That beeing said, it´s nevertheless an extremely fast battleship, faster than what IOWA was able to attain.
That speedometer had been calibrated against range speeds during the standardization trial, and accordingly, the model basin trials crew was able to reconstruct an additional data point at 31.0 knots of 198.2 RPM and 186,400 SHP. This being substantially closer to full power than the World War II data, extrapolating it as a cubic is correspondingly more accurate. The result is 32.36 knots, in very good agreement with the classic design number of 32.5.** Since this trial was run 71 days after the last drydocking, suggesting there could be some hull fouling, I am inclined to believe that 32.5 knots at 212,000 SHP is within experimental error of what actually happened on the sea trial in 1985. "
"There has been lots of interesting discussion on this thread since my original article was posted. The biggest questions seem to be extrapolation from trials data since there was in neither sea trial a point at 212,000 SHP.
I have looked at this more carefully and indeed the commonly used cubic interpolation method does not work well in the region around 30 knots for the BB 61. In the sea trial numbers, taking the point at 28.08 knots and extrapolating it to 29.41 knots, the next point measured on the trial, the cubic results in a power about 12.3% less than what was measured. Coincidentally, extrapolating the 29.41 knots to the reconstructed 31 knot point gives a result that is about 12.5% short of the measurement. The reason why these numbers are so similar is "coincidental" because of the shape of the drag coefficient curve for the ship. This can be modeled somewhat (resistance only, ignoring what happens to the propeller efficiency) by making a Taylor series calculation.
The Taylor Series is a large number of model tests, first published in 1910 as “Speed and Power of Ships” by RADM David W. Taylor, USN, that can be interpolated to create a resistance curve for a ship with a wide range of dimensions and coefficients. Modelling an Iowa class battleship this way results in the following table (I can’t seem to make graphs as some of the others in this thread did).
If we look back at the table as a model for the resistance behavior implied by the trials curve, however, there is a difference. Extrapolating the EHP curve in the table from 29 to 31 knots (representing, approximately, the last pair of points on the trials) as a cubic results in a predicted EHP at 31 of 91,417, or 6.6% short of the calculated resistance – about half as bad, in short, as the trials behavior. The trials behavior is quite complicated, with propeller efficiency steepening the curve in the region of 30 knots. To get SHP from EHP we add in the appendage drag and still air drag and then divide the total by the propulsive coefficient, PC, which is typically around 0.64. If we take the next pair of speeds, 31 and 33 knots to represent the extrapolation I performed as a cubic on the trials data, we find that the result is 117,585, or about 5.3% less than a cubic. There is no reason to assume that the propeller efficiency will decrease less in this speed range than it did between 29 and 31.
Accordingly, I agree that the Iowa’s top speed in the conditions of the trials will be less than the cubic extrapolation, maybe about 32 knots. However, this changes the stature of these battleships very little since this is still a very creditable speed for a ship this size.
The question of the difference between Iowa and New Jersey is one I didn't really look into when I first saw the DTMB report. The Rockland measured mile was in use for a substantial time and was considered deep enough to get good results for most ships. While the displacements were different, the differences were small, and I tend to agree with the authors of the report in saying ship surface smoothness was probably the biggest difference that could affect results.
However, one of my main reasons for requesting the new trial was that I had access to the original BuShips report of the New Jersey trial. I analyzed it and concluded it was inadequate. I don't recall why I concluded this -- possibly I didn't like the fact that the highest speed point was so far away from full power. (I didn't realize, of course, that "my" sea trial would end with the same basic defect). I'm a naval architect with no special training in statistical analysis and I didn't want to do the kind of fancy extrapolations that delcyros has done in this thread. But, it's also possible I found some other problems with the data. Anyway, I do not think it is correct from a physical standpoint to attempt to make a composite curve from the two trials. The two ships were different for one reason or another and the two trials must be analyzed separately. If that means Iowa in her last commission was a little faster than New Jersey was in her first, so be it. The ships of this class were built in two different yards, which means the lines were lofted by two different teams and the "as built" hull forms may not have been identical. The underwater paint was certainly different. While propellers were to the same design, those used in the Iowa trial might have been reconditioned using modern facilities; I looked at them in drydock on one of the ships after renovation and they were certainly beautifully polished. Accordingly, I would caution against lumping the two curves together. Iowa in 1985 was a little faster than New Jersey in 1943, but those differences were relatively small and did not affect the ships' tactical usefulness had they been operating together. They still deserve bragging rights as the fastest (well, maybe just one of the fastest) battleship classes ever built.
dunmunro wrote: For example Iowa would not be able to allow her displacement to fall below ~55000 tons due to the need to maintain liquid in the outboard layers of her TDS.
Users browsing this forum: No registered users and 1 guest