Cash prize of 250 GBP - Dinghy Design Competition

  • 09 Jul 2021 10:13
    Reply # 10739505 on 10211344
    Anonymous member (Administrator)

    Thanks for your feedback and comments David, I have been thinking about this a lot, and I partly agree with you.

    But not entirely.

    “The whole process of stability testing” is most certainly not invalid. I think it might be true that some of it is irrelevant, at least to most people, but that is not the same thing as invalid.

    I am concerned though, because I think the way I have been going about it is invalid, and I am at present trying to think of a better way to do it.

    I also think that looking at the results and trying to figure out which dinghy is “best” is invalid, and I think that might be what you are saying. If so, I agree. I am not really very interested in the “competition” and wish to have no part in the judging.

    I intend to continue, because I think the testing can potentially show that the various dinghy shapes are different, and perform different tasks differently. In the case of drag, none of them are all that very different (they are, after all, only 8’) but even that is something that is worth demonstrating, rather than just taking for granted. I think it is worth it, though I would not take the actual numbers too seriously, as I have stressed repeatedly. As for stability at various angles of heel – these different section shapes perform VERY differently, and while some people, such as yourself, have a very good idea of form stability and how it depends on shape – some of us are not so knowledgeable and I think these practical demonstrations can teach us a lot. I do, however, want to try to do it better, because I can see some real flaws in the way I went about the stability testing.

    I think you are probably right in saying that beyond 30 degrees of heel, stability curves probably don’t mean much in relation to dinghies. After all, we are not expecting “recovery from a knockdown”. Never-the-less, some of these dinghies do continue to remain stable right up to 80 degrees or more, and do still provide some of us with something to learn, on the subject of form stability. How the struggle to return to an upright position plays out, might be of interest to some people. (How many people in this world still believe that boat stability only comes from a lead weight slung beneath the body of a keel boat – most people, I would suggest – but not JRA members I hope!)

    The nice thing about these forums is that one can skip what is of no interest, and I imagine most people are probably now skipping this thread because I have had far too much to say on it. I would be very grateful though, if you would continue to monitor this thread, and continue to criticise. That is good, and probably necessary.

    By the way, the shape of a grand banks dory has little to do with its motion in a seaway. They evolved from a need for something that can be built quickly and cheaply from flat planks and can be stacked up one on top of the other and "nest", on the deck of the mother ship when being carried to the fishing grounds. Horses for courses. (I have seen factory-made moulded fibreglass “grand banks” dories, which completely misses the point, and shows little comprehension of the concept or its raison d’etre). For purely motion in a seaway, there are probably better models. I am sure you are right, however, in observing that the “section” which has evolved, (the angle of flare etc) is quite important in the way they behave at sea, and they seem to have proved to do the job well.


    (While I was doing this post, my dinner burned on the stove - and also Arne posted a reply).


    Arne, I just saw your post. Thanks, you have said in a few words what I was trying to say in a hundred. I want to do the stability testing all over again, and then I will try to find a way to post all the graphs, and how they relate to the various mid-section shapes. I am sure none of it will be a surprise to people like you and David, but for some of us it will be quite educational I think.

    Last modified: 10 Jul 2021 00:02 | Anonymous member (Administrator)
  • 09 Jul 2021 09:10
    Reply # 10739461 on 10211344
    Anonymous member (Administrator)

    I can see David’s point here. The absolute numbers will not be of that much value, since other factors play inn her, like the builder’s chosen weight (thickness) of the plywood, the sort of interior (tanks, bulkheads or not), and not least, the chosen height of the thwarts.

    However, the stability curves obtained so far are far from useless:
    The shape of the stability curves gives us a good idea of the character of the boats.

    • ·         For instance, the 5-plank versions have a slack slope, indicating initial tenderness compared to the ultimate stability.
    • ·         The 3-plank versions with flaring topsides have a steeper slope,
    • ·         while the box boats with vertical sides have the steepest stability curve.
    • ·         The ultimate stability is then just a function of the beam of all the tenders.

    I don’t think there is much use in going into deeper science with this, but the curves surely are interesting to watch, so I hope you let us see all of them, Graeme.

