Welcome, Guest
Username: Password: Remember me
All tutorials related to equalization of a system with miniDSP platforms.

TOPIC: Some bi-quad help please

Some bi-quad help please 8 months 1 week ago #29119

  • denver8me4dinner
  • denver8me4dinner's Avatar
  • Offline
  • Expert Boarder
  • 3038564518
  • Posts: 132
  • Thank you received: 17
  • Karma: 3
I have a set of 2x4HDs
I use the APF function and it's fantastic for in car.

So I can measure acoustic response and get it pretty good however is a 2nd order APF

My question is

How do I use biquad to make a phase shift by 11.25 deg and 22.5 deg?

Would that be a first order APF ?
I tryed using spreadsheet and it won't work right?

Can anyone point me to a calc that doesn't use APF order but rather the degrees in phase (even tho one in same) just easier for me to figure out.

Or explain exactly what a APF at what order and what Q would be a 11.25° and 22.5° Phase shift (at whatever corner frequency....I'll probably use the crossover point which could be anything. But that's important too...


When I look at spreadsheet the Q determines how fast the phase changes but it never stops at a specific degree it keeps going to -180 any help much appreciated

Cheers
The administrator has disabled public write access.

Some bi-quad help please 8 months 1 week ago #29124

  • dreite
  • dreite's Avatar
  • Offline
  • Platinum Boarder
  • Posts: 1027
  • Thank you received: 504
  • Karma: 81
A first-order all-pass has a phase shift from 0 to 90 degrees. You could use the supplied miniDSP spreadsheet to create a first-order all-pass by inputting different formulas to process the Frequency and Fs values. In that case, the b2 and a2 coefficients will be zero.

However, I'm puzzled by your objective. First-order and second-order all-pass functions are for duplicating the phase-shifts of appropriate crossovers. What characteristic/application could you have that would need a 11.25 or 22.5 degree correction??

Dave.
The administrator has disabled public write access.
The following user(s) said Thank You: devteam

Some bi-quad help please 8 months 1 week ago #29125

  • denver8me4dinner
  • denver8me4dinner's Avatar
  • Offline
  • Expert Boarder
  • 3038564518
  • Posts: 132
  • Thank you received: 17
  • Karma: 3
dreite wrote:
A first-order all-pass has a phase shift from 0 to 90 degrees. You could use the supplied miniDSP spreadsheet to create a first-order all-pass by inputting different formulas to process the Frequency and Fs values. In that case, the b2 and a2 coefficients will be zero.

However, I'm puzzled by your objective. First-order and second-order all-pass functions are for duplicating the phase-shifts of appropriate crossovers. What characteristic/application could you have that would need a 11.25 or 22.5 degree correction??

Dave.

Thanks for the reply and help Dave!

So , I use my dsps in a car as stated ,
There's another dsp on the market (helix dsp pro)
For a car that has the ability to shift phase in 11.25° Increments.

On another forum some folks say that feature is a APF but the corner frequency is fixed at crossover. It has a slider in the dsp that moves ALL the phase of that channel in even amounts.

I've been using APFs trying to get the phase shift desired. I just recently got enough experience to gate my responces good enough for a car environment because early reflections are louder than source. So until now I haven't really been able to see the effects of my APF. Well it's shifting too much in some areas and I'm realizing a even shift across an entire channel (midrange in 4way) on one side need to be moved.

I have the ability to do it in FIR or biquad just don't know how. I would like to keep it in IIR to free up my FIR Coefficients. However I've been successful using a even spaced shelf filters for phase in rephase, but than The 1024t bank doesn't have enough power to do the crossover I'm using it on. So I need to figure out a way to do it in biquad.

I need it for car. Offset seating I have one midrange that moves between 11-22deg OFF fromdriver side midrange through the passband.
I tryed using a low Q (.5) APF and it still isn't shallow enough it moves too much phase.

I haven't been able to get 1st order to work using spreadsheet so I've been only able to use 2nd order (-180 reversal) :(

If you know a way I'm all ears ;)

Much appreciated
Andrew
Last Edit: 8 months 1 week ago by denver8me4dinner.
The administrator has disabled public write access.

Some bi-quad help please 8 months 1 week ago #29155

  • dreite
  • dreite's Avatar
  • Offline
  • Platinum Boarder
  • Posts: 1027
  • Thank you received: 504
  • Karma: 81
You can create a first-order all-pass filter with this setup:

b0 = (g-1)/(g+1),
b1 = 1,
b2 = 0,
a1 = b0,
a2 = 0,

Where g = tan(pi*(Fc/Fs))

b1,b2,a2 are always 1,0,0, respectively.

Dave.
The administrator has disabled public write access.
The following user(s) said Thank You: devteam

Some bi-quad help please 8 months 1 week ago #29165

  • denver8me4dinner
  • denver8me4dinner's Avatar
  • Offline
  • Expert Boarder
  • 3038564518
  • Posts: 132
  • Thank you received: 17
  • Karma: 3
dreite wrote:
You can create a first-order all-pass filter with this setup:

b0 = (g-1)/(g+1),
b1 = 1,
b2 = 0,
a1 = b0,
a2 = 0,

Where g = tan(pi*(Fc/Fs))



b1,b2,a2 are always 1,0,0, respectively.

Dave.

Dave,
Thank you!
I had to re-watch some biquad building videos, I think I can get it now. Sorry I'm such a rookie .
But this I'm pretty sure I can get to work. It might take a few try's:)

Andrew
The administrator has disabled public write access.
Time to create page: 0.125 seconds