Hi M, thank you for taking the time to point out exactly what i said to cause consternation.
And I now see my imbedded assumption that most folks probably don't share.
Which is... on a sealed sub with either a Linkwitz transform or low-shelving used to extend low freq response ....
I view a high-pass filter as mandatory.
Excursion has to be frequency limited when there is boost all the way to DC, ime.
Or, to avoid boost to DC without a high-pass, use narrow band PEQ's instead, to get the desired low end boost/extension.
Which also accomplishes frequency-based excursion limiting.
Both of these methods of limiting excursion will of course increase phase rotation / group delay.
(which may or may not matter)
So yes, I stand by my statements...realizing i have an imbedded assumption that a high-pass is needed, and that should be stated upfront.
And i guess that assumption rests on another imbedded assumption...which is we all generally overestimate our home audio subwoofers to maintain linear response as volume is turned up.
Which is no doubt built on yet another assumption...... that occasionally, we all turn up the volume past linearity. 😀
ASSumptions are eating me alive! lol
Never knew people could get so "passionate" and strongly emotional attached about something as simple as an highpass filter......😆 😳
Hi mark100,
Regarding the high pass filter, what do you think about my idea from post no. 10,987? I have been thinking about it further, and even if one does not make the two lines parallel, as long as the equalization does not exceed the difference at "DC" as you put it, there does not seem to be a need for such a filter. Or, am I missing something?
Ha, ha, we all are guilty of that, and not only during communications on informal platforms. Hence a subsequent discussion (hopefully) clarifying the raised issues.
Hi b-force,
Kindest regards,
M
Yes, with the undisclosed assumption, you are, of course, correct. Thank you for clarifying.
Regarding the high pass filter, what do you think about my idea from post no. 10,987? I have been thinking about it further, and even if one does not make the two lines parallel, as long as the equalization does not exceed the difference at "DC" as you put it, there does not seem to be a need for such a filter. Or, am I missing something?
Ha, ha, we all are guilty of that, and not only during communications on informal platforms. Hence a subsequent discussion (hopefully) clarifying the raised issues.
Hi b-force,
I do believe that you had missed the point, at least on my part, which is to understand mark100's arguments.
Kindest regards,
M
Hi b_force, you made me laugh, thx. 🙂
Yes, i'm passionate about audio, for sure. But emotional ? Nah....it's all just another day in engineering trenches in the end....
btw, thanks for both the informative content of your posts lately, and their pragmatism as well.
Hi ya mefistofelez, ok, here's a quick input of the parameters you gave me into Hornresp...could be mistakes...
(are you certain of Le...never seen anything that low) (or Vas that high)
And perhaps most importantly, is 30 watts rated power....that's what i use below.
Box volume is 160L....about 0.6 qtc. I just choose something rather large to extend the low end naturally.
Here is 1 watt with no shelving boost on left, and a couple of shelving filter applied on the right.
And here's the displacement of the 1 watt responses. As xmax is 4.6mm, the shelving filters are still fine to use without a hpf even if there was 10Hz signal.
Here's the two shelving filters applied. An f-3 of about 30Hz...not going for anything lower...
Here's 30 watt SPL, followed by 30 watt displacement.
No EQ on left again, two shelves on right,
No EQ is still safe at 30w. But In my mind, the shelving displacement chart is screaming for help.
Here's 10w displacement chart, still doesn't stay within xmax once below 40Hz.
Yes, I do believe in a hpf when using a Linkwitz transform or shelving 🙂
(But maybe I screwed up something doing this so fast, or T/S are way off.)
Hope that helped /made sense...
Hi mark100,
thank you for the simulation.
I am rather surprised by the difference in the simulation by the different software packages because my understanding was that the simulation is accurate for the lowest frequencies. Could that be attributed to the difference in Vb (Vb=160l vs Vb=130l)? Can you please p.m. me the image of the Hornresponse input parameters that you used?
Ideally I would measure the test enclosure, but it has been windy and my REW is on strike, reporting wrong format received from UMIK-I. When/if I fix it, do you think that near-field measurement would be a good representation at these frequencies?
Kindest regards,
M
thank you for the simulation.
Well, the parameters are from the Altec data sheet. Although I am not sure that they can be trusted because GM, who is knowledgeable regarding Altec, remarked that the T/S parameters have not been consistent. However, that is immaterial for our purpose, since (hopefully) we both use the same parameters set.
I am rather surprised by the difference in the simulation by the different software packages because my understanding was that the simulation is accurate for the lowest frequencies. Could that be attributed to the difference in Vb (Vb=160l vs Vb=130l)? Can you please p.m. me the image of the Hornresponse input parameters that you used?