    Arne


    Last modified: 09 Jul 2021 10:29 | Anonymous member (Administrator)
  • 09 Jul 2021 07:44
    Reply # 10739339 on 10211344

    Graeme,

    Sorry, but I have to say that I think this whole process of stability testing is invalid. The crew is live weight, not dead weight, and in the event of a lurch to one side will instinctively try to correct, and then will finally slide or fall down to the lee side. In a kayak or small dinghy, one instinctively pivots at waist level to keep the upper body upright. I don’t see any realistic point in these tests, certainly not beyond 30˚ of heel. And anyway, an unstable, round bottomed boat is better in a seaway - eg greenland kayak - but a flat bottom boat is better for standing up and hoisting water jugs onto the side deck. It all depends on what kind of boat you want. Probably the 5-planks are the best all-rounders. Though a 3-plank can also be good in a seaway, if the section is right - eg grand banks fishing dory. 

  • 09 Jul 2021 03:14
    Reply # 10739065 on 10211344
    Anonymous member (Administrator)

    Dave W came to visit this morning and brought another model dinghy – this is his DD which is one of his three entries in the dinghy design competition.

    I was pleased to see this somewhat radical model – it is one that I had wanted to include, but thought it would be too difficult for me to make. It is an “extreme” design, which is good for testing, because these “outliers” provide good contrast.

    Here is Dave W.’s “DD”.

    Dave stole an advantage here, by clever use of the “8’ sheet of ply” constraint. The bottom is made in two parts (it’s a shallow vee-bottom) and the centre line comes from the diagonal of a plywood sheet, rather than the length. So, it is a 2.4m dinghy which is actually just under 2.7m. That will make a difference to its hull speed, and to some extent its stability. In addition, the model itself is very over weight, to scale – so that gives it extra stability over and above what it will obviously have from its shape and its beam – compared with the other models. (That “unfair advantage from over weight" is a flaw in the testing procedure I have used.)

    We put the mast on it, and put it in the tank.


    The result was – the stability curve went right off my scale!

    Here is the graph again, with its own special vertical axis

    Massive stability, right up to the point where the gunnel is immersed. A few more degrees of heel is possible because of the buoyancy tanks, but once the dinghy goes beyond about 45 degrees and floods, like most of the others, it continues to float, but with little or no stability.

    I have not yet been able to test these models in choppy conditions. Dave says that his full size prototype DD, with its fine entry, is good in a chop.

    I think I will have to test all the models again, and find some other way of compensating for differences in the over weight factor. Putting a discount-weighted “passenger” up on the rowing thwart has disadvantaged those little dinghies, when compared with big ones like this whose over-weight is built into the hull, and which do not need a “passenger” sitting up high, which affects the metacentric height.

    (Maybe I'll just put the "passengers" on the floorboards this time - which then confers a slight stability advantage - this time to the "little" ones.)


    Last modified: 09 Jul 2021 04:57 | Anonymous member (Administrator)
  • 07 Jul 2021 16:28
    Reply # 10734987 on 10211344
    Anonymous member (Administrator)

    Each of the dinghy models has now undergone a kind of “stability test”.

    Each dinghy is rigged with a simple, temporary mast (0.5m in length) from which it is hauled down into a heeled position, the hauling line attached to a tension gauge calibrated in Newtons. The position of the tension gauge is moved, between trials, to try to get the hauling line to remain roughly perpendicular to the mast, as the mast rotates from trial to trial.

    Each dinghy is restrained by two lines which go under the hull and prevent it from moving sideways when hauled down from the mast head.

    When making an "observation", the angle of heel is estimated by lining up the mast with a large “protractor” scribed on the far wall.

    The righting moment in Newton-metres, at a series of angles, is given by force x 0.5m

    There is a problem: the weight of the model has a big effect on its stability. I have tried to compensate for over-weight in the models by placing a “passenger”  (jar of sand) whose weight is reduced by the approximate over-weight of the dinghy model (see previous post, and see the following examples).

