Signal Conditioning

cancel
Showing results for 
Search instead for 
Did you mean: 

Bessel filter in SCXI-1531 module

Hi,

I have a question regarding National Instruments SCXI-1531.
The specifications of the product state that the Low Pass Filter is a Programmable 4-pole Bessel filter. Most of the signal conditioners for IEPE accelerometers like Iotech WBK4 or Wilcoxon PR710 or Endevco 136 or Wilcoxon IT111 adopt instead a Butterworth filter with filter order >=4.

In terms of noise, does the choice of a Bessel instead of Butterworth filter represent a disadvantage for SCXI-1531?

A Bessel filter has a very gradual initial falloff, while a Butterworth has a faster initial falloff that would provide more attenuation near the cut frequency.

Am I missing some important considerations in this comparison of SCXI-1531 with these other signal conditioners?

Thank you in advance for your help.
0 Kudos
Message 1 of 4
(3,330 Views)
Hello GFrau,

I believe that the main advantage of a Bessel filter over a Butterworth filter is that a Bessel filter has characteristics that are optimized for phase linearity. In terms of noise, it would depend on what frequency of noise you are expecting to filter out. If the noise you are trying to filter out is not close to the cutoff frequency, then I don't think that the Bsessle filter represents any disadvantage at all.

I hope this information is useful!

Take care,
E.Lee
Eric
DE For Life!
0 Kudos
Message 2 of 4
(3,317 Views)
Hi E. Lee,

Thank you for your reply.
The phase linearity you mentioned implies that the input waveform is phase shifted an amount proportional to frequency. In other words, the output is again the input signal but with higher frequency components removed by the filter. But does the Bessel filter choice in the SCXI-1531 have to do with possible overshoots?
I wonder whether an accelerometer signal conditioner with higher order Butterworth filter would offer more attenuation but higher risks of overshoot and ringing and then a worst output waveform.
I guess the selection of the signal conditioner is not easy.

Thank you for your help.

Best regards,
Giuseppe
0 Kudos
Message 3 of 4
(3,310 Views)
Hello Giuseppe,

To be honest with you, I'm not sure if the designers of the SCXI-1351 were concerned with possible overshoots in the signal. I'm also a little rusty on my filter design, but I believe that a higher order Butterworth filter offers a steeper rate of attenuation at the cost of ringing and phase linearity just as you mentioned. Like most designs, the choice of any filter offers both advantages and disadvantages over other filters. It is really up to you in terms of what you believe your signal will benefit most from (steeper roll-off, less overshoot, phase linearity, etc.) in order to determine which signal conditioner is best for your application. And as you mentioned, it's not always an easy decision.

Take care,
E.Lee
Eric
DE For Life!
0 Kudos
Message 4 of 4
(3,303 Views)