![\"\"](\"\/\/ir-na.amazon-adsystem.com\/e\/ir?t=svrshdw-20&l=am2&o=1&a=B005CQ2ZY6\")
(which I really like) and ran them through an FFT program (FFTExplorer by Paul Lutus, at arachnoid.com, it\u2019s free as long as you don\u2019t whine).\u00a0 I found the best recordings were made with the microphone in contact with a cabinet in the aft cabin.<\/p>\nThe data in the first chart, taken at about 1800 engine RPM, shows a large peak at 47Hz, or 1.5 times the engine frequency. This is the sound of the three blades passing the deadwood \u2013 a 2:1 reduction ratio and three blades give a factor of 1.5, so eg at 1800 engine RPM (= 30 Hz), the primary prop peak is at 45 Hz. There is also a very tall but very narrow peak at twice this value, showing that the prop is indeed making noise at both the upper and lower sides of the aperture. Alan Watts, at Axiom, told me this might be the case \u2013 it had not occurred to me before. Nice to see it in the data.<\/p>\n
In between these two peaks is another at 62 Hz which is the engine RPM x 2.<\/p>\n
Also of note, there is no significant component at the shaft rotation frequency (15 Hz). This is important because I would expect that component to be large if the shaft is vibrating, for instance due to a loose or poorly fitting bearing, or if the prop were unbalanced.<\/p>\n
![\"\"](\"http:\/\/svrainshadow.com\/wp-content\/uploads\/2011\/08\/PropFFT1.png\" "\\"FFT")
<\/p>\n
The next chart is for 2200 RPM. At this point, the noise seemed very loud. The vertical scale is the same in both charts. The first peak is at 56 Hz, the next at 114, then 170, then 226 corresponding to the prop frequency x1, x2, x3, and x4.<\/p>\n
It\u2019s very interesting that so much of the energy is in the higher harmonics of the prop frequency. But only up to the 4th harmonic \u2013 not the 5th. The 4th harmonic might be related to shock waves from the leading and trailing edges\/corners of each blade \u2013 that would give you twice the frequency. So the x2 peak would become a x4 peak. But I don\u2019t know where a x3 peak would come from!<\/p>\n
Qualitatively, these two plots are very different. It\u2019s as if a new process is occurring. The subjective difference in sound between 1800 and 2200 RPM is dramatic.<\/p>\n
![\"\"](\"http:\/\/svrainshadow.com\/wp-content\/uploads\/2011\/08\/PropFFT2.png\" "\\"FFT")
<\/p>\n
What we learn from the FFT data, is that the interaction between the blades of the prop and the deadwood is dominating the sound. That\u2019s a bad sign.<\/p>\n
Looking at the design of the Axiom propeller, you can see from the photo below that unlike a screw prop, the blades do not lean backwards from the hub like a screw does. (Yes, we painted the hull blue). This means that the clearance between the blades and the deadwood is much less \u2013 in fact, we measured it at about 1.2 inches, which is far less than the recommended 2.5 inches (at least 15% of the prop diameter, per Gerr). This probably explains the excessive noise from the blades passing the deadwood.<\/p>\n
![\"\"](\"http:\/\/svrainshadow.com\/wp-content\/uploads\/2011\/08\/Axiom1_Screw.png\" "\\"Axiom")
<\/p>\n
At this stage, we contacted Axiom. They had always assured us they would get us a prop that fit or would give us our money back, and they stuck by this promise. (This was a much more pleasant experience than we had with North Sails last fall).<\/p>\n
We agreed with them it would be unfortunate if we had to return it and get our money back \u2013 and agreed they would make us a second prop. This one would have the blades set further back on the hub. And to correct the over propping, it would have a 15.25 inch diameter with a 19 degree shear.<\/p>\n
Next: Part III<\/a>, where we report on the 2nd replacement Axiom Prop<\/p>\n