Hydrodynamic model

Dear all,

I've been contemplating the effect of seperation on transom stern hull forms. While trying to envisage the effect that has on a resistance; ive come up with a few comments to help me get a grip on things. Please comment on the validity of the following:


Assuming wave pattern interference effects are negligible at Fn=0.5; take the case where the flow on the transom seperates at Fn=0.51; would it be correct to assume the following:

1-) seperation of the flow causes the shifting of the sink ( in a kelvin wave analogy) further aft and thus wave interference effects could rise up again to create an unfavourable constructive interference effect further aft. What would the effect of seperation be on Fn (i.e. is there a dynamic length somehwere that should be considered as a result of seperation).

2-)Also, there would be a net increase in viscous pressure resistance as a result of hydrostatic pressure loss in the stern region.

How is that ( if necessary) accounted for in thin ship theory (i.e michlet)?

 

You can try to find one, but it is unlikely to be consistent, i.e. it might work well for one type of hull and produce poor results for another.

Michlet is an inviscid code. It doesn't attempt to simulate or estimate visocus pressure resistance.

Michlet has many different transom stern models that can be used: e.g. a completely consistent thin-ship method, and many different (inconsistent) empirical formulations such as those developed by Doctors and Day, and Molland, Couser et al. The free version (8.07) allows only one of the many variants available.

The results generated by Flotilla (see the pdfs in http://www.boatdesign.net/forums/design-software/flotilla-demo-2-07-a-32276.html) were obtained with a series expansion model of the transom flow and a single empirical constant for all 10 monohulls and 40 catamarans. The results compare very well with experiments for the NPL series. I also have very good predictions of the DTMB 5415 destroyer hull, AMECRC series, and Series 64 hulls, but I wouldn't bet it would be appropriate for many other hull series. That's the danger of using inconsistent formulations.

If you are keen to study these issues further, please search for papers by Kevin Maki, Lawrence J. Doctors, Alexander H. Day, Patrick Couser, Robert Beck, Anthony(?) Molland, and Jerome Milgram. I think you will find that nobody really has a handle on this very tough fluid flow problem. CFD can give you a wrong answer just as well as other approximations, but it will take longer to do so

Good luck!
Leo.

 

Thank you leo for reverting back. I've read a lot of your posts in this forum and your genuine contribution in the field of hydrodynamics deserves true credit. truly a hydrodynamic wiz .

I have read through numerous papers in the past few hours ( including some of the papers you have co-written about linear potential theory).

Regarding the molland,couser, and wellicome paper about virtual appendage model; is it fair to attribute the different between model Cw and theoretical Cwp due to the presence of wave breaking resistance and interaction between viscous pressure and wave pattern resistance?


Also regarding the work on linear potential codes ( or even nonlinear); i have noticed few studies on low-L/B hull forms for minimizing of wave pattern interference drag ( i.e. for tugs and other offshore vessels) Is there a particular reason for that. i have seen alot of commercial software companies advertising codes to calculate effect of bulbous bow and hard shoulder on wave interference....Do these codes really offer what they are claimed to give?


I'm trying to dig deep into hydrodynamics for commercial applications. Are these codes to be trusted in commercial fields where mistakes of an order of 0.1 of a knot can result in serious economical reprocussions. In some extreme case, 0.5 of a knot difference can lead to vessel rejection from owner.



Thanks in advance!

NavArch007

 

If you believe that resistance can be usefully broken up in such a way. I'm not entirely convinced.

Patrick Couser might be able to help you more. I suspect he would say that the appendage model is one way of improving agreement between predictions and experiements for large Froude numbers. I'm not sure he would attribute the difference to those causes unequivocally.
Personally, I have grave concerns about virtual appendage models and the way they generate waves, but that's an old bug-bear of mine and E.O. Tuck's I won't bore you with.

I suspect they can give excellent, perhaps perfect, results if they are able to use a few free "empirical" parameters. On the other hand, "give me two free parameters and I can fit a curve to my grandmother's fat ***". (As a professor once said to me).

Remember that bluff ships tend to make large waves that break. Nobody really knows what happens to the drag then because turbulence is still so poorly understood. CFD will give you beautiful pictures of such flows if you want to see ignorance in colour.

No, they can't be trusted to give that sort of accuracy. Ultimately you will still have to rely on towing tank experiments. Even then, the experimental results you get will depend on the towing tank you choose from a hat.

There is a world-wide collaboration in progress that is measuring the drag and squat of two 3.0m and 5.7m geosims. I have seen some results from the first few tanks (of the 40 in the scheme) and the results are not all that consistent for the smaller model. There could be differences of around 10% for the resistance at low Froude numbers.

On the same note, Grigson attributed the 5% under-powering of some large ships to inadequacies of the ITTC skin-friction line and to extrapolations using that line.

Accurate to 0.1Knot? Har-de-har, mate!

All the best,
Leo.