Group Delay/phase responce
Printed From: Speakerplans.com
Category: General
Forum Name: Advanced Discussion
Forum Description: Advanced discussion area for higher lifeforms
URL: https://forum.speakerplans.com/forum_posts.asp?TID=5824
Printed Date: 16 April 2024 at 2:37pm
Software Version: Web Wiz Forums 12.06 - https://www.webwizforums.com
Topic: Group Delay/phase responce
Posted By: kevinmcdonough
Subject: Group Delay/phase responce
Date Posted: 22 July 2006 at 5:58am
|
Hey all I kind of have a vague idea of what group delay and phase responce is, but wanna learn the detail about it. In terms of cabinate design, what exactly causes it? What design features can you implement to improve it and what makes it worse. Any info or links appreciated. kev |
Replies:
Posted By: kevinmcdonough
Date Posted: 25 July 2006 at 6:05pm
|
LOL anyone?  |
Posted By: soundsystemdan
Date Posted: 25 July 2006 at 6:24pm
|
well what type of cabinet ae you refering to? afaik these things affect reflex and bandpasss designs. I probably know little more than you about what causes it, i dont think phase response is much of an issue, except at crossover frequencies or if you're combining non identical cabinets. As far as making group delay better or worse, it would depend what you want out of your cab, be it depth, kick or efficiency, or maybe size is important too. volume and tuning frequencies can affect power handling too. I expect you know some of this all ready. I'm in the process of trial-and-error mdf prototyping different reflex designs. Hopefully by the end of the week I'll be able to post some findings. ------------- http://dancomsound.com" rel="nofollow - http://dancomsound.com |
Posted By: djtosco
Date Posted: 26 July 2006 at 1:22pm
|
Any linear system could be represented by a transfer function, which contains all the informations about its behaviour. Even a loudspeaker+enclosure, under a certain degree of approximation, is a linear system with a transfer function (TF). A linear system with a sine wave at the input, gives you a sine wave at the output with the same frequency, but eventually different amplitude and phase. The frequency response is the plot which represents the ratio between the amplitude of the output ant the input (usually in dB) at the different frequencies. The phase response plot represents the difference in phase between the input and the output at the different freqeuncies. A phase difference is a time difference, and thus the group delay. A non-linear phase response cause the fact that the system produces a frequency-dependent delay. Any waveform (Fourier's theory) could be represented by a sum of sine waves. But because of the group delay, the waves at the different frequency are differently delayed. This cause a different shape in the input and output waveforms. So, we want an almost constant group delay (the value is not so important, and could be easily compensated with a DSP), thus an almost linear phase response (the group delay is the opposite of the derivative of the phase respect to the frequency in rad/s) |
Posted By: kevinmcdonough
Date Posted: 26 July 2006 at 5:37pm
|
LOL thanks for your replys guys. I already know most of the info, i understand WHAT phase is, what fourier transforms are and the graphs etc. I suppose my question wasn't phrased very well but i kind of ment more WHY Why does a speaker play different frequencies at different delays/phases. and what building meathods are to be avoided or used in order to minimise this effect. I'm refering to mostly to reflex as thats what i'm working on at the moment but just all designs in general really. kev |
Posted By: godathunder
Date Posted: 28 July 2006 at 6:51am
|
Im not sure that Im remotely correct here but my understanding is that the cabinet produces differing resistance to cone movement at differing frequencies. This resistance to cone movement alters the reactive component and hence impedence of the coil. since the phase angle is the ratio of the coil resistance (constant) to impedence if the impedence changes then the phase changes. Corrections anyone? ------------- LOUDER THAN LOUD |
Posted By: tb_mike
Date Posted: 09 August 2006 at 8:38pm
|
http://www.xlrtechs.com/dbkeele.com/PDF/Keele%20%281981-05%20AES%20Preprint%29%20-%20Monitor%20Loudspeaker%20Systems.pdf -
See blauert and laws criteria as in the below pdf
http://www.xlrtechs.com/dbkeele.com/PDF/Keele%20(1981-05%20AES%20Preprint)%20-%20Monitor%20Loudspeaker%20Systems.pdf
Go to audioasylum,search authoer 'tom danley' or 'tomservo' youl get heaps on different topics there.
|
Posted By: unknown_artist
Date Posted: 15 August 2006 at 1:16pm
|
Hm, the best case is that you design your system as a minimum phase system(it's not possible to have "faster" system without the use of FIR filters[which quit have delay to work even at low frequencys]). It's a disputed question whether you can hear group delay. Definitely you can hear it at low frequencies and you can hear it when theres a jump in the group Delay at your crossover (for example at 1500Hz)-> then transients will be torned up(don't know if it's the right word)- for example a snare or somthing like this... So you have to try to get a flat group delay(group delay means the derivative of phase -> a spectrally delay) ------------- I'm sorry for all faults, my knowledge of the english language is not the best |
Posted By: tb_mike
Date Posted: 16 August 2006 at 9:51pm
|
I find it easy to visualise with step response/impulse response measurements. The acoustic response is easily understood here.
Although this may be a simplification...I havent bothered with FIR study yet.
Once people realise what it is,then you have to see if your ear can even hear what looks bad on a graph. It certainly is sensitive to certain things in certain frequency ranges but not others.
|
Posted By: unknown_artist
Date Posted: 17 August 2006 at 2:49pm
If you're able to understand the german language, there are great publications available(including the question:what kind of improvements are "hear"able....) ------------- I'm sorry for all faults, my knowledge of the english language is not the best |
tb_mike wrote:Posted By: Paulo Duto
Date Posted: 17 August 2006 at 6:17pm
|
See also John L Murphy's collection of posts to Bass list mentioning Journal of the Acoustic Society of America Vol 63, nr 5, pp 1478-1483 (May 1978) Blauert and Laws... http://www.trueaudio.com/post_010.htm - http://www.trueaudio.com/post_010.htm
and
http://www.trueaudio.com/post_011.htm - http://www.trueaudio.com/post_011.htm
Regards,
Paulo.
------------- Paulo Duto |
Posted By: tb_mike
Date Posted: 17 August 2006 at 10:11pm
Sorry,my german only consists of 2yrs at secondary school,not quite enough I think!
|
Posted By: tb_mike
Date Posted: 17 August 2006 at 10:12pm
Hey Paulo those are good links!
|