Dennis
Your result is almost identical to mine. You must have let it do more iterations because you got a touch lower drag. I think that should be a good shape to aim for with the lw hull.
You can force the symmetry by narrowing down the range of the shape functions. But I always fair out in Delftship and with lowest drag hulls that are going to be pushed above hull speed they will not have much rocker if any.
There are a couple of things that happen as you keep pushing up the design speed. The first is that the hulls eventually start to get shorter and narrower. Then they will switch to very short. This will probably happen when you work on the ww hull. If it happens you know you would be better off with a planing hull if drag at the that speed was your only concern. However you need to design it to suit a range of speed and a range of loading.
Rick
On 07/08/2010, at 12:28 PM, Dennis Cox wrote:
Rick,You're a scholar and a saint for putting me out of my misery.I was just certain I was missing something. There is enough differences from the documentation and the input files, I had to figure there was some undocumented flag you knew about. I figured the flag might be in the "Special Hull Constraints" section. It seems like a nice enhancement to have just such a flag to force fore/aft symmetry for Proa designs. Maybe, canoe and racing shells might be able to take advantage of it also. Just a thought... maybe he'd listen to someone like you. hint, hint.Anyway, I did the "3", 300kg, 7 m/s analysis this afternoon, but threw it away because it wasn't totally symmetric. I'm re-running it now...Its currently running through 50,000 evals.D=0.2927L=8.7994T=0.1518B=0.3576Cp=0.761HRT (@7m/s) = 0.2646 kNI understand what you're describing with the DelftShip... no problem! Its pretty much what I expected I was going to have to do, unless you gave me the magic bullet. I'll hit it tomorrow and send you the stuff. Won't be able to do it in 30 minutes...BTW, I was running some studies using Hull Shape Function 1 and letting it go wild. At least it was symmetric. I noted that at lower speeds, it was bumping up against my max length of 30' and using lower Cp's. At higher speeds, it was settling for something shorter with high Cp. It made the transition around 12 knots.Anyway, thank you for the information... I'm off... Dennis
From: Rick Willoughby <rickwill@bigpond.net.au>
To: harryproa@yahoogroups.com.au
Sent: Fri, August 6, 2010 6:56:25 PM
Subject: Re: [harryproa] Godzilla... how'd you do it?
Dennis
Setting the section shape factor to 0.2 will give you a tight round chine. It should have a little bit of lift in the bow and maybe a little or none in the stern. I do not force fore-aft symmetry.Compare this result with both fully constrained section shape (0) and unconstrained (3). If the fully constrained is not much different then use that version as it requires less eyeballing.Once you have decided on the partial constraint or full constraint make the following steps:Import the resulting offsets into Delftship.Delete all the longitudinal lines (waterlines) apart from the one closest to the centre of the chine; leave the keel line and the top line. You now have three lines that define the half hull. Delete all the leftover points on the stem as well.Crease the one remaining longitudinal line - this will be the line of the hard chine.Delete all vertical lines (stations) aft of the one just aft of centre or on centre.If the hull markers that you imported is not turned on then turn it on. These are the guides to use when fairing the chine.Once you have the chine where you want it line the keel line using the profile view with the chine line.If you have used the partial constraint then the sides will have slight flare. Extend the gunwale line up to give you the required deckline. The station lines should just intersect the top of the imported hull marker lines. This keeps the original waterplane as produced by GODZILLA.Mirror what you have in a transverse plane to create the stern (or other bow).Once you are happy with the result export to Michlet and check how close the resistance is to the original result. If it is more than 1 to 2% higher then you may not have eyeballed very well. Irrespective it is not too critical because the aim is to see how well you can predict performance.What drag have you got for 300kg at 7m/s? I just want to confirm you are close to what I got.I can do this loop in about 30 minutes but I have done it hundreds of times so am well practised. It gives a good appreciation of what is important as far as the shape goes. For these lightly loaded, fast displacement hulls they end up long and slender. The result for a target speed of 8m/s might surprise you though!Email me the fbm file you end up with so I can have a look. There are a few little things in the way Delftship exports to Michlet that can result in slight errors. It pays to use the Gaussian curvature to check for any panel oddities.RickOn 07/08/2010, at 6:51 AM, Dennis Cox wrote:
