[...]... D.W., Dudleys Gear Handbook, Ch 13 Gear vibration, McGraw-Hill, New York, 19 92 Newland, D.E.N., Random vibrations, spectral and wavelet analysis Longman, Harlow, UK and Wiley, New York, 19 93 Analysis Techniques 9 .1 Types of noise and irritation One of the most difficult problems in gear noise investigations is that the final "detector" and arbiter (on whether or not a noise is irritating) is an extremely... cycle The sampling rates needed for the envelope are much lower than for the original vibration out ground ampl Fig 8 .10 Former method of enveloping vibration signal 13 4 Chapter 8 rectified signal ampl Fig 8 .11 Preferable method of enveloping This type of information may be relevant for looking at I/tooth frequency and its modulation due to varying misalignment or torque effects or looking at high frequencies... on gears which have already worn out but it is extremely difficult to take a decision on the time scale for killing off old records This is one problem to which there is no satisfactory solution but the more compact the storage the longer can the decision be delayed Recording and Storage 13 7 References 1 2 3 American Gear Manufacturers' Association AGMA Standard 6000A 88 Dudley, D.W., Dudleys Gear Handbook,... at 12 kHz, which is an ear-splitting frequency The third aspect which can be different in the particular case of gear noise investigations and checks is the permissible frequency range The text book approach to vibration analysis may be extremely worried that "aliasing" problems with, say, a 10 kHz sampling rate might mean that a 6 kHz vibration is wrongly identified as a 4 kHz vibration As far as gearing... driven by a TTL oscillator such as an 80 38 at 10 0 times their required rolloff frequency There is also a limitation that the maximum input voltage is limited to about 4 V when the rails are at 5 V This requires that an input signal is reduced to below 4 V, filtered then re-amplified to return to the original size 13 1 Recording and Storage +5V +15 V output BNC Fig 8. 8 Circuit for double 5th order low... frequency and so can be ignored This allows the use of a lower performance and hence more stable 13 2 Chapter 8 filter which is less prone to ringing or the use of a much reduced frequency margin between harmonic and carrier 8. 6 Information compression Although modern PCs have relatively large (tens of gigabytes) hard disc memories and the initial investigations of a problem will require raw vibration. .. quantities but are not in general use yet and formats have not standardised A suitable compromise for vibration work or T.E tests is to store selected small files such as the most interesting mesh cycle averaged files (as in section 8. 6) These averaged files contain typically less than 4 k points and so are only 8 kB, allowing noise test results from thousands of tests to be stored on a CD Rom, easily... revolutions, average the vibration signal over the revolutions so that only vibrations associated with that shaft remain, as all other nonsynchronous vibration (and electrical noise) has averaged to zero Displaying the vibration on an oscilloscope synchronised to once per rev has much the same effect since we tend to average out visually what we see on the screen If we have a standard 15 00 rpm motor driving... tooth pinion meshing with a 11 9 tooth wheel, then we must complete 11 9 revs of the pinion to complete a meshing cycle, and all subsequent meshing cycles should be identical so there is no point in measuring any more complete meshing cycles since the same information should appear again and again This will take 4.76 seconds for the cycle, and with tooth frequency 600 Hz and a requirement to measure... rules quoted in textbooks 13 5 Recording and Storage output input Fig 8 .12 Circuit for accurate rectification of small signal The standard sampling rules do not seem to apply for problems such as this where the main requirement is to have the area under the envelope roughly right Practical testing with an artificially generated signal with bursts of perhaps six cycles of vibration and testing by varying . delayed. Recording and Storage 13 7 References 1. American Gear Manufacturers' Association. AGMA Standard 6000- A 88. 2. Dudley, D.W., Dudleys Gear Handbook, Ch 13 Gear vibration, McGraw-Hill, . exactly. Recording and Storage 12 7 ampl 1 f II /I I 1 maxfreq of interest N ' 1 1 i i i i 1 1 1 r i i i i i t 20 Hz T sample frequency filter characteristic 4kHz 5kHz frequency 15 kHz Fig 8. 5 Typical . rev Fig 8 .1 Simple T.E. trace. 12 1 12 2 Chapter 8 vel 1 rev Fig 8. 2 Complicated vibration recording. " ;Noise& quot; in this context is used to describe any electrical or mechanical vibration