Delft resistance formula

Studying Delft formula I conclude that, by the formula, minus all (Cp, LCB in % Lwl, Bwl/Tc, etc) minus resistance, totally against the old rule:
more velocity ---> more Cp

You can see this using the annex spreadsheet. Change the values in green cells only.

What you think about?

Fred

www.frediom.wordpress.com

 

I have not found all the errors but it does not seem to produce anywhere near realistic drag.

One error is in cell g77.

As you have noted the useful range of Cp is very narrow so if you are using it at the top value or beyond this value the results will be doubtful.

You should have the spreadsheet test the input data to ensure it is in a valid range. At this stage you can enter completely inconsistent parameters and it still gives a result.

I entered data for a 3 tonne yacht with a LWL of 8m, BWL 2.5m, draft of 0.6m and it yields a drag of 12N at Fn of 0.275. I think highly optimistic.

Rick W

 

Hi Rick

Ok, an error in G77, but this cell do not enter in the calculations, I Use only to correspond the Froude Number to a velocity.

The only parameter that is out of the rstrictions is Bwl, but we can change to Bwl = 0.20 m. I go test with this value, but I think that do not change nothing.

The problem is: to a specific Froude number, so to a certain velocity, if we change Cp from 0.52 to 0.60 the little resistance is with 0.52.
If you do the same for several velocities the same occur, the less resistance is with 0.52.

For hight velocities the lesser resistance we ever assume that is with hight Cp.

The problem that I do not understand is this.

Annex the spreadsheet corrected.

Fred

 

Fred
I know it does not enter into the resistance term but I was looking to see how well your spreadsheet works with a boat I am familiar with. It was immediately apparent that there was an error in the speed for the Fn once I was at a familiar scale.

You need to check the model with some measured data to make sure it makes sense before you use it to draw conclusions. The drag predictions seems lower than I would expect. Work out how much power you need to drive say a 5t boat with LWL of 10m at 8kts. See if it relates to something realistic. Provide an answer using your spreadsheet and I can see if it is realistic - I would expect of the order of 10kW on the hull.

Also the limit on the ratios are not the only consideration. I am not sure if the Delft series was ever intended to work with 4kg boats. The boundary layer conditions are quite different at this scale than at typical yacht scale.

Rick W

 

Rick

Ok, but I do not know if you see that only the residuary resistance is analised.

The results is only a part of total resistance.

I will go test with bigger boats but I think that if we maintain the parameters in the contest of the formula restrictions is valid for any displacement, is a matter of scale, but it is possible that I am in mistake.

Thinking well I only will change the numbers proportionally. I think that this will not affect the results. For change the results iI will need numbers with different signals in determined parts of the formula, that this will not occur.

Well, go to the numbers. Tomorrow I will test for big boats.

I am very thanks full to exchange ideas with you, I was needed for this.

Really the flow is in different Reynolds number but I think that this influences more frictional resistance, and the formula do not take this in consideration in residuary resistance.

Fred

 

Fred
Yes I do realise this now that you are only determining residuary. My comment on discrepancy remains applicable.

I have changed lwl to 10m, Bwl to 2.5m, Tc to 0.5m, Vol to 5Cu.m and Displacement to 5000kg. The residuary resistance it provides is 159N at 8.16kts. This is about 10% of what I would expect.

I have looked carefully through your formulas and cannot find any errors. I have also checked the polynomial constants and they all look correct. The only thing I have not been able to confirm is that the units are consistently MKS. If the drag was kgf rather than N it would be close to what I would expect.

I know Martyjn of Delftship has incorporated the Delft Yacht Series polynomial into the software so he might be able to give you an idea.

Rick W

 

Using:

Lwl = 5.57 m
Bwl = 2.03 m
Volume = 1.71 m3
Displacement = 1710 kg
LCB = 5.6 % Lwl
Tc = 0.47 m

We have for:

Fn = 0.4 --> V/Lwl¹/² (knot. ft) = 1.34 lesser resistance is in Cp = 0,59

Fn = 0,375 --> V/Lwl¹/² = 1,26 lesser resistance is in Cp = 0.58

Fn = 0,325 --> V/Lwl¹/² = 1,09 lesser resistance is in Cp = 0.53

Fn = 0,30 --> V/Lwl¹/² = 1,00 lesser resistance is in Cp = 0.52

The old rule is:

V/√L Cp

1.0 0.52
1.1 0.54
1.2 0.58
1.3 0.62

Well, I think that's ok

The reflections are:

1 - that according with rangers of parameters we need different formula's to represent the resistance and I can not use these formulas to preview the resistance of my IOMs.

2 - Are we in hour to review the numbers of Cp versus Fn?

Much thanks to Rick to give energy for my thinkings and if someone else want say something I will very happy.

Thanks Rick, it's possible that we have some mistake, but is difficult to find, but I wiil continue in the work to test the spreadsheet and the results.

 

Fred
If you are looking to optimise hulls for a particular purpose then you may find Michlet/Godzilla to be far more useful.

The Delft Series covers quite a narrow range of hull shapes that could be far removed from what is the optimum of all possible shapes. So you may come up with the best result for that series but there could be a whole range of other hull shapes that should be explored.

Rick W

 

Thanks Rick, I will go to study Michlet/Godzilla.

Really my intention is optimize IOM sailboats design a one meter RC. As I use Delftship to design the hull naturally I was studying the formulas that they use. But I see, thanks to you, that is not a good way.

 

Fred
Post #25 on this thread gives some ideas on running Michlet/Godzilla:
http://www.boatdesign.net/forums/design-software/racing-canoe-19110-2.html

There are some files there that you can use to start with by setting parameters to suit what you are doing.

If the program is not available on the cyberiad anymore then you will find it posted in one of the Michlet threads on this forum.

GODZILLA will get you up the learning curve very quickly. You can see how the hull changes to changing constraints.

Rick W

 

Thanks Rick I shall need informations about.

Fred

 

Very good Michlet/Mozzilla.

I use linux with wine and I run Michlet without problems.
I run Michlet with the file exported by Delftship with the offsets and parameters of my IOM without problems.
Meantime I not attain run Godzilla. I change in.mlt in Michlet directory by gin1.ml, change the name to in.mlt . Michlet open and do all.
The problem is : what I do to run godzilla.bat?

 

Why did you use the 1992 paper, and not the 1998 paper ? http://www.hiswasymposium.com/pdf/p...ion of the hydrodynamic forces on a saili.pdf

Note also what you have computed is residual resistance of bare hull, flat water, no appendage, no heel , no leeway , no wetted surface (viscous) .



Now, I am not sure a model can fit delft parameter validity range.

 

fcfc

Was the unique that I obtain. I am thanks to your ṕaper indication I will study it.
My objective is study and optimize a hull for IOM class, a one meter Lenght radio control sailboat.
In this class, lenght, canoe draft and hull total draft (with keel) and sail area is fixed.
My first problem is study the hull downwind, more later heeled.
When I tested the Delft series for hull I realized that I do not have convergence and less the values for the parameters, lesser the resistances.
I conclude that delftship formula 1992 do not served for IOM boats, still the delft parameter validity range is attended.

Thanks

Fred www.frediom.wordpress.com

 

You need to double click on godzilla.bat. Michlet will open the same way as before. You then have to press <shift> <G> to make it run. You should see the hull shape change as it searches for the optimum.

Rick W