Rick,I've been using Godzilla most of the afternoon. I've re-read the section on Shape Series over and over. Besides, Series 1, I don't see how you achieved the Godzilla optimization in your picture below. Even the in.mlt file you sent me some time ago does not enforce symmetry... and as far as I can see, I haven't found any way to get the prismatic station plan your flat bottom version exhibits. Square yes, but not strait sides that are flared out.Are you saying... that you take this rounded, 0.2 optimization into DelftShip and flatten out the sides and bottom till its develop-able and then export that back into Godzilla as a designed hull and then let it optimize on that?Dennis
From: Rick Willoughby <rickwill@bigpond. net.au>
To: harryproa@yahoogrou ps.com.au
Sent: Sat, July 17, 2010 5:22:59 AM
Subject: Re: [harryproa] construction
Dennis
The result I arrived at has surprised me but that is nothing new when using these tools. The linesplan for the round section and flat bottoms hulls I compared can be seen on this link:I had to constrain the minimum length to avoid getting planing hulls even with 4.5t hence the 18.3m.Making an estimate that the sail CoP is 10m above the centre of hull resistance, the round section hull actually trims 98mm bow up on net lift of 123mm at 15m/s while the flat section has a slight bow down trim of 30mm on a net lift of 217mm.I expect the drag I get from Flotilla is lower than you have seen with Michlet because both hulls exhibit lift at 15m/s. The round section shows drag of 5.02kN while flat bottom has drag of 5.34kN.Unlike the lighter hulls I have looked at before the round hull exhibits higher bow up moment than the flat bottom hull throughout the speed range to 30 kts.As noted in previous posts the flat bottom may get some benefit from dynamic lift if it actually trimmed bow up but that would require a lower CoP than 10m.Either hull works reasonably well and should be capable of 30kts with an efficient rig and rudders. The round section hull shows a slight advantage from drag perspective but then when you look at simplicity of building consistent quality with a flat panel hull it is likely it would have better performance.RickOn 16/07/2010, at 11:49 PM, Dennis Cox wrote:
Rick,I was just trying some things with the Godzilla after you sent it to me and I put it through an optimization that ended in a flat bottom hull. It did have very high speed potential, but if I remember correctly when I spit out its resistance curve, that it had a very high spike down around 10 knots... (far higher than the peak resistance at 30 knots). At first, this seemed reasonable to me... that the high surface area would be still in displacement mode and thus making a lot of drag. That once above that speed it would start planing and loose friction. Then... I recalled you said that Godzilla doesn't do lift... so my massive rationalization went out the window. So, I just assumed gigo happened.If your offering... what about a lee hull for a Harry...Length <= 18.3 mDisplacement when flying windward hull = 4500 kgBeing optimistic.. . shooting for 30+ knots.Dennis
From: Rick Willoughby <rickwill@bigpond. net.au>
To: harryproa@yahoogrou ps.com.au
Sent: Fri, July 16, 2010 12:48:47 AM
Subject: Re: [harryproa] construction
Your thoughts are similar to where I was 2 years ago. My flat panelled hull was aimed at easy building. I could not muster the will to carve another plug out of foam. I was prepared to give a little in performance with something easier to build.
I was very pleased and a little surprised when the end result actually performed better. So the flat panels was not a performance compromise. This was not quite what Michlet had predicted. The Flotilla software goes some way in providing the theoretical understanding of why it is better.It would be reasonably simple to make a long slender lw hull for a proa using ply that has very good performance. AND likely better performance than could be achieved with a round section hull. It could be designed to trim flat or slightly bow up through a wide speed range.If you want to have a go I can provide input on the shape that should perform best.
Rick
On 16/07/2010, at 2:21 PM, Doug Haines wrote:
--- On Fri, 16/7/10, tsstproa <bitme1234@yahoo. com> wrote:
From: tsstproa <bitme1234@yahoo. com>
Subject: [harryproa] Re: Rudder lift?
To: harryproa@yahoogrou ps.com.au
Date: Friday, 16 July, 2010, 13:25
What if the same 12m. hull was only 1/2 tonne displacement?
Todd
--- In harryproa@yahoogrou ps.com.au, Rick Willoughby <rickwill@...> wrote:
>
> The length constraint or non-constraint is a function of design speed
> and weight.
>
> As an example a 1 tonne hull with a flat bottom designed for minimum
> drag at 25 knots results in a length of 12m. However this hull will
> not generate enough bow up moment through wave making to counter the
> moment from the rig to drive it at that speed. Extending the length
> to 15m will get close.
>
Rick Willoughby03 9796 24150419 104 821
Rick Willoughby03 9796 24150419 104 821
Rick Willoughby03 9796 24150419 104 821
Rick Willoughby
03 9796 2415
0419 104 821
__._,_.___
Messages in this topic
(308)
.