this has also been done.

I assume you have looked at realizable displacement numbers in a 10 to 12 foot boat?? How much do you or your designated driver weigh? There is often a problem in model testing to hold the scale weight to a realistic number. I'm guessing the model displacement will need to be about 1/4 of the full scale boat. That might be difficult to do since the driver weight doesn't scale along with the boat.

Poor pensioner that I am, I will pledge $100 to the testing and increase it if not enough is gotten from others.

Some have called the Atkin tunnel a box keel but I don't think this fits at all. The glider form and some of Bolger's hulls fit the box keel designation more closely. I think the canoe/box keel with nearly flat upper hull has a lot of promise at the speed range of our tests but that is another and different test program. That also means a different thread as we should not hijack this one.

 

Only one model

Ken,
what about using only one model? If you really want only to check the prop tunnel contribution, you should use a horizontal axle, same wetted area, same wave drag, etc for both cases.
Here a simple and leakproof solution:
Use one hull only, and move a horizontal axle sail drive (or something similar) to two places in hull keel line.

redu

 

It has seemed to me that most of the ballyhoo regarding the purported efficiency of the Atkin tunnel stern hull is based on claims by Rob White concerning his modified Atkin designed Rescue Minor. I don't doubt Rob's claims, but I do think his boat's performance is mostly due to her incredibly light displacement, with a D/L ratio on the order of 42, loaded. I don't know of any other inboard diesel powerboats capable of 17 knots with a D/L ratio that low.

As Billy and John didn't often publish displacement figures it's hard to know what their intention might have been without modeling and running the calculations on candidate hulls. I don't know what the designed displacement of Rescue Minor is, or that of any of the other hulls discussed. But this weight and its relationship to other factors is a key to the boat's performance, therefore it must be modeled accurately. I would like to see each boat weighed on a scale before testing, and corrected to within a few pounds of the other.

To maintain that D/L the 1/2 size version will have to weigh about 95 pounds with crew. Not really possible. Though it could be done with Radio Control and very light ply.

You have already severely limited the "optimization" of the second hull by stating that it must be simply built of ply. This eliminates much of the competition, say Maine Lobster Boats, for instance.

I have done two boat testing in the towing tank. As Tom states above, it just raises more questions than it answers. Especially when you are working with unusual hull forms or features. You cannot (despite Godzilla and Michlet) optimize anything with a two boat program. You need to develop a family of hulls to discern trends in the effect of various features. One data point does not tell you much, a string of points can inform.