How do I improve the filtering quality of the fractional octave analyzer without increasing the reporting time of data

[J-Roc]

I want to aquire a 1/3 octave analysis from 3 analog input accelerometers. These are put into waveforms and are then put through labviews frational octave filter. The results from the filter are giving an incorrect distribution of the freqeuncies. I known this beacuase the signal being input is of known frequency (sine wave 1 Hz.) produced by a signal generator and also checked with another device. Labviews results shown wide distribution when in reality the majority of the distribution should occur at 1 after the filtering.

[Gerald Albertini]

If you're using the sound and Vibration Toolset 2.0, you should be able to visualize the "filter settled" indicator.
The settling time of the filter is inversely proportional to the lowest frequency you define in the "frequency range [Hz]" settings of the 1/n filter and it may require a certain amount of time before the filter returns valid measurements.

Could you please provide more info about your settings ?
Are you using IEC or ANSI VI's ?
What are your frequency range settings ?
What Sampling frequency are you using ?
I guess you're using an exponential-slow averaging.

I attach a VI using Sound and Vibration Toolset 2.0 that simulates your measurements and shows the effect of changing the frequency range settings on the settling time p
arameter.

I hope this will help.
Regards

Gerald