• 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?

DonH56

Master Contributor
Technical Expert
Forum Donor
Joined
Mar 15, 2016
Messages
7,835
Likes
16,497
Location
Monument, CO
And you think with a simple EQ to equal curves you will hear a difference despite the maybe differences in dispersion??? No way....

I am not sure where this is directed, since I have not really been following this thread, but note modern DSP systems for room correction and such do much more than "simple EQ"; they can correct phase as well as amplitude, and can fix problems in the time-domain.
 

haraldo

Active Member
Joined
Jun 23, 2020
Messages
118
Likes
13
Location
West Norway
Here is an AMT against a ribbon. You see that the ribbon is more extended over 20kHz and has more sensitivity.
One reason is the very low moving mass. That is more resolution to me.

How high up do we need?
Like the Townshend supertweeter that extends to 100Khz ... how much do we need?

Guys at OHC claim supertweeters are amongst the cheapest speaker upgrade you do... and that it works even with hi-end statement speakers...

I never played around with supertweeters but would like to try out to see what happens? :p
 

Lbstyling

Addicted to Fun and Learning
Forum Donor
Joined
Jan 2, 2019
Messages
500
Likes
461
Everyone is justly raving about the performance of the Kii3 and the D&D 8c.

The Toole school preaches that everything which matters is expressed in the frequency response measurements.

Yet there is a chance that the time-domain performance of these speakers may contribute to the overall results.

719DD8Cfig5.jpg

719DD8Cfig6.jpg

https://www.stereophile.com/content/dutch-dutch-8c-active-loudspeaker-system-measurements

917Kii3fig1.jpg

917Kii3fig7.jpg

https://www.stereophile.com/content/kii-audio-three-loudspeaker-measurements

The thing is, unless it's a coax design, this means nothing unless you sit with your head at EXACTLY the right height, as the distances between drivers and your ears changes as you move above and below on axis.

Even then, it would need to be listened to in an anechoic room as the ceiling and floor reflections would arrive at different times for the bass and tweeter, (tweeter arrives earlier from ceiling reflection, and bass driver arrives earlier from the floor reflection again, unless it's a coax driver/point source.

Having said this, I do think the research on time delay importance is useless unless it's at least done using a speaker that meets the performance requirements for an exceptional speaker before you even start to think about evaluating time domain stuff.



For this test, a EQed KEF R3 would be a good start point, but if I was serious about doing this I would also want a radian 951be on a SEOS or OSWG 300hz horn and a AE TD15m, along with an electrostatic speaker to be sure.

If you don't hear any time domain effects on these 3, then you can be reasonably sure there is no way to currently hear it, if you ever even will.

You can't test this on a typical 3 way.
 
Last edited:

Joachim Gerhard

Active Member
Industry Insider
Joined
Apr 29, 2020
Messages
169
Likes
131
How high up do we need?
Like the Townshend supertweeter that extends to 100Khz ... how much do we need?

Guys at OHC claim supertweeters are amongst the cheapest speaker upgrade you do... and that it works even with hi-end statement speakers...

I never played around with supertweeters but would like to try out to see what happens? :p
I think it is not Extension PER SE but a lower mass membrane stops faster then one with higher mass. Bending wave transducers can be an exception.
 

Joachim Gerhard

Active Member
Industry Insider
Joined
Apr 29, 2020
Messages
169
Likes
131
It's interesting but I can't say I am sure I can always relate it to frequency respons.... I auditioned the whole piega coax line up through the master series to the Master Line Source ... and I can't say I can hear the same issues here.... there are other things with Piega though, I am not sure I like the sound of a grand piano, a Steinway Model D sounds not fully real (that is my subjective opinion)
Maybe these are phase related issues, the harmonics may be out of phase ...

Piega Coax 711 also show this rising trend ....
View attachment 74648

Piega writes on their webpage: Furthermore: As our ribbon systems ensure that the direct sound components and the sound reflected from the surfaces of the room always reach the listeners in the correct phase
So much for that claim o_O
I don't like when a "credible" speaker manufacturer tell me rubbish
View attachment 74649
That impulse or what that is looks terrible I am sorry to say.
 

KSTR

Major Contributor
Joined
Sep 6, 2018
Messages
2,690
Likes
6,013
Location
Berlin, Germany
The thing is, unless it's a coax design, this means nothing unless you sit with your head at EXACTLY the right height, as the distances between drivers and your ears changes as you move above and below on axis.

Even then, it would need to be listened to in an anechoic room as the ceiling and floor reflections would arrive at different times for the bass and tweeter, (tweeter arrives earlier from ceiling reflection, and bass driver arrives earlier from the floor reflection again, unless it's a coax driver/point source.

Having said this, I do think the research on time delay importance is useless unless it's at least done using a speaker that meets the performance requirements for an exceptional speaker before you even start to think about evaluating time domain stuff.
[...]
While I fully agree on all points, basically, I would come to a less harsh conclusion: Even with the bog-standard 3-way floorstander in someone's typical living room with all its issues, phase linerarization, that is undoing the crossover's allpass response (excess phase) will give a worthwile overall improvement perceived by many people.
More subtle than with the ideal setup that you have described but it's still there. One level of blur removed. Well, that is unless drowning in a real mess of room modes and a sea of strongest early reflections... blank conrecte cellar type of "listening room".

