Michlet/Godzilla availability

Thank you very much, Cesar!

The good agreement for the NPL4b might be a bit lucky at low Froude numbers. There is a lot of scatter in the actual experiments (about 25% at Fr=0.24). The experimental points in the graphs are from fits through the scatter of points.

Cheers,
Leo.

 

And I still do´nt know how "to craunch a marmoset" ! You?

 

Good agreement between Michlet and Tank tests

Some if you may be interested in this graph (attached) that compares tank test data and Michlet calculations for a 130m heavy duty trimaran. Both graphs show ideal outrigger position. Tests and calculations did not come up with the same ideal position (effect of bow wave due to bulb). But still, the overall minimum resistance for concept development purposes using Michlet showed remarcably good results.

Tank tests and calculations were performed by me at MUN, St.John's NL Canada 2005 as part of my PhD studies. I also used the software back in 2000 on a 280m 40 knots trimaran and compared to Marintek tank test results and eded up within a margin of +/-10% on resistance. I still have not figured out what Marintek did wrong

Thanks for making the software free Leo, a great tool for "first pass" analysis.

Regards,
Erlend

 

Erlend

Yes very interesting. Especially at around 8.5knots...around 30% difference, not a small amount!

There is an interesting Japanese paper by Suzuki, Ikehata in 1993 that looked at optimal position of outriggers.

 

Erland
Jokes aside, I actually agree - I would be looking for the error in the tank testing. Comparisons I have done against Michlet were initially within 2% of actual measured power to a prop at the design point. I asked Leo to adjust the allowable range of viscosity for the cold water testing we were doing, which he did, and the results were then in close agreement. Certainly within the 1% error in our power measurement.

There are many sources of error in tank testing and unless they are all account for they will be poor compared to what you will get with Michlet.

The little hump in your drag curve is interesting as it is opposite to what Michlet does at that point. I am sure Leo would have a view on the cause.

If I get data that varies with what Michlet predicts I look for other errors before I question what Michlet is producing.

Rick W

 

Thanks for the kind comments, Erlend.

I agree with Ad Hoc that some experiments are not very reliable, particularly for Froude numbers around 0.35.

Most codes (CFD and classical) predict large bow-stern wave interference around Fr=0.35, and that (beneficial) wave interference is not always realised in practice. Predictions around Fr=0.35 should always be interpreted as the the best attainable under ideal conditions; towing tank experiments should be treated with great suspicion because small variations in towing speed, vibrations and many other factors can induce large changes in the measured resistance. Wave cancellation (and reinforcement) can be quite sensitive to those changes.

For trimarans the difficulties are compounded. Towing tanks are not always wide enough to comfortably handle multihulls. Predicted wave cancellation can be significant (not just at Fr=0.35) and not realised.

If you are looking for other papers on trimarans, see:
Sandy Day, David Clelland and Ed Nixon,
"Experimental and Numerical Investigation of "Arrow" Trimarans,
Proc. FAST 2003, pp. D2-23 to D-30.

That paper was an attempt to verify the predictions that E.O Tuck and I made for optimum placement of trimaran outriggers, but it used demihulls with transom sterns.

The ITTC is starting to compile the results of the "Facility Bias World Wide Campaign". This campaign aims at testing the same model hull (one with L=3.048m and the other L=5.72m) at about 30 towing tanks. (See attached Proc. 25th ITTC - Vol I, The Resistance Committee
Final Report and Recommendations, 2008, Fukuoka).

The first few results are not particulary encouraging. For example, the uncertainties of measured trim are very large in some cases. Maybe the results will be better from other tanks. Or maybe not. It looks like a bit of a gamble at the moment.

For preliminary and "First pass" design, I tend to take the view of experienced naval architects Carl Scragg and Bruce Nelson who, after conducting an extensive experimental program with rowing shells, concluded:
"...the empirical approach is less reliable in discerning small differences in performance than the systematic results obtained from numerical hydrodynamics."
Scragg, Carl A. and Nelson, Bruce D.
"The Design of an Eight-Oared Rowing Shell",
Marine Technology, Vol. 30, No. 2, April 1993, pp. 84-99.

All the best,
Leo.

 

.."...the empirical approach is less reliable in discerning small differences in performance than the systematic results obtained from numerical hydrodynamics."..

Said without a hint of bias

Very interesting ITTC report. Shocking how many of the centres did not or do not conform to "standard procedure". I'll say there is a serious QA issue to be had at the very least.!!

 

I'm biased in that I am all for the hydro approach for thin hulls. For bluff bodies I doubt both experiments and theoretical predictions.

 

Look at the experimental results around Fr=0.35 in the attached small paper. There it can be seen that the results rise and fall very sharply. I would really like to see more results for that Fr region.

Without knowing the hull shapes, the demihull stagger, whether the demihulls have transoms, whether the vesssel was allowed to squat, and the depth of the tank and its width, all I have are empty guesses, Rick.

 

The coarseness of Erlend's data would hide such rapid variations. The test data shows a kick up and the Michlet data shows a kick down. Very small error in speed measurement would put the tank data on a peak rather than a trough.

Knowing how hard it is to measure accurately in the real world I would take the Michlet data. More data points, both from MIchlet and the tank data, would highlight the rapid variation in the region.

When we were tuning CP2 (http://www.adventuresofgreg.com/HPB/uploaded_images/P1060945-720324.JPG) last year I decided to do all testing at the design power level as it was one thing we could measure accurately.

We thought we were measuring speed accurately in an enclosed lake using a GPS. However there was a small flow into the lake and out of it. Hardly noticeable but it was enough to set up a Coriolis current. Everything looked good until the boat was circulated in the opposite direction. Current was around 0.2kph so difference of 0.4kph between the two directions. Got into a lot of debate over the offset prop and the amount of drag on the rudder to correct for the off-centre thrust. I had calculated 0.3W so this did not explain the speed difference because it required more like 10W. Was not resolved until the boat was tested in a larger lake that had no such current and the result was the average of the other two.

As you know once we corrected for the cold water in Calgary in May we got exactly what Michlet predicted. That was after getting the hull surface finish sorted, trimming the weight to design and removing unnecessary and oversize appendages per the design.

The lesson for me is that if you do not get what Michlet predicts then look for the error in the test data or input to Michlet.

Rick W

 

"...We thought we were measuring speed accurately in an enclosed lake using a GPS. However there was a small flow into the lake and out of it. Hardly noticeable but it was enough to set up a Coriolis current...."

So the lake, all by itself, decided to become effected by the earths rotation that is in itself enclosed, but at the same time has flow of water into and out of it???

Please explain that further....

 

ngineering is not enough because, according to Prof. Georg Weinblum, "The theory of ships is too difficult for engineers."

The quote comes from an interesting historical look at ship hydro by Prof. Ada Gotman of U. Novosibirsk.
"Navigating the Wake of Past Efforts"

well Leo there some of us that do things by feel still, feel, eye, and 'sperience, in small ships, can you understand this?
Keep up the good work,

 

"...well Leo there some of us that do things by feel still, feel, eye, and 'sperience, in small ships,.."
Indeed, just like this:
http://www.boatdesign.net/forums/metal-boat-building/metal-frame-detail-28371.html

...yes we can understand this kind of 'sperience easily.

 

You don't sophisticated software to show you how to make a log pointy at both ends

Cheers,
Leo.

 

Question for Michlet

Hi, question from a amateur beginner,
Trying to learn this excellent michlet program, interested in fast slender hull.
As i understand there is only this program able to calculate hydrodynamics for narrow hulls.
I have red the manual but cant find out how to change the beam width B and
the displacement Dtot ?