The Simplest Proa?
On my AmbiProa model I used a hullform which went some way towards the mini keel idea, but without actually having a keel as such. The hulls (vaka and ama are similar) have a deep vee midship section, the bow sections are a shallower vee. All in an attempt to reduce CLR migration.
Understood Mal, but I wonder… T2 from Gary (below) is pretty similar, and her CLR migrates a great deal, by most accounts. I’ve a feeling that, unless there is a discontinuity in the underwater shape (a knuckle or crotch), the CLR migration is going to be more like the hull-as-chord rather than daggerboard/keel/deeper hull as chord.
Dave, I agree that the differences are going to be small rather than dramatic. The idea was that a wider, shallower hull section should provide less ‘grip’ on the water than a deep, narrow section, as I’m sure you would appreciate. On a hull like the T2, the ends are usually finer (a deeper vee) than the midship section, whereas on the Ambi hull the ends are relatively blunt. Wheras on the T2, the CLR may be positioned 25% aft of the bow, on the Ambi the CLR might be 30% or 35% aft of the bow. Not a dramatic improvement, but a bit better.
The Ambi style hull can be made more extreme, with the keel line made concave fore and aft of midships, with a more defined stub keel like shape. To do this and maintain a good volume distribution in the hull, the waterlines around midships need to become waisted so that you end up with dumbbell shaped waterlines. Makes for an interesting hull shape.
Some towing tests would be good.
Mal.
Dave, I agree that the differences are going to be small rather than dramatic.
Who knows, might be a breakthrough, too. Too little data.
The idea was that a wider, shallower hull section should provide less ‘grip’ on the water than a deep, narrow section, as I’m sure you would appreciate. On a hull like the T2, the ends are usually finer (a deeper vee) than the midship section, whereas on the Ambi hull the ends are relatively blunt. Wheras on the T2, the CLR may be positioned 25% aft of the bow, on the Ambi the CLR might be 30% or 35% aft of the bow. Not a dramatic improvement, but a bit better.
All of this may be dead on. But we’re only surmising. I think I understand the mechanism which causes CLR migration with increased AoA in shallow hulls, as I expect do you, but we don’t really have a clue in the real world, do we? 😉
The Ambi style hull can be made more extreme, with the keel line made concave fore and aft of midships, with a more defined stub keel like shape. To do this and maintain a good volume distribution in the hull, the waterlines around midships need to become waisted so that you end up with dumbbell shaped waterlines. Makes for an interesting hull shape.
Perhaps. Not disagreeing with you, but achieving “good volume distribution” as you suggest may—or may not—be the very thing which causes larger CLR migration
Some towing tests would be good.
If you’ll build the models and ship them to me (in California), I’ll build the test setup around my pool and test them. Seriously.
Dave
Really interesting discussion!!
No idea how to calculate CLR or its movement but may spend some time to see how difficult it would be to flush some of these models thru Mitchlet.
Cheers,
Skip
I ran a bunch of simple proa models through michlet when designing my boat. I came to the conclusion there was no significant (to me) speed difference between v shaped, narrow and wider box hulls at my target design speed of around 8 knots. You can almost export freeship models directly into michlet. With the program versions I was using you had to add a line to the input file somewhere before the latest michlet would read it. I could probably find out exactly if you’re interested.
-Thomas
I ran a bunch of simple proa models through michlet when designing my boat. I came to the conclusion there was no significant (to me) speed difference between v shaped, narrow and wider box hulls at my target design speed of around 8 knots. You can almost export freeship models directly into michlet. With the program versions I was using you had to add a line to the input file somewhere before the latest michlet would read it. I could probably find out exactly if you’re interested.
-Thomas
Thanks, lets wait and see, I’d have to learn Freeship and there are a lot of other things already in the hopper.
