Sound & Vibration, Order Analysis: parameters and strange results

[peper2001]

Hi,

I'm using the Sound and Vibration toolkit to do order analysis. My VI takes temporal data as input (analyzed channels + tachometer) and generates a graph of amplitude vs tachometer frequency.

I basically use the two express vi depicted hereunder:

My problem is that there are very few informations about the effect of the parameters of the method and their relationship.

For instance:

- what is exactly the "bandwidth [order]" parameter? It seems to determine the number of revolution taken into account for the analysis. Am I correct?

-  does the size of the block of data that is inputed to the VI have an influence on the results? If yes what is the influence?

- It seems that the time the order analysis VI takes is longer for each new iteration. Isn't there some kind of memory leak somewhere or   memory not correclty prealocated?

- Does the frequency at which the tachometer is computed have an influence on the resulting order tracking graph?

- I get very "edgy" results like if the computation was made in steps (see picture). How is it possible?Is there a way to correct that effect?

Thanks in advance

peper

[ParisL]

Hi Peper,

Bandwidth [order]—Specifies an order span centered at the order that you specify in Orders to track. Specify a Bandwidth [order] value between 0.01 and 1. If Bandwidth [order] is less than 0.01, the Bandwidth [order] sets to 0.01. If Bandwidth [order] is greater than 1, the Bandwidth [order] sets to 1. The default is 1. It can be realted to revolution, time or speed(rpm), and it is set when you configure the order tracking VI.

The size of the block data could delay the results, bringing also memory leakages, but I cannot say that for sure without seeing your code.