In that sense, while one is at it anyway, do a full "room correction" at the listening point which usually incorporates phase linearization. Then you might get away with way less acoustic treatment and less than perfect transducers. When that is done, one can still use the allpass response of the XO to re-introduce the phase error (but letting everything untouched for the room correction) and by that one can compare the effect of phase response in isolation, almost as good as on the perfect system.
 

fosti

Member
Joined
Jul 19, 2020
Messages
36
Likes
16
I agree, but for crossover phase equalization (if needed) you need inverse allpass filters.
 

pma

Major Contributor
Joined
Feb 23, 2019
Messages
4,591
Likes
10,727
Location
Prague
Hi all,

I'm glad to find this place because I think measurements are important.

Am I overlooking time domain measurements in the reviews?

Time domain measurements are extremely important if you are an amplifier designer. You need them to verify stability and stability margin, especially of the feedback systems. IMO step response of the amplifiers should be measured even here in ASR, to give a complete view. The reason why it is not done is that it needs different measuring equipment than the AP, if done seriously.

I also insist on step response measurement of the amplifiers at full output voltage swing, not at some 1 - 5V only. Full swing response tells quite a lot about large signal non-linearities and transient behaviour. Again it will not be popular to measure, because many units would fail or switch off itself.
 

fosti

Member
Joined
Jul 19, 2020
Messages
36
Likes
16
Time domain or frequency domain. It is all the same ....people who try to ignore that are foolish.....sorry to say so, but this is physics!
 

UliBru

Active Member
Technical Expert
Joined
Jul 10, 2019
Messages
123
Likes
337
Time domain or frequency domain. It is all the same ....people who try to ignore that are foolish.....sorry to say so, but this is physics!
Yes, but many people do not understand the frequency domain in its full meaning. They just look e.g. at the frequency response plot of a loudspeaker. And this magnitude chart does not contain any time information, it simply shows the steady-state frequency behaviour. This typically leads to misinterpretations of frequency charts as you cannot fully tell from a chart how a speaker sounds.
A simple example for better understanding: take a piece of music and calculate its frequency response by Fourier transform. Now time reverse the music and compute the frequency response. The resulting frequency responses are the same. But the music clearly sounds different by the two music samples.
So frequency domain is the same like time domain if it contains both magnitude AND phase information.
 

edechamps

Addicted to Fun and Learning
Forum Donor
Joined
Nov 21, 2018
Messages
910
Likes
3,620
Location
London, United Kingdom
Yes, but many people do not understand the frequency domain in its full meaning. They just look e.g. at the frequency response plot of a loudspeaker. And this magnitude chart does not contain any time information, it simply shows the steady-state frequency behaviour. This typically leads to misinterpretations of frequency charts as you cannot fully tell from a chart how a speaker sounds.

Could you please provide a reference to research that shows clear audibility of phase response of real loudspeakers in a typical listening room?

A simple example for better understanding: take a piece of music and calculate its frequency response by Fourier transform. Now time reverse the music and compute the frequency response. The resulting frequency responses are the same. But the music clearly sounds different by the two music samples.

I don't think that's a reasonable analogy. A piece of music is multiple seconds long, at least. In comparison, the impulse response of a loudspeaker is typically measured in single-digit milliseconds. That's a difference of at least 3 orders of magnitude. The perceptual mechanisms at play are so completely different that the comparison does not make any kind of sense.
 

pma

Major Contributor
Joined
Feb 23, 2019
Messages
4,591
Likes
10,727
Location
Prague
Time domain or frequency domain. It is all the same ....people who try to ignore that are foolish.....sorry to say so, but this is physics!

It is all the same, if you are a theoretician. Yes the domains are dual and there is a Fourier transform and Hilbert transform, convolution and deconvolution. We all know it. But once you are a real designer engineer, you have a piece of amp electronic at your workbench that is close to stability limit and its behaviour depends on the load, what do you think is a faster test approach - to send a unit step or a square of proper frequency and amplitude to it, or to measure frequency responses? In case you are a real designer and practitioner, you will hardly choose the later. If you are a computer guy who never held a soldering iron, then you would say it makes no difference.
 

SIY

Grand Contributor
Technical Expert
Joined
Apr 6, 2018
Messages
10,383
Likes
24,749
Location
Alfred, NY
It is all the same, if you are a theoretician. Yes the domains are dual and there is a Fourier transform and Hilbert transform, convolution and deconvolution. We all know it. But once you are a real designer engineer, you have a piece of amp electronic at your workbench that is close to stability limit and its behaviour depends on the load, what do you think is a faster test approach - to send a unit step or a square of proper frequency and amplitude to it, or to measure frequency responses? In case you are a real designer and practitioner, you will hardly choose the later. If you are a computer guy who never held a soldering iron, then you would say it makes no difference.

