- Joined
- Jan 15, 2020
- Messages
- 6,927
- Likes
- 17,026
Interesting is also this paper on why bassreflex is often not suited for very low tunings, for example 20 Hz subwoofers
-
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!
Transient response.
- Thread starter Pearljam5000
- Start date
- Joined
- Jan 23, 2020
- Messages
- 4,343
- Likes
- 6,723
Interesting read, though most of it was over my head. Seems to give logical and well reasoned explanations as to why sealed sounds "faster".
The 2 problems with acoustic port loading for very low frequencies are non-linearities and pipe resonances.
At low f with insufficient port area the air velocity gets too high, which creates loss of efficiency, distortion and noise.
Trying to increase the area gives a very long port, which moves the 1. pipe resonance so low in frequency that it enters down into the pass-band of a subwoofer.
Both problems can be solved.
Small-signal transient response for a ported will be equal to a sealed with eq to bring the frequency response back in shape. This is a minimum-phase system, where the resulting step response simply is a result of the frequency response, the ringing at cut-off is a result of the roll-off characteristics.
What is more important, is the large signal transient response. Now the construction of the driver and for a ported/acoustically loaded its port design as well, is what will determine how it performs. Acoustic loading can increase output capacity significantly due to reduced cone displacement.
In all practical situations, what happens when the bass-system is placed inside a room, is what determines (small-signal) transient response. What we observe now, is a frequency response that is very different from the calculated, where now the room completely dominates. Time domain behavior follows the frequency response, with similar disastrous effects. Fortunately, this can often be significantly improved with eq. Then we understand that whatever frequency response the subwoofer was born with does not matter, it is the output capacity that limits what can be achieved.
- Joined
- Jan 23, 2020
- Messages
- 4,343
- Likes
- 6,723
This is what I've slowly come to believe. It's all about output and extension. With enough of both, I can EQ a really great in room response. It's why I think CEA-2010 is the best overall spec we have for subs.
Yes. Room interaction effects are vastly greater than the minor frequency response differences between well-designed subwoofers.
At the tuning frequency, a vented box has a theoretical 6 dB advantage over a comparable-sized sealed box. Of course this is not across the bass spectrum, but it does imply that EQing the sealed box for similar low-end extension calls for significantly greater thermal and mechanical power handling for the woofer, and perhaps four times as much amplifier power, for equivalent output capacity down to the vented box's tuning frequency. My observation has been that high-power compact equalized sealed subs are more likely to have something fail than are comparable-output-capacity vented box subs because of all that additional heat.
And then add the increased output capacity due to reduced excursion to those 6dB, and it adds up to a quite significant number.
For power, the impedance must be taken into account. A ported/loaded design will have a low impedance, resistive, at the tuning frequency, which gives a power loss close to Rdc at this frequency. Above tuning, a loaded design can have a high, nearly resitive load across a quite wide frequency range - nice, because then the power loss will be very low with ordinary music signal.
Look how far this got off the original track. Some new aspect was brought in, because otherwise the case was lost.
However, bassreflex / ported is exactly complicated due to weakly formulated requirements. How deep, and how loud at each frequency on average and peak. Once the requirements are set, the design decision is straight forward. Granted!
However, bassreflex / ported is exactly complicated due to weakly formulated requirements. How deep, and how loud at each frequency on average and peak. Once the requirements are set, the design decision is straight forward. Granted!
Wow! This is beefy (as Holland's frikandel is, though, couldn't resist, appologies ... ). It's irrelevant, as far as I didn't read it. You would otherwise notice, no Thiele/Small parameters, no alignment ... and the electromagnetic drive of the driver from a constant voltage source is dealt with as a special case? He states that that would (we know: by Qe exactly) introduce "some damping". Really?
dasdoing
Major Contributor
the high pass obviously creates a bump on the waterfall since the group delay isn't compensated (it's not an artifact). but the waterfalls of the ported speakers show more than that; clearly visable as they adobt a triangular shape. also the tail starts steep and gets flatter over time. clearly a resonance at play.