Skip
The idea was that a wider, shallower hull section should provide less ‘grip’ on the water than a deep, narrow section, as I’m sure you would appreciate. On a hull like the T2, the ends are usually finer (a deeper vee) than the midship section, whereas on the Ambi hull the ends are relatively blunt. Wheras on the T2, the CLR may be positioned 25% aft of the bow, on the Ambi the CLR might be 30% or 35% aft of the bow. Not a dramatic improvement, but a bit better.
VERY interesting. Sweet Rhode Island Red does the same thing—the bottom gets flatter near the ends and steeper towards the middle.
Shouldn’t the concentration of lateral area in the middle of the profile , from the flat section in the keel on SRIR, cause even less migration than your fairer keel shape, for a given draft?
The idea was that a wider, shallower hull section should provide less ‘grip’ on the water than a deep, narrow section, as I’m sure you would appreciate. On a hull like the T2, the ends are usually finer (a deeper vee) than the midship section, whereas on the Ambi hull the ends are relatively blunt. Wheras on the T2, the CLR may be positioned 25% aft of the bow, on the Ambi the CLR might be 30% or 35% aft of the bow. Not a dramatic improvement, but a bit better.
VERY interesting. Sweet Rhode Island Red does the same thing—the bottom gets flatter near the ends and steeper towards the middle.
Shouldn’t the concentration of lateral area in the middle of the profile , from the flat section in the keel on SRIR, cause even less migration than your fairer keel shape, for a given draft?
Entirely possible, but (I believe) this may be new territory: In terms of CLR migration, just what comprises a “representative chord”? We know it’s often the entire waterline length of the hull when there’s no daggerboard, and we know that it’s usually the board’s chord when there is one, but I’m not aware of any studies on the ‘tweens. There are all sorts of things we might surmise, but that’s just guessing, isn’t it?
I’m not intimately familiar with Michelet (Heck, let’s be honest, if it doesn’t have a Mac interface, I have *zero* familiarity!) However, I suspect that somewhere in the inputs the program asks the user whether the hull is to include a daggerboard or not, and then chooses the appropriate algorithm for CLR migration already baked in. These will reflect the current state of the art, not the future. GIGO in other words.
Dave
I think Michlet only does resistance calculations, I could be wrong of course.
-Thomas
Regarding testing models, if you just want to test the location of the CLR, all you need to do is to attach the towing line to the deck of the model at roughly where you think the CLR is. Then tow the model (by hand is all you need) with the tow line at an angle of, say, 70 degrees to the centreline, to simulate the thrust line from the rig when close hauled. Adjust the location of the towing point fore and aft until the model goes in a more or less straight line. The towing point will then be roughly aligned with the CLR.
Sweet Rhode Island Red does the same thing—the bottom gets flatter near the ends and steeper towards the middle.
Shouldn’t the concentration of lateral area in the middle of the profile , from the flat section in the keel on SRIR, cause even less migration than your fairer keel shape, for a given draft?
A very low aspect ratio lifting device such as a proa hull functions similarly to a delta wing vortex lifting device, even though the flow may not be separated as a distinct vortex. Most of the lift is generated very close to the leading edge. The leading edge, for a proa hull, is the sloping (swept) portion of the keel line. By introducing the flat section on your keel line at midships, you may have inadvertently moved the leading edge further forward. The important point is that the CLR is not located near the centre of lateral area, it is located towards the leading edge. As we know, for a wing type section, the centre of lift is usually around 25% - 30% back from the leading edge. For a vortex lifting device, it is much closer to the leading edge due to the sharp pressure peak close to the leading edge.
Consider a (symmetrical) proa hull with a keel line that runs in a straight line from the waterline at the bow to the lowest point at midships, then straight up to the waterline at stern, and with constant section shape (vee angle) for all sections along the hull. The CLR for this hull will be very close to the mid point of the leading edge, which is 25% aft of the bow. Varying the section shape along the hull will affect location of the CLR, basically, the sharper the section, the greater the lift.
Mal.
I just laid my hands on enough door skin to make a couple of 1/6 scale models of a 24 footer. Any good glue suggestions for modeling other than epoxy, or should I go grab some of the quicky stuff from home depot?