Or you see which is already hooked up and running and use it. Because it really does make no difference.

You do realize that an AP has squarewave generation capability?
 

Bjorn

Major Contributor
Audio Company
Forum Donor
Joined
Feb 22, 2017
Messages
1,286
Likes
2,562
Location
Norway
Could you please provide a reference to research that shows clear audibility of phase response of real loudspeakers in a typical listening room?
Depends what you describe as a typical listening room.

Phase issues become audible in well treated rooms and more and more audiophile to treat their rooms. Something that was discovered decades ago by the way in the development of different design studios. However, in a room with no treatment it will be masked because everything is a blur to some degree. A very directional speaker with some rear wall treatment might be sufficient though to hear phase distortion. Exactly where the limit goes isn't something easily said or researched. It will depend on several factors.
 

pma

Major Contributor
Joined
Feb 23, 2019
Messages
4,591
Likes
10,727
Location
Prague
You do realize that an AP has squarewave generation capability?

I want short test signal rise time, not the AP system rise time. Do you think I am a fool who does not understand duality principle?? Do you think I cannot use GHZ analyzers? AP tells you nothing if you are in a real trouble. See below preamp 250MHz oscillations. I am not interested in childish discussions.

inp+634.PNG


buffer_step_4898_smallsig.PNG


Scren112_d2.gif

Preamp HF oscillations
 

SIY

Grand Contributor
Technical Expert
Joined
Apr 6, 2018
Messages
10,383
Likes
24,749
Location
Alfred, NY
I want short test signal rise time, not the AP system rise time. Do you think I am a fool who does not understand duality principle?? Do you think I cannot use GHZ analyzers? AP tells you nothing if you are in a real trouble. See below preamp 250MHz oscillations. I am not interested in childish discussions.

View attachment 74785

View attachment 74786

View attachment 74787
Preamp HF oscillations
You’ve been spending too much time with Curl.

Parasitic oscillation is different than the sort of stability measurements one does to get feedback and compensation correct. And there’s all sorts of tools that can be used to detect them.

If there’s a need for gigahertz measurement, you’re doing something terribly wrong. These are audio amplifiers.
 

pma

Major Contributor
Joined
Feb 23, 2019
Messages
4,591
Likes
10,727
Location
Prague
You’ve been spending too much time with Curl.

Parasitic oscillation is different than the sort of stability measurements one does to get feedback and compensation correct. And there’s all sorts of tools that can be used to detect them.

If there’s a need for gigahertz measurement, you’re doing something terribly wrong. These are audio amplifiers.


Please once more and again, do not bore me with your oversimplified statements. The 3 plots above had nothing in common. The plot with 250MHz oscillations was the local issue in the output diamond buffer. People who do not make deep analysis just may not know the issue exists. On AP, or soundcard in-audio band measurement, the issue appeared as higher mains voltage spuriae. Without the GHz analyzer, the real issue would never be found. It is too easy to say If there’s a need for gigahertz measurement, you’re doing something terribly wrong. These are audio amplifiers, only someone incompetent can say that.

oscillations.png

The only sign of 250MHz oscillations were the elevated mains spuriae and some light audible hum. Who cares? And who finds the real reason? Audioband measurement?
 

SIY

Grand Contributor
Technical Expert
Joined
Apr 6, 2018
Messages
10,383
Likes
24,749
Location
Alfred, NY
Please once more and again, do not bore me with your oversimplified statements. The 3 plots above had nothing in common. The plot with 250MHz oscillations was the local issue in the output diamond buffer. People who do not make deep analysis just may not know the issue exists. On AP, or soundcard in-audio band measurement, the issue appeared as higher mains voltage spuriae. Without the GHz analyzer, the real issue would never be found. It is too easy to say If there’s a need for gigahertz measurement, you’re doing something terribly wrong. These are audio amplifiers, only someone incompetent can say that.

View attachment 74792
The only sign of 250MHz oscillations were the elevated mains spuriae and some light audible hum. Who cares? And who finds the real reason? Audioband measurement?
If you’re stuck in inappropriate ways of doing audio amp design rooted in old fashioned thinking, then I’m afraid that people are going to bore you. :D

Oscillations are a different thing than the sorts of loop stability issues which square waves and scopes can pick up. And if I’m not trying to do something useless like design for response up into the shortwave region, these show up perfectly well on digital scopes and yes, AP analyzers. So please spare me your condescension.
 

Geert

Major Contributor
Joined
Mar 20, 2020
Messages
1,937
Likes
3,526
Phase issues become audible in well treated rooms and more and more audiophile to treat their rooms. <... >. Exactly where the limit goes isn't something easily said or researched. It will depend on several factors.
I agree. My listening room at home isn't treated and I can easily hear the phase correction my DSP does in the low end (bass reflex group delay). However I can't hear 4th order crossover phase correction. Blind tested extensively with a script that randomly switches DSP's correction filters, succes rate 100% when using music with sufficient and accurate low end content.
 
Top Bottom