    This way, over-all weight is near as possible to what it should be for a real-life dinghy and one person rowing. However, this results in anomalies in the DISTRIBUTION of the weight. The weight of the "passenger" is placed, high, on a rowing thwart - instead of the extra weight just being built into the hull. The trouble is, the dinghies are all over weight by differing amounts, and thus carry "passengers" of differing weights. The dinghies which carry a heavy "passenger" (generally the smaller, lighter dinghies are the least "over weight") now have a higher centre of gravity than those which carry a lighter passenger, and are thus at a disadvantage when comparisons of righting moment are made.

    Lighter dinghies, such as YouYou, KISS and Sibling Tender are particularly disadvantaged in this respect. 

    I don't know what to do about this - I should have worked within a budget for weight, when making these models. As it is, the project is somewhat flawed.

    [Edit: because of this, if I can find a place to post a spreadsheet for all the results, I will group the dinghies according to the weight of their "passengers". It probably all means that although there might be some valid comparisons possible, the absolute numbers probably don't mean much, which is why I have not bothered to scale them up.]

    Here are a couple of the results:

    Halibut

    David’s Box Barge


    The stability curves all differ greatly – a wide range of results – and some of them with a very wide and some with rather narrow stability range.

    I will try to find a way of getting all of these posted on the website.


    Edit: with their decks up near the gunnel, and full fore-and-aft buoyancy, Simplicity 8, General Purpose Dinghy and the NZScow can be heeled to almost horizontal and still have some reserve stability. They will float on their side tanks and come up dry. Little, light, easily-propelled open dinghies such as KISS and YouYou will flood and capsize as soon as the gunnel is hauled down to the water line. Big, beamy, stiff dinghies such as Halibut, and AD maintain high stability until the gunnel goes under - then suddenly its "all over, Rover". Easily-driven, narrow-beam dinghies with lower-placed built-in buoyancy, such as the baby SIBLIM have their own special merit but do not carry such heavy loads. Oyster, with its inverse v-bottom and two bilges has an interesting stability curve, and, by the way, actually behaves quite well when partially swamped.

    don't believe there is a perfect combination of ease of driving, convenience for rowing, stability, load-carrying ability - every one of these attributes comes with a price. Especially in the range of 8', which inevitably means an extreme displacement vessel. "You gets what you pays for." No 8' dinghy will ever row worth a damn, yet they are probably all good enough for a tender - and they all seem to have at least one outstanding feature - at the expense of some other feature. Then there are the unmeasurables: the presentation of plan and instructions, suitability for a schoolroom project, ease of construction and use of materials.

    Well done, those who submitted entries! I'm just glad I am not one of the judges!

    Last modified: 08 Jul 2021 03:47 | Anonymous member (Administrator)
  • 01 Jul 2021 05:46
    Reply # 10716255 on 10211344
    Anonymous member (Administrator)

    AD is a Mack truck alright! We piled the equivalent of 375 kg into it and it still had plenty of reserve buoyancy for more. You can hear the little boy (my photographer) chuckling.

  • 01 Jul 2021 00:01
    Reply # 10715624 on 10211344

    Great work Graeme!!

    Very interesting seeing the models in real conditions. Pity that DD was not there for comparison, I may try and build a model to the same scale so that you can test her as well, also my Webb 8. I will try and call in on my next visit to Auckland.

    David.

  • 30 Jun 2021 16:49
    Reply # 10714487 on 10211344
    Anonymous member (Administrator)

    Testing for drag – part 1 towing in static water

    There are two ways we can test for drag. Here is the first – it was suggested by Marcus, who saw someone testing in this way.

    It comprises hoisting a weight to the top of a 10m (approx.) mast, then releasing it. The halyard runs back through a turning bock and thence to the model where it is attached to a towing point. When the weight falls, the model is towed down a runway alongside my shed.

    Here is the towing apparatus (53 grams hoisted to 10m) and the “tow path” at spring high tide.

    The idea, as I see it, is that a fixed amount of potential energy (mgh – the weight at its highest point) is released and imparted to the apparatus. After friction losses (which should be approximately constant between trials) the energy is converted into forward motion of the dinghy (kinetic energy) and wave motion. I don’t intend to attempt any calculations, hoping that timing the run will be a sufficient measure of the model’s drag. We will also be able to observe the dinghy’s behaviour under tow, and perhaps the waves it generates.

