ISO/TS 10811-1:2000 pdf download

ISO/TS 10811-1:2000 pdf download.Mechanical vibration and shock —Vibration and shock in building swith sensitive equipment — Part 1: Measurement and evaluation.
The time history of the vibration should be recorded and analysed along three orthogonal axes (preferably in the vertical direction and in two orthogonal horizontal directions).
The measurements shall, where practical, be made with the shock-and-vibration-sensitive equipment in place, or with a dummy having the same mass and similar dynamic behaviour as that of the equipment under consideration. (The effective mass of the equipment on raised floors or tables may significantly change the response levels and frequencies. At the equipment’s resonance frequencies, it will act as a “dynamic absorber” and reduce the vibration.)
It is recommended that the measurement be performed, whenever possible, with the sensitive equipment (including auxiliary systems) in both the operating and the non-operating modes in order to distinguish between the various possible vibration sources.
4.2 Instrumentation requirements
Velocity transducers or accelerometers may be used. Because vibration is often at very low amplitudes, the transducers shall be highly sensitive.
Depending on the application, transducer sensitivity as high as 100 mV/(m/s2) (for accelerometers) and
25 mV/(mm/s) (for velocity transducers) may be required. Transducers shall be mounted in accordance with
ISO 5348.
In any case, the instrumentation noise (including electrical disturbances) measured as total r.m.s. value in the frequency range 2 Hz to 200 Hz should be less than 5 % of the maximum measured vibration amplitude. The use of battery-powered instrumentation is recommended to prevent power line disturbances.
The frequency range of the measurement system shall be 1 Hz to 315 Hz (—3 dB each), with a roll-off at both ends of 12 dB per octave. This may be accomplished by using a second-order high-pass filter of the Butterworth type with a cut-off frequency of 1 Hz in the low end and a second-order low-pass filter of the Butterworth type with a cutoff frequency of 315 Hz in the high end. For a digital system, the recommended sampling frequency is at least 2 000 Hz. Additional anti-alias filters with a proper cut-off frequency should be used.
4.3 Analysis
The time history shall be analysed using a set of filters corresponding to the response-equivalent peak velocity. The rationale for the method and description of the filters are given in annex A. Each filter corresponds to the Q normalized pseudo velocity response of a single-degree-of-freedom system with a defined resonance frequency and Q value. The method gives the amplitude of a sine wave having the same maximum relative displacement response as the studied vibration. The analysis result is presented as peak velocity versus resonance frequency.
The (resonance) frequencies of the filters shall be 40 per decade, logarithmically distributed. If possible the appropriate value of Q should be obtained from the instrumentation manufacturer. In the absence of such information, it is recommended that three Q values be used in the analysis: 5, 10 and 20. If only one Q value is used, the recommended value is 10.
The algorithms for the digital computation of the filtered time series are given in annex B.
CAUTION: When the time history is fed to the filters, a transient response will occur at the start. It is recommended that the calculation of maximum response be delayed until the initial transient has died away.
4.4.2 One-third-octave band spectrum
If the spectrum is given in r.m.s. values (averaged, not peak hold), the estimation of the maximum value from the r.m.s. velocity given in 4.4.1 may be applied for each frequency in the spectrum. If the vibration is known to be periodical, the factor of 1,4 may be used. The result should be presented as the vibration velocity.
4.5 Statistical considerations
Each vibration event (e.g. from a passing train, from a blasting, from a series of piling blows or a recording of vibration from rotating machinery) will result in one response-equivalent peak velocity spectrum according to 4.3. The number of events that should be analysed must be subject to sound engineering judgement. It is recommended that a complete cycle of events be treated (for example, a full day and night cycle). The final result is then calculated by taking the maximum value for each frequency over the calculated spectra.