Acceptable group delay

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
gedlee said:


If you are saying that edge diffraction can occur with GD then you would need to show this mathematically not "the way I see it". Its either true and can be proven or its not.


No, you was claiming this not me. I don't see GD in a reflection which is a pure time delay. GD is where a group of frequencies (not all frequencies) are delayed by some amount.


Also I do not need to proove anything mathematically. This is a discussion board and I am free to express my view and my understanding of a subject.

Also your last line doesn't fit well with this that you wrote further up:
I think that edge diffraction is both GD and comb filtering.


How can you "think" either it is or it is not... right? (and you better proove it..)


/Peter
 
I have done an analysis of the simplest case- two sources, one delayed and attenuated. This system has nonzero and nonconstant group delay. Since I can't attach a word document, I've attached a picture of the word document that I have the details in. I will post some graphs shortly.
 

Attachments

  • analysis.jpg
    analysis.jpg
    66.5 KB · Views: 231
Here goes... not sure if the resolution will be very good though.

A = .5, tau = 1 ms

An externally hosted image should be here but it was not working when we last tested it.


A = .25, tau = 1 ms

An externally hosted image should be here but it was not working when we last tested it.


A = .5, tau = 2 ms

An externally hosted image should be here but it was not working when we last tested it.


Also, if anyone wants the matlab function that I'm using, go ahead and email me at rybaudio@hotmail.com and I'll send it to you.
 
In case you can't read what is going on, the plots are as follows.

Upper left:

system transfer function magnitude in dB vs frequency
vertical scale: linear, -30 dB to 10 dB
horizontal scale: log, 20 Hz to 20 kHz

Upper right:

system phase angle in degrees vs frequency
vertical scale: linear, -90 to 90 degrees
horizontal scale: log, 20 Hz to 20 kHz

Lower left:

system group delay in ms vs. frequency
vertical scale: linear, -2 to 2 ms
horizontal scale: log, 20 Hz to 20 kHz
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.