Lowpass filter design please help me

Go To Last Post
10 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hi folks

I have a particular problem.
I want to design a lowpass filter which would have passband freguency = 80 Hz and passband frequency 100 Hz. Stopband attenuation should be 30 dB or something like that.

Obviously I need this filter for microcontroller :)

I have designed sucha filter and it consist of 4 stages. N=8 Overall gain is 60 [v/v]
It is Chebshev filter.
Unfortunately on the second stage there are high oscillations!
I don`t know why they are occuring but even if I don`t provide signal to the input they are existing.
I really need your help cause I am not very proficient in designing filters. I used special program for designing filters.

Below I am enclosing filter characteristics and circuit designs:

Stage1:

Stage2:

Stage3:

Stage4:

Why oscillations occur on the second stage ??

I join each stage in series.
Is this design is wrongly designed ??
What should I take into consideration ??
I supply each op-amp MC4558 +-5 [V]

Please help me any clues and maybe other designs are welcome :)

Thanks in advance

Adam

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I have never seen a filter stage with a feedback
from the output to the positive opamp-input (positive feedback).

Ususal filter topologies as shown in:

http://www.science.unitn.it/~bas...

ground the positive input.

Often "positive feedback" leads to instability or
oscillation. So could you give us hint where this
topology is described. Perhaps there is a hint
under which constraints a stage really is stable.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I used special program for designing circuit.
Filter Multiple Feedback (MFB) Second Order Lowpass II inverting

That`s how it is called and it vary from the first one that it has these 2 resistors.In this program is written that tunning this circuit is fair. So maybe it is not so easy to get rid of these oscillations and you have to adjust these two resistors.
Program describes other circuits as good for tunning.

Only hint I have found about Filter Multiple Feedback (MFB) Second Order Lowpass II inverting
was:

http://www.technick.net/public/code/cp_dpage.php?aiocp_dp=cir_filter_lp2i_mfb_02

What do you think about this design ??

Maybe it is better to change for

Filter Multiple Feedback (MFB) Second Order Lowpass I inverting

or maybe for :
Sallen_Key second order lowpass

Now my suspision is that my choice was cause I wanted to obtain the maximum gain. I chose this design and I obtained desired gain but I loose stability.
But that`s only my suspition.

What do you think aout this all ??

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Remove R5 from each stage of the filter. R5 doesn't add anything to the filter, except Hysteresis. I would predict that R5 is why the second stage of your filter is oscillating. The positive feedback from the output of the filter thru R5 to the non-inverting input of the Op-amp is setting up the filter stage for oscillation.

Worst case, you may need to disconnect R5 from the Op-amp output and connect that end of R5 to VCC or V+.

You can avoid reality, for a while.  But you can't avoid the consequences of reality! - C.W. Livingston

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

From what source do you have the filter
design/simulation software ?

Perhaps you could find a design software
that uses the more traditional
multiple-feedback or sallen-key topology
and redesign the filter.

Do you really need an 8.th order filter ?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

There is one nice program called "filter wiz pro"
You can find many circuits for filters.
I forgot about stability.
Now I think that I redesign my circuit.
As far as I am concern I should also avoid positive feedback. It always does some oscillations if you don`t tune it properly ;)
Well I need 8.th order filter cause my algorithm in AVR microcontroller needs really nice selective waves.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Things are not as easy as it seems.

There are circuits with positive
feedback that are stable.
(A negative-Impedance-Converter NIC for example)

There are circuits without positive feedback
that are unstable.

Therefore I am not completely sure what causes
the oscillation in your circuit but I still consider
it a good route to avoid positive feedback.

If designing an 8.th order filter the
tolerances of the components may become important !

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I was using elements which have tolerance +-5%
But sometimes they have varied a lot more.
Maybe this caused problem.
But you know how to is with electronics stores.
They don`t care about tolerance they are just happy that they have resistors or capacitors of desired value. For example I used 2 capacitors for 200V !!!
because my store didn`t have for lover voltage. So I suppose tollerance was far from desired ;)

Uff many things to consider before soldering filter ;)

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

All the other filters on the site EXCEPT that one show the + input grounded. Try the Sallen & Key 2nd order for example... (Cool collection of filters... hadnt seen some listed!)

Imagecraft compiler user

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I recommend the filter design software from Texas Instruments. It's free of charge and did already a lot of good jobs for me:

http://focus.ti.com/docs/toolsw/...