• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Time Domain measurements?

haraldo

Active Member
Joined
Jun 23, 2020
Messages
118
Likes
13
Location
West Norway
We use Dr. Brüggemanns UB Filters. They are a further development of the Horbach-Keele. They are steep but the result is optimal time domain behavior nevertheless.

Thx :cool:

I can't however find any wiritings on Dr. Brüggemanns UB Filters, do you have any pointers, links or info? :)
 

UliBru

Active Member
Technical Expert
Joined
Jul 10, 2019
Messages
123
Likes
337
Thx :cool:

I can't however find any wiritings on Dr. Brüggemanns UB Filters, do you have any pointers, links or info? :)
The UB filters are based on the idea that kinks or non-smooth transitions in the frequency response cause some ringing in the time domain. Thus the target is to get smooth frequency responses. An indication is the behavior of derivatives, like kink (1st derivative) or jerk (2nd derivative). A curve without jerk requires at least a polynomial of 5th order. The UB filters are now given up to 11th polynomial order = smoothness up to 5th derivation.

An example of first order UB jPol11 with corner frequencies 300 Hz and 2 kHz:
Amplitude.png


As you can see the filters are quite steep. They add perfectly.
The linearphase step response:
Time.png


I hope this answers your question.
 

haraldo

Active Member
Joined
Jun 23, 2020
Messages
118
Likes
13
Location
West Norway
The UB filters are based on the idea that kinks or non-smooth transitions in the frequency response cause some ringing in the time domain. Thus the target is to get smooth frequency responses. An indication is the behavior of derivatives, like kink (1st derivative) or jerk (2nd derivative). A curve without jerk requires at least a polynomial of 5th order. The UB filters are now given up to 11th polynomial order = smoothness up to 5th derivation.

An example of first order UB jPol11 with corner frequencies 300 Hz and 2 kHz:
View attachment 72404

As you can see the filters are quite steep. They add perfectly.
The linearphase step response:
View attachment 72405

I hope this answers your question.

Thx a lot for your quick response :)

I don't understand, however, the negative part of the step response... do you have something that show the summed response of all the drivers?
 

haraldo

Active Member
Joined
Jun 23, 2020
Messages
118
Likes
13
Location
West Norway
The "pre-ringing" is somewhat confusing to me, to me it looks like the woofer and midrange is ringing before the signal should start, but in opposite phase, so I guess ideally they would cancel out .... but I can't get my head around that :rolleyes:

Woofer and midrange is not in the same physical location so I wonder if you will never get a 100% full cancellation, isn't there first reflections that do NOT cancel out (especially in vertical plane), so isn't this audible? :rolleyes:
 
Last edited:

tuga

Major Contributor
Joined
Feb 5, 2020
Messages
3,984
Likes
4,281
Location
Oxford, England
The "pre-ringing" is somewhat confusing to me, to me it looks like the woofer and midrange is ringing before the signal should start, but in opposite phase, so I guess ideally they would cancel out .... but I can't get my head around that :rolleyes:

Woofer and midrange is not in the same physical location so I wonder if you will never get a 100% full cancellation, isn't there first reflections that do NOT cancel out, so isn't this audible? :rolleyes:

Are they reproducing the same frequency bands?
 
Last edited:

haraldo

Active Member
Joined
Jun 23, 2020
Messages
118
Likes
13
Location
West Norway
Are they reproducing the same frequency bands?

I think so, the wavelength is about 0.0032 seconds, which seems like around 300+ Hz which is exact crossover point between woofer and mid
 
Last edited:

KSTR

Major Contributor
Joined
Sep 6, 2018
Messages
2,690
Likes
6,013
Location
Berlin, Germany
For a linear-phase crossover, the pre-ringing is extremely benign and is most certainly not audible even off-axis where the perfect summing/cancellation falls apart. Note that a typical minimum phase XO's like LR4 have actually more and longer ringing, just that it is post-ringing and that may be more audible in the end as this XO.
In general, symmetrical or coaxial arrangements are somewhat preferable with linear-phase XO's to mitigate the degradation of perfect summing.
 

Joachim Gerhard

Active Member
Industry Insider
Joined
Apr 29, 2020
Messages
169
Likes
131
We like the UB filters the most from a standpoint of soundquality. This is our subjective impression, sure. The result is unique though in terms of micro details and sheer musicality.
 

Joachim Gerhard

Active Member
Industry Insider
Joined
Apr 29, 2020
Messages
169
Likes
131
In my speakers I use quite low crossover frequencies. That also helps the off axis response.
 

haraldo

Active Member
Joined
Jun 23, 2020
Messages
118
Likes
13
Location
West Norway
We like the UB filters the most from a standpoint of soundquality. This is our subjective impression, sure. The result is unique though in terms of micro details and sheer musicality.

isn’t there quite many things you can’t measure?

how do you measure the magic of a speaker?

how do you measure a speaker’s capability to bring tears in the eyes while listening to music?
 

Joachim Gerhard

Active Member
Industry Insider
Joined
Apr 29, 2020
Messages
169
Likes
131
I have 50 years of professional experience. I was on at least 300 shows. I was at many famous reviewers homes. I won many test reports and got title pages and interviews world wide. My hearing is intact at 62. In the end i listen. When it sounds bad it is bad, no matter the specs. I work mostly in the time domain. Linear frequency response alone is easy. Every student can do that with modern simulation tools. Measuring off axis and distortion is a good idea too.
 

haraldo

Active Member
Joined
Jun 23, 2020
Messages
118
Likes
13
Location
West Norway
I have 50 years of professional experience. I was on at least 300 shows. I was at many famous reviewers homes. I won many test reports and got title pages and interviews world wide. My hearing is intact at 62. In the end i listen. When it sounds bad it is bad, no matter the specs. I work mostly in the time domain. Linear frequency response alone is easy. Every student can do that with modern simulation tools. Measuring off axis and distortion is a good idea too.

It's what I mean .... I experienced that a good measuring speaker is only good measuring, it doesn't have to be good sounding at all
You certainly convinced me.... I got to listen to ARA before I get my next speakers :cool:
There are several people in the industry that I talked to that say that every design that come from your hands is amazing, so I need to check ....

Sorry if I confused you... when I wrote you in the post above, I did not mean you, but a speaker designer in general :)

Just a sidenote ..... I thought you had your hands on Virgo III, is that incorrect?
 
Last edited:

Joachim Gerhard

Active Member
Industry Insider
Joined
Apr 29, 2020
Messages
169
Likes
131
didin’t you always rely on really low crossover points, especially for tweeters?
Modern tweeters can be crossed very low because they have low distortion down there, much lower then 20 years ago. That is limited by linear mechanics though. In the end is is surface times throw. The UB filters are very steep. That helps too.
 

Joachim Gerhard

Active Member
Industry Insider
Joined
Apr 29, 2020
Messages
169
Likes
131
It's what I mean .... I experienced that a good measuring speaker is only good measuring, it doesn't have to be good sounding at all
You certainly convinced me.... I got to listen to ARA before I get my next speakers :cool:

Sorry if I confused you... when I wrote you in the post above, I did not mean you, but a speaker designer in general :)

Just a sidenote ..... I thought you had your hands on Virgo III, is that incorrect?
I think the Virgo 2 was kind of better. Yes, I was the co designer of the Virgo 3 too but curiouslx it was not as successful in terms of sales.
 
Top Bottom