Haslar
8th June 1921
D.N.C.
SPEED TRIALS OF H.M.S. "HOOD".
Analysis of Results
1. The Speed Trial records of the "Hood" forwarded with your reference of the 20th August 1920, have been analysed, and the results are shewn on the accompanying tracing in the usual "pillar" diagrams i. These apply to the results for all four shafts collectively. Plottings are also made for the S.H.P./D.H.P. (see id) and "x", (i.e. 
, see Mr. Froude's 1908 I.N.A. paper) for the individual shafts. Sheet I shews the results for the Deep, and Sheet II for the Light draught.
2. It is a matter for regret that screw experiments for the "Hood" model (YA), were unfortunately only made at high speeds (see Report of 17th Jan 1917), so that the values for the Hull Effy. Elements had to be inferred for the lower speeds. This was done by adopting the same character of curve as was used for the "Renown" and "Repulse" trial analysis, which is fairly well established by a study of the results for similar models (see report of 2nd Feb 1917). The curves adopted are shewn on Sheet I.
i
3. There is no evidence of cavitation although the pressure coefficient (from the Model T.H.P.) at the top speed exceeds 80 tons per sq ft of propeller area. The propulsive coefficient is well over .5 at the high speeds. The analysis of this results as recorded reveals a few inconsistencies, notably a fairly large apparent underestimate of speed or over-rating of revolutions in Trial No. 6 (at a reported speed of 28.3 knots) of the Deep Draught Trials. The basis of the evidence bearing on this, and other apparent inconsistencies is detailed in Appendix I.
4. The drag due to the inner screw running idle at 13.09 and 15.17 knots in the Light Draught Trials is seen from the isolated "pillar" plottings of the H.P. constant at the speeds stated (Sheet II). The extra power needed to overcome the negative thrust of the screws appears to be about 12%, and this (when corrected for the screw efficiency) may presumably be regarded as a measure of the total friction of the propeller, bearings, gear wheels, and "windage" of turbine. The revolutions of the trailing screws were not recorded on the mile, but can be estimated from the appropriate three hours trial, which indicate a negative slip of nearly 18%. It might be expected that these trailing runs would give some idea as to the speeds of the wake stream past the screw disc, but the large negative slip experienced renders any accurate investigation of the wake from such experiments quite out of the question. Exterpolating our propeller data to negative slips (which is of course quite and assumption) it is found that the negative thrust developed by the propeller is in fair accord with the above, and this may be regarded as a rough confirmation of the wake values adopted at these speeds.
5. A striking feature of the analysis lies in the excess of 
over 
at all speed, i.e., that the T.H.P. as calculated from the naked model resistance experiment is always greater than that calculated for the screws at the revs and speed of trial. The excess becomes even more marked if we add to the resistance due to the appendages, and points of course to error in our estimates of , or , or both. This is a point of great importance, and of the trial data of capital ships available, a careful survey has been made to find similar cases. The only other available results shewing a comparable excess are those of "Temeraire", "Vanguard", "Iron Duke", "Indomitable", "Inflexible", and the "Renown" and "Repulse". There seems to be no common characteristic of these vessels with the "Hood", apart from the last two, as regards length, and of the trial results examined (43 in all), the excess of over was greatest in these vessels.
6. It is of interest to examine the possible causes to which this excess of over may be attributable. These are considered to be:-
(a). Over-rating of E.H.P. The extension of our model resistance results to ship figures necessarily involves, as is well known, a purely empirical assumption concerning skin friction, the data for which was obtained from experiments on short planks (maximum length 50ft.), moved at low speeds (maximum 900 ft per min. - this speed was only associated with shorter lengths than 50ft.). The method employed at Haslar of calculating skin friction is essentially the same as that suggested by Mr W. Froude when the skin friction expts. were made in 1872, and any error involved may reasonably be expected to increase with increased dimensions and speeds. Another semi-empirical method of obtaining the skin friction for the ship evolved by Mr G.S. Baker (of the N.P.L.) in 1915, gives E.H.P. values generally less than those obtained by the Haslar method - the general effect can be judged from the plottings of Sheet I, from which it is seen that Mr Baker's scheme of allowing for skin friction leads to an estimated E.H.P. os about 18% less . . .
End of transcription