Transient response.

[Pearljam5000]

I don't see it mentioned here a lot.
How can you tell if a speaker has good transient response?

 

[fieldcar]

Transient response and frequency response are essentially the same thing. Edit: it's explained a little past half way into the video.

Watch this:

 

[aarons915]

This is what I love about this forum, very first response and the thread is done. No endless pages about fast woofers and fast tweeters lol

 

[Beave]

If my sister's dog catches a bird in the back yard, does that mean her woofers are faster than her tweeters?

 

[dfuller]

Transient response and frequency response are essentially the same thing. Edit: it's explained a little past half way into the video.

Watch this:

mm, yes and no. Part of transient response is also how quick the speaker comes to a stop, which we can see in CSD waterfalls.

 

[Beave]

mm, yes and no. Part of transient response is also how quick the speaker comes to a stop, which we can see in CSD waterfalls.

...which are also related to frequency response.

 

[dfuller]

...which are also related to frequency response.

Related yes, the same no.

 

[Rock Rabbit]

In simple terms: the driver is a resonant circuit, driver in a box makes a high pass filter. The Q of the filter defines if the frequency response have a peak at the cut frequency, for fast transient response the Q must be less than one.
More simple: a closed box design with flat response is fast!...more than a vented design

 

[800]

In simple terms: the driver is a resonant circuit, driver in a box makes a high pass filter. The Q of the filter defines if the frequency response have a peak at the cut frequency, for fast transient response the Q must be less than one.
More simple: a closed box design with flat response is fast!...more than a vented design

Does this apply across the spectrum, or just near Fs?

 

[Rock Rabbit]

Does this apply across the spectrum, or just near Fs?

It's valid in the "rigid piston" range up to cone break up (not single piece movement), and very complex at xover points. But any driver at "high frequency" (few octaves over resonance) start to rise impedance and produce little output.
There's a tradeoff between speaker efficiency and transient response and due to the bandwidth limit of a driver the transient aspect is less relevant in the design

 

[theyellowspecial]

Psychoacoustically speaking my subjective impression is speakers with more early reflections have worse transient response. Of course, that's completely unrelated to the mechanical properties of the speaker.

 

[Duke]

Psychoacoustically speaking my subjective impression is speakers with more early reflections have worse transient response.

Your subjective impression is correct:

"Transients are not corrupted by reflections if the room is large enough - and 10ms of reflections free time is enough." - David Griesinger.

Imo the implication for smaller rooms is, speakers which are directional enough that they can be AIMED to avoid significant reflections within the first 10 milliseconds of the direct sound, at least in the horizontal plane. The floor and ceiling bounces are more difficult to avoid, but are psychoacoustically not as detrimental as early ipsilateral reflections, according to Geddes and according to his understanding of Griesinger.

 

[ppataki]

In simple terms: the driver is a resonant circuit, driver in a box makes a high pass filter. The Q of the filter defines if the frequency response have a peak at the cut frequency, for fast transient response the Q must be less than one.
More simple: a closed box design with flat response is fast!...more than a vented design

I think this is very true, also based on the below:

I would have a question though: if I design a closed speaker with a low Q value (to respect the above) what will happen if I use a low shelf filter to "flatten" the frequency response? Will that have a detrimental effect on the transient response?

I have made some tests with my Audio Nirvana full range speakers

This is the normal step response:

And this is the step response after applying the low shelf filter:

I believe the latter one looks better since it resembles more to the ideal step response curve but I am open to any comments

Thank you

 

[andreasmaaan]

Transient response is not only a function of the enclosure type and tuning, but also crossover filters.

For min-phase filters (found in the vast majority of crossovers), the steeper the slope of the filter, the higher the degree of phase rotation (approximately 180 degrees for a second-order filter, 360 degrees for a fourth-order filter, and so on). And the higher the Q of the filter, the faster the rate of rotation.

Internal reflections within enclosures will also affect transient response, as will baffle edge diffraction, higher order modes in horns/waveguides, and essentially any other source of linear distortion.

The thresholds at which departures from ideal transient response become audible is another question ofc

 

[andreasmaaan]

I would have a question though: if I design a closed speaker with a low Q value (to respect the above) what will happen if I use a low shelf filter to "flatten" the frequency response? Will that have a detrimental effect on the transient response?

Just to clarify, was there actually a boost in the bass before you applied the low-shelf filter, which the filter corrected? And if so, was this boost in the speaker’s anechoic response or just in the in-room steady-state response?

 

[ppataki]

The low shelf filter is the boost itself
In the listening room at the listening position

 

[trl]

...which are also related to frequency response.

You need to elaborate a bit more, please.

AFAIK this is what a step response means: https://en.wikipedia.org/wiki/Transient_response#:~:text=In electrical engineering and mechanical,the equilibrium of the system and in my opinion two audio sources with identical freq. response across audible spectrum might have different transisnt responses to, let's say 1 kHz short signals, something similar with DAC filters.

Another good example of a transient response might also be: "For example, the switching time of a bipolar transistor and the reverse recovery time of a diode are examples of transient response", source https://en-support.renesas.com/knowledgeBase/16978650.

Extrapolating the above to headphones/speakers drivers, my guess is that transient response would be the recovery time of the driver before the applied electrical signal ceases and the driver settles down completely. Now if this recovery time of the driver (after the audio signal stops) has any audible energy out of it, it is another story.

L.E.: I do understand that an audio driver with a low transient response may have a poor freq. response, due to driver's inability of reproducing higher frequencies, but the transient response is not related to freq. response, but vice-versa.

 

[andreasmaaan]

The low shelf filter is the boost itself
In the listening room at the listening position

Yeh thanks, I understand that, but what I mean is: when you say the boost “flattens” the response, do you mean (1) that it was not flat (ie there was a lack of bass before the boost) and (2) if so, was that lack of bass measured anechoically or in-room?

 

[ppataki]

There was a lack of bass, hence I boosted it with a low shelf filter, measured in-room
Then I measured before/after the step responses, that is what you can see above

 

[trl]

An interesting similar thread could be read here: https://www.audiosciencereview.com/forum/index.php?threads/headphones-transients.13113/.