    Using a sail-maker’s thread for a halyard, and tiny blocks – and much trial and error – a 53 gram weight was found to be sufficient to tow each model through the water at a fairly constant speed. (The first metre, while accelerating, was ignored, but after inertia was overcome the rest of the run was at a kind of “terminal velocity” and that part of the run was timed). A crude assumption was made that the time of the run would be a measure of the dinghy’s drag.

    (I think this might be the nearest we can get to learning if there is any difference in how “easy” each dinghy would be to row).

    Here is Sibling Tender under tow with one “passenger” (a jar of sand weighing 600 grams, equivalent to a 75kg person) – less a factor allowing for the model’s over weight.

    Each model trails a retrieval line, adding a little more drag, but the same amount for each.

    Timing the runs using a “stop watch” app on a cell phone proved problematic – wet fingers etc – and eventually, after many foul-ups, much bad weather (and bad language)  I recruited a grandson as technical assistant, and he simply made a video of each run, enabling rough timing (to the nearest second) between two recorded events (a) passing the start point after initial acceleration – and (b) running into a stop at the end of the tow path. Not very precise – but probably consistent enough.

    The video timed-runs have been edited into a clip and placed on Youtube – here is the link

    https://youtu.be/M9PFpAM5zUs.

    This will allow some observations of wave creation and behaviour under tow, though wind and other conditions probably affected some of the runs.

    There were two stand-out features regarding behaviour, which can be observed in the video: (a) the steady course (directional stability) of Oyster and some others with deep or long skegs – and (b) the yawing behaviour of David T.’s three, none of which have skegs. These three were OK when lightly laden, but with two or three "passengers" would tow straight only when trimmed by the stern which probably then added more drag. Note however the “horses for courses” motto – these three were not designed for towing, they were designed to have minimum drag and maximum manoeuvrability for rowing – and as it turned out, they were among the fastest on the tow path. David has agreed that in order to test like-for-like we might add skegs to the models, for future trials, and if I have time, I will do so. KISS also has no skeg – but this dinghy has its centre of buoyancy much further aft, trims down by the stern anyway, and does not seem to need one.

    (As a digression, directional stability is the inverse of manoeuvrability, so you can’t expect lots of both. And it may not just be a matter of skeg or no skeg. I think that if the centre of buoyancy is further aft, with respect to the centre of mass, a vessel might tend to have more directional stability. [Edit: I have since realised that is nonsense. The distribution of weight and reserve buoyancy might, however, have some effect on directional stability when the dinghy is overloaded.] By design, baby SILBIM and particularly Sibing Tender both have full forward quarters – whereas KISS has her main body fullness aft).

    Here are the numbers for the first trials of “drag while under tow”.

    Some of the trials were repeated, in order to check consistency, and where there seemed to be a range of results they are shown here. The asterisk refers to tows in which the vessel yawed, and the trial was repeated with a down-by-the-stern trim.

    The next set of trials will be with the dinghies stationary, on a tow line, in moving water.

    Down at the launching ramp, where there is a proper pontoon. When the weather improves a bit.

    There will also be some static testing for stability, posted on this thread in due course.


    Last modified: 01 Jul 2021 21:51 | Anonymous member (Administrator)
  • 30 Jun 2021 07:15
    Reply # 10713075 on 10211344
    Anonymous member (Administrator)

    There you go, David. It still doesn't show the real disparity in the sizes of these things - in terms of volume there must be a range of nearly 100%.

    Back row LtoR

    Halibut, Sibling Tender, YouYou, baby SIBLIM, KISS, GP Dinghy, Medium Boy

    Front row

    AD, Oyster, Boxer, David's Box Barge, NZScow, Simplicity 8


    I'll try to post some numbers later this evening.

    Last modified: 30 Jun 2021 11:42 | Anonymous member (Administrator)
  • 29 Jun 2021 08:20
    Reply # 10709989 on 10211344

    Good work, Graeme! This is the way to evaluate all the designs.

    Could we see a photo from an upstairs window looking vertically down, to get a better idea of relative size and planforms?

       " ...there is nothing - absolutely nothing - half so much worth doing as simply messing about in junk-rigged boats" 
                                                               - the Chinese Water Rat

                                                              Site contents © the Junk Rig Association and/or individual authors

Powered by Wild Apricot Membership Software