Problem Michlet import from freeship

On the DDG51 destroyer example referenced above, replacing the final row of offsets with a copy of the second row causes it to underestimate total resistance by roughly 3.5% at full speed (20 m/s), compared to the case where the actual transom offsets are used.

Because Michlet takes as inputs a set of offsets, a length, a displacement, and a draught- but not a beam value- the slightly larger than actual transom (thus slightly larger than actual volume between the last two stations) created by copying the second-last row serves to narrow the overall beam very, very slightly, thus there is a very small drop in wave resistance when compared to the actual transom. The viscous drag term is not significantly affected.

(Incidentally, this can be legitimately interpreted to mean that the DDG51 as drawn is a tad too wide overall and a tad too narrow in the stern for a 20 m/s target speed. But the military may have other priorities- like stability.)

To avoid this problem, it is obviously best to use the actual transom offsets, but if you must copy the second-last row over, use as many stations as possible so that the affected volume will be minimal. (I always use 81.) Hulls with fairly straight lines aft will not be affected as badly. In the Michlet line drawings screen, always check the horizontal cuts view (H) to be sure the straight bit near the transom does not significantly change the shape of the hull.

Incidentally, Freeship/Delftship does get it right now and then, maybe one time in three for me. No particular pattern has emerged as to when it will or will not translate the transom correctly.

 

3.5% is quite a lot. I have never get a proper transom from freeship,
About transoms from Godzilla: is it always giving transoms with shape and size of midsection for high speeds?

By the way do you have any mean to account for changing trim and sinkage with speed? I know that there is such an option in in.mlt file but how can I pinpoint right values at given speed?

One more curious thing for me is what is ship loading type used for... is it only for KG and GM calculation or it affects drag results somehow?

I know it's a lot of questions but i do not have any other source to get answers. It is so good to find someone who is also playing with Michlet Fighting with it alone is not an easy task...

Michal Klimek

 

The Series 20 produces the transom without any constraint other than a maximum length and an optimum speed somewhat higher than hull speed.

Rick W

 

As I said, the 3.5% is for the DDG51 example; if you have a high station count and are using a hull that doesn't taper in too quickly at the stern, I have found about half this error to be typical. In any case, I always assume at least a +-5% margin of error on calculated resistance values, sometimes even more if the hull might be prone to trim issues.

Michlet as you and I know it cannot calculate trim and sinkage, although it can adjust the calculation accordingly if you already know the trim and sinkage. Leo's "Flotilla" code is reportedly capable of calculating this, but it's not available yet and we haven't been told how much it will cost when it is.

Ship loading- the only valid option is type 3, and this determines the formula used to compute the height of the centre of gravity above the baseline. CG height has no effect on drag, but is an important factor for the stability numbers.

I often get a transom with Series 8 when optimizing for higher than hull speed, very rarely or never when optimizing for lower speeds. I would love to play with the higher-parameter series, but thanks to a corrupted alleg42 dll (that I know Leo fixed, but I lost the corrected version somehow), my 8.09 executable can't run.

 

These questions are answered elsewhere on the forum. Search for Leo Lazauskas and read his posts for the last couple of years.

The loading type will not alter the drag calculation.

Michlet does not iterate sinkage or trim automatically. There is a later version of the software called Flotilla that I understand takes this into account but it is has limited distribution and is still being developed.

If you are operating at planing speeds then Savitsky calculator will give more reliable results.

Rick W

 

It will be great to try Flotilla then... Hope it wouldn't be too expensive.
About Savitsky calculator: length to beam ratio 10:1 and design speed about 20 knots with 22.5 meter 18,5 tonn hulll... It gives planning speed at about 30 knots so it will be probably not planning ( i don't know if my formula can be used for such slender hulls - displacement Froude number equal to 3 or more=planning).

Thanks for help

Michal Klimek