I just laid my hands on enough door skin to make a couple of 1/6 scale models of a 24 footer. Any good glue suggestions for modeling other than epoxy, or should I go grab some of the quicky stuff from home depot?
Some industrial type hot glue sticks would probably be my choice, forget the clear hobby type stuff but the tan sticks are quick and hold well. If you’re doing mainly fillet type joints and can work outside the premium “gold” bondo is a surprisingly(to me) good product. Epoxy for model work is a mixed bag IMO regular stuff takes too long to set and the fast is a pitiful product.
Skip
I just laid my hands on enough door skin to make a couple of 1/6 scale models of a 24 footer. Any good glue suggestions for modeling other than epoxy, or should I go grab some of the quicky stuff from home depot?
Some industrial type hot glue sticks would probably be my choice, forget the clear hobby type stuff but the tan sticks are quick and hold well. If you’re doing mainly fillet type joints and can work outside the premium “gold” bondo is a surprisingly(to me) good product. Epoxy for model work is a mixed bag IMO regular stuff takes too long to set and the fast is a pitiful product.
Skip
I also try to avoid epoxy where possible. I’ve had reasonable success with polyurethane glues, just make sure you wear gloves though, the stains on your fingers last for days! There are also a lot of higher strength, semi-waterproof wood glues becoming available, like PVA only better, just make sure the model has a few good coats of varnish or paint so the glue doesn’t get soaked.
Every time I go to the hardware store, there is some new glue to try out 😊
I’ll probably look for the glue sticks or Bondo. Gap filling fillet joints are gonna be the order of the day. Thanks guys.
I just laid my hands on enough door skin to make a couple of 1/6 scale models of a 24 footer. Any good glue suggestions for modeling other than epoxy, or should I go grab some of the quicky stuff from home depot?
Some industrial type hot glue sticks would probably be my choice, forget the clear hobby type stuff but the tan sticks are quick and hold well. If you’re doing mainly fillet type joints and can work outside the premium “gold” bondo is a surprisingly(to me) good product. Epoxy for model work is a mixed bag IMO regular stuff takes too long to set and the fast is a pitiful product. Skip
I also try to avoid epoxy where possible. I’ve had reasonable success with polyurethane glues, just make sure you wear gloves though, the stains on your fingers last for days! There are also a lot of higher strength, semi-waterproof wood glues becoming available, like PVA only better, just make sure the model has a few good coats of varnish or paint so the glue doesn’t get soaked.
I learned something kind of cool from the R/C airplane guys. I don’t like “instant” glues, like CA or 5-minute epoxy, and I never really trust the bond strength of hot glue unless it’s a whole industrial system—super hot gun, air powered, expensive, etc.
Polyurethane glues have gotten much better—you can now buy it in clear (dries white) and also fast setting (well, faster)
Anyhow, here’s what they taught me: Choose your slow-setting glue, whatever you like (I like Elmers, but your mileage may vary!) When you spread glue on the pieces to be joined, leave about 1/2” of dry material (no glue on either piece) every few inches of glue line. Just before you lay the parts together, take your hot glue gun and put a dab on all the dry spots. Slam the parts together, wait 10 seconds, and it’s done. The hot glue dots have become your “clamps” while you wait for the slow stuff to set/dry/kick. The join is strong enough to clean up glue, add more if you have gaps, even (gently) move onwards with the next operation. If you’re using polyurethane glue, the"dots” help keep the foaming glue from pushing your joints apart, too.
Oh, and a trick for Polyurethane glue (either white or brown). With rubber gloves on, wipe the glue lines with a wet cloth, or even your wet finger, after it’s in the clamps. This stuff absorbs water in order to set, so give it some… the 2-hour brown stuff will set in 20-30 mins and the white 1-hr stuff will set (mostly) in 10-15 mins. Some recommend wetting the parent wood on one piece and putting glue only on the other, but I’m not for it—too much chance of poor adhesion to the “wet only” side, IMO.