I would love to see those plots in the spectogram view as it clealy shows what is delay, and what is ringing.
also my example comparing similar roll-offs ported and sealed was totaly ignored
here we can see the resonance continue after the easiely identifed HP delay
others are almost as obvious, others less
Sorry for revitalizing this again. I wish to turn it into positive.
Regarding the paper (https://www.grimmaudio.com/wp-content/uploads/RMS-white-paper-3-Bass-reflex.pdf), seen in its context, it is just a part of a series. From this it can be understood, why the reference to Thiele/Small is missing (yes, I read it all):
Eventually the author admits, that his efforts didn't yield sufficient outcome. Kudos for that honorable statement. Again kudos! Thank You.
But how this singled out "white paper" made it onto the Grimm Audio website, I won't comment.
Your enthusiastic attitude cherished. I developed a measuring mike with active feedback. It was absolutely impossible to take even a crude number of low frequency noise. All that allday rumble around, still late at night was obviously far above its self-noise. I think it is mostly from traffic.
Then room reflections will contribute in the near field also, at some point in time.
Taken many more things into account, such measurements You present as evidence for open questions have to be taken with a grain of salt. From my very personal experience they are more estimates.
Anyway, in nearly all the scientific community the issue of group delay, especially in-room and that low in both frequency and value is considered a non-issue. Think of the steepest brickwall filters the industry uses everywhere.
Wouldn't it be nice to stop worrying?
KSTR
Major Contributor
You mean the tail at 140Hz of the Acoustic Energy AE2 here? This not a port issue, at any rate not directly related to being ported. It's a different ill-effect hard to nail down without hands-on inspection what's going on here. Same for the HHb Circle 5A, also at ~140Hz. Something is resonating mechanically I'd guess.
EDIT: Could be a measurment problem as well, something gets excited to resonate, speaker stand, mic stand, whatever (I note this all was done in anechoic chamber by Studio Sound Magazine but no info given how exactly). The 140Hz tail is often there to varying degrees which is suspicous.
Last edited:
dasdoing
Major Contributor
there is indeed always a rumble almost anywhere, but is there in anechoic chambers? youtubers in one always act like they are completly silent
It would be nice if I could ignore ringing as most can. I can't even sing and relax in the shower as all I hear is resonances lol
Didn't check the sources but this is what Wiki says:
Bass reflex - Wikipedia
the increased time delay you guys say exist, but as I understand it you guys say that damping is not afected? this is afaiui what causes the frequencies to "stop later". are the sources wrong? it just matches what I see on those plots. I don't want to be right in any way. I just want to stop (or not) worrying
I have read through it too, but his reasoning for can't tune port to very low frequency is only because the mass needs to increase and stiffness needs to decrease. I think maybe he does not want subs too be too big, but not everyone is bothered by that.Sorry for revitalizing this again. I wish to turn it into positive.
Regarding the paper (https://www.grimmaudio.com/wp-content/uploads/RMS-white-paper-3-Bass-reflex.pdf), seen in its context, it is just a part of a series. From this it can be understood, why the reference to Thiele/Small is missing (yes, I read it all):
Eventually the author admits, that his efforts didn't yield sufficient outcome. Kudos for that honorable statement. Again kudos! Thank You.
But how this singled out "white paper" made it onto the Grimm Audio website, I won't comment.
He talked about the tuneing frequency will ring longer than closed box. From the graph posted by KSTR bove, the driver start and stop at the same time, it is just the phase of that frequency lags a bit. I am puzzled by this because the two sayings are completely different. Is it because the paper used a ported speaker that has different frequency response as closed speaker? Then his reasoning of ring for longer will be wrong regarding the box type, because ring only depends on frequency response.
He also talked about when we damp the port and passive radiator to optimal state we will lose the benefit of port or passive radiator. I am thinking can this be true? I think we can use another tuning that will not result a small peak at tuning frequency so there is no need to damp since frequency response is perfect, then we will have the benefit of port, isn't it?
For me the case is closed, as every aspect seems to be mentioned. My point of view is having a degree relevant to this topic. From that I know how hard it is to understand the details and then bring them back into alignment for the full picture. A randomly picked "white paper" which supports the own standpoint by the title alone (Grimm Audio) doesn't help with that. If You are interested in the design of ported You may want to open a new thread.
Similar threads
