Uncategorized

cfd, bfd

counter counter rant

cfd, bfd

I have to respond to Mr. Falsone’s AC model claims in the article titled size doesn’t matter?

A smaller model is not cheaper than a big one – the cost is mostly labour
not materials and the labour is not less. Small models cost just as much
as large ones, and they take just as long to do the (meaningless)
experiment.
The cost of the models he quotes are hugely inflated and must include solid
aluminium foils for each model and a day rate for some consultant. Using
interchangeable foils and the cost is approximately 1/3 of those mentioned.
The error bars are not just greater at 1:5 scale – the results are
meaningless.

I suggest he speak to those with direct experience at this
scale, especially recently. It is true that work has been done at these
scales and smaller – but not by anyone who has ever won. PACT did indeed
do 1:8 scale and if my memory serves well, no winner resulted from that.
The absolute values of the drag predicted by CFD is still not within 20%
of the measured values. Users believe trends only after considerable
calibration with experiments. You cannot use the absolute values of CFD
drag in a Velocity Prediction Program. The promoters of CFD will not often
admit this, and it is even more difficult to judge the value of such work
if they are not embedded.
There is no longer any restrictions on where one goes to do model
experiments, and while there is now only one facility in the USA ( and no
they do not give yotties a break on the price), there are numerous such
facilities around the world.
Ken Davidson did indeed work at smaller scales, but they were bigger boats
he was modeling and differences between designs was huge. We now work in a
much smaller design space.

There have been many rumours about what happened with Mariner, this is new
one to me. This does not explain why someone went with a slower design.
Whatever the real reason it ruined the American taste for tank work, and
allowed the Aussies to win. To go back to a meaningless experiments is to
set the stage for similar thing to happen again.
The performance prediction of sailing vessels is arguably the most
difficult experiment in naval architecture, and there are hundreds of nav
archs around the world who think they know how to do this. The reality is
there are less than a dozen people in the world who have demonstrated they
in fact do know how to do it. Every one of them will tell you 1:5 scale is
meaningless.

Restricting experimental work to 1:5 scale is a deeply subversive means to
restrict the number of serious contenders. It does not reduce cost. It
does give established groups an almost insurmountable advantage. – Anarchist Bruce