LPF resonance control error | Algorithm Design | Forum


Please consider registering

sp_LogInOut Log In sp_Registration Register

Register | Lost password?
Advanced Search

— Forum Scope —

— Match —

— Forum Options —

Minimum search word length is 3 characters - maximum search word length is 84 characters

sp_Feed Topic RSS sp_TopicIcon
LPF resonance control error
August 12, 2017 - 12:42 pm
Member Since: August 7, 2017
Forum Posts: 5
sp_UserOfflineSmall Offline

Not sure if this is the right section to post this on... this might be already posted somewhere else, but anyway:

While working on the 2nd order LPF project (as well as the Massberg filter design) I noticed that an error occurs when the resonance (Q) becomes lower than 0.707. This might be some instability. In my DAW this also causes a shutdown. So I set the minimum value for Q to be 0.707 (like the initial value). Seems to be better.

EDIT: When switching between the Massberg option and the "normal" option while Q is set to 0.707, the gain blows up (the audio breaks down in my DAW). Could this have something to do with the fact that Q is in the denominator in the equation

float d = 1.0 / fQ;

and maybe some kind of rounding happens where 0.7 becomes 0 (even though actually every variable is set as float, not int).


Edit 2: I also doubt that I might have done the implementation of the Massberg switch wrong because everytime I change the frequency the Massberg switch goes back to default.

if (m_f_MassbergSwitch2 == SWITCH_ON) // first it was SWITCH OFF which would make more sense but the spectrum showed that using this option leads to the non-Massberg design

{ not Massberg (this is the implementation from the book, including the coefficient calculation) }

else { Massberg stuff }

Some feedback would be very helpful! For further information about what I did, I uploaded my project folder here:


Thank you so much.

New Member
August 30, 2017 - 4:47 am
Member Since: August 26, 2017
Forum Posts: 2
sp_UserOfflineSmall Offline

As a filter pole is moved to provide Q, it may at some point exceed the limits of digital stability. For example, some filter designs move 2 poles apart to provide Q, the rising one will eventually exceed a gain of one of its elements in the feedback of 1.0. At this point an IIR filter becomes unstable. It is better to do some limit checking to prevent effectively CR going critically low, and with the sin or tan Nyquist compensation, making some gain total too big. Run at an over sampling. I think a state variable had this problem above 8kHz for a 44.1kHz sample rate, and needed triple oversampling at least.

Can't help you on the messy massey effect.

Forum Timezone: America/New_York

Most Users Ever Online: 152

Currently Online:
3 Guest(s)

Currently Browsing this Page:
1 Guest(s)

Top Posters:

Chaes: 56

Skyler: 48

StevieD: 46

Derek: 46

Frodson: 45

Peter: 43

TheSmile: 43

Nickolai: 43

clau_ste: 39

jeanlecode: 37

Member Stats:

Guest Posters: 2

Members: 784

Moderators: 1

Admins: 6

Forum Stats:

Groups: 13

Forums: 42

Topics: 850

Posts: 3372

Moderators: W Pirkle: 693