CFD state of the art

The way I interpret the heave-sink curve, VCG is again higher than static at Fn o,55 or so. Although probably much of the VCG motion is due to the pitch attitude, rather than real heave-sink. Is the combination of heave & pitch what Leo usually calls squat?

 

My first impression was that there must be a mistake. I double checked it once more, but all the coefficients in my program are identical to the ones published. I cannot find an error. May be somebody else has the time to cross-check.

This regime seems difficult to simulate. The Open-Foam results are so far off, that I left them out.

I am currently re-thinking my whole simulation-strategy. When I use a curved waterplane and extend the wetted surface to the transom, I get the correct viscous forces. I must accept that the hydrostatic waterplane at rest is not a good approximation for such a hull.

 

The still-water level (SWL) is at Sink = -0.01
above FN=0.5 the boat rises above the SWL in the Star-CCM simulation. It stays below the SWL in the Open-Foam simulation. Which one is right?
Uli

 

The boat knows.

Thank you for great contributions in this and other threads Uli!

 

the wave pattern along the hull is a function of Fn. One can very easily calculate how many wavelengths along the waterline exist at each Fn. What is then missing is the height od the peaks an troughs. This can be back calculated from the wave resistance you get from your regression. You just need to back calculate what wave height for a kelvin system gives that energy.
Then you have all you need to get a better wetted area estimate.

 

Yes. It is more than just the heave, or sinkage, component.

It isn't quite that simple because every part of the hull produces waves
that interfere (constructively or destructively) with waves produced
upstream, and on the bow wave in particular.

Remeber too, that the trim and heave are relative to the carriage,
not to the water surface around the hull. The water around the
hull also sinks and trims because of the pressure field around
the hull.

 

I am rather confused now.
I compared the different sink-data in the attached diagram. I can understand, that the CFD-data was shifted by 10 mm from the line z=0 to the real still water level at 197 kg. What I don't understand is the tweaking of the Star-CCM results. The only explanation that is given in the files is the text, that I have copied into the diagram. I am not trained in the exegesis of cryptic texts.
May be someone else can explain.
Uli

 

As I read it, it says that the heave data was originally measured from the model origin (what was that, the base line?), and then corrected to show the distance from the displacement waterline (still-model waterline, I guess).

 

My impressions about the heave data were from Figure 22 of the published report. It is the relative heave from static (zero speed) that is important - and the heave is negative (sinkage) across all speeds. (Trim may cause the bow to appear higher...)

Don

 

Semiplaning??

I am calling for a better word for semi-planing, because it has little in common with the planing. The word is misleading for me, as it has far more of wave formation and wave interacting at speeds around Fnl 0.7 to do.

All suggestions are welcome.

The wave-forming length is furthermore not the same as the waterline length. So Fnl is also not entirely correct.

js

 

Yes, I realized that. I was only looking at the uncorrected Star-CCM results. It appears that they were moving LCG forward by about 36 cm from Fn= 0,1 to Fn 0,8 (to simulate a pitching moment from sail drive?), so with the bow up trim that would make a difference in heave, measured at the LCG vs. measured at the original towpoint 2.484 m aft of bow (Fz - Heave (m); vertical displacement, parallel to the waterplane measured at 2.484 m aft of datum referenced to the static zero speed and LCG condition, positive upwards).

But then, as Uli points out, the influence of the correction on the Open Foam sim should have been similar, and it was not (?).

 

Mikko:

I was only looking at the CB-1 (canoe body only) case to isolate the hull from any other influences. This gives me a sense for how the calcs handled the simplest case...

Don

 

I have done some "reverse engineering" to understand the correction to the Star-CCM raw data.

In the report from Cape Horn Engineering one reads:
"Sinkage report: in our case is just the vertical shift of the Centre of Gravity of the boat, i.e. a different point for each run."

In the report about the tank tests one can find the sentence:
"Heave (m); vertical displacement, parallel to the waterplane measured at 2.484m aft of datum, referenced to the static zero speed and LCG condition, positive upwards."

To correct the Star-CCM-values it is necessary to subtract the following value:
(delta Z) = trim angle * (delta X). Where delta X = (distance between X-position of COG and X=2.484m).
This (delta X) varies between 0 and 0.357m because the position of the COG varies with speed. Backward calculation reveals, that instead a constant "delta X" value of 2.37m was used to calculate the "delta Z" that was applied to the Star-CCM raw data. Where this large value is coming from and why it is necessary is nowhere explained.
Uli

 

Uli, I think that you have done a great job with the available data. Now it's up to the authors of the research to come out with some explanations and answers to the issues you have raised.

 

Just a note that I've passed the question to SYRF/Myles Cornwell. This is probably a question that should be addressed to Andy Claughton but I'm afraid he's busy with more important things (like BAR AC racing).