Ideally I would measure the test enclosure, but it has been windy and my REW is on strike, reporting wrong format received from UMIK-I. When/if I fix it, do you think that near-field measurement would be a good representation at these frequencies?
Kindest regards,
M
Increasing Direct sound raises the truenesses of the signal from the loud speaker to your ear. As long as the speaker is outputting an accurate signal, the end result is more accuracy to the source.
The room messes up the signal more than the speaker does a lot of the time. Hence, I focused my design to include the most Radiation area possible
It wasn't the beaming, in terms of diminishing polar width, it was the directivity increase over a larger part of the spectrum, that I liked about the horn.
Having large drivers, baffle and mouth width, will only increase directivity
I am also under the impression that larger radiation area increases Direct Sound, in a small room as well.
All in all, I did all I could think of to increase Direct Sound except for Cardiod. Though maybe I'll figure out how to do that passively one day
Accuracy of the loudspeaker+room can be judged by the Taking an impulse response. Its not the whole picture but a good tesr to be used at different spl's, and proximities.
There can be some obvious differences due to the different Q parameters used in some software that may not be duplicated in the other.
Ql is the system's Q at Fb due to leakage losses (sealing of the cabinet, etc.).
Qa is the system's Q at Fb due to absorption losses.
Qa can have an effect on apparent volume in the same way stuffing does in real life.
Nearfield with baffle diffraction added is usually accurate below the frequency where you get a nearfield dip.
There is no problem with your idea if you are happy with the frequency response and SPL capability of the sealed box you show.
The problem comes when you want to make it flat to 30Hz before rolling off and get to 110dB, the excursion is going to take off as mark shows.
If it still is this woofer, it shows a max 100 dB at 30 Hz, let's stay within that limit? That was the original plan, wasn't it?
Hi fluid,
thank you for your response.
Hi wesayso,
Kindest regards,
M
P.S. Cince you both mentioned 30 Hz, I reviewed my posts, but cannot find mentioning the figure. Am I overlooking something?
M
thank you for your response.
Indeed, there may be more variables than I noted in my reply to mark100. That was why I asked for his input to the Hornresponse, so that I can make a meaningful comparison.
Thank you for confirming what I gathered so far from reading about such a measurements. I had been contemplated ground plane measurement, as I have a nice clear area, but the winds . . .
Indeed, but, as I noted elsewhere in my posts, the simulations I posted have been intended for support of my posts, and not simulation of an enclosure/driver combination for implementation.
Hi wesayso,
Nope, please see my response to fluid above. I have literally tens of simulation of this particular Altec driver in different enclosures because the driver had been successfully (my opinion), used for a high (about 1000 Hz) crossover, which I contemplated for a 2-way speaker. So I just picked the simulation illustrating/supporting my arguments.
Kindest regards,
M
P.S. Cince you both mentioned 30 Hz, I reviewed my posts, but cannot find mentioning the figure. Am I overlooking something?
M
You didn't mention 30 Hz, but did mention/speak about a 9 dB allowance, Which would bring you at 100 dB at 30 Hz maximum (to stay within x-max I supposed).
Mark100 boosted the output towards about 108 dB at 30 Hz, that's going to make a difference. That was my point.
Ground plane is good but still needs some diffraction simulation to account for the doubled baffle size from the ground plane, wind is usually a big problem at high frequencies bit less so at low, it might not be as bad as you think.
For simplicity in measuring a standard woofer in box that is not a dipole, cardioid or have some other intentional directivity feature, nearfield and baffle diffraction is very accurate.
This is kind of the problem with hypothetical situations, are they representative of a realistic use case? I think that a woofer that follows a sealed roll off beginning around 100Hz can sound really good in a room and the reinforcement from the room tends to compensate the roll off quite well. In an anechoic chamber or outside it would sound bass shy. But if the response does need to be flat to 30Hz then the reality is very different.
Mark equalized to that frequency in his simulation. You are asking about a hypothetical apple, but mark says a real peach is much better 🙂
Hi wesayso,
Hi fluid,
I think that we had beaten the issue to death, with the result of understanding it.
Kindest regards,
M
Thank you for the clarification. To my defense though, it was hard for me to understand your point form your laconic statement.
Hi fluid,
O.K., I am currently reinstalling the OS as nothing helped with the REW problem, so hopefully I will be able to perform some measurements soon.
The point was my disagreement with mark100's statement re GD and phase rotation, which has now been clarified. The intended measurement will answer the question of reality.
Yeah, I understand it now form wesayso's reply. Now, ice-cream beats both mentioned fruits. 😉
I think that we had beaten the issue to death, with the result of understanding it.
Kindest regards,
M
I agree with your ice cream comment. Make mine Cherry Garcia or homemade vanilla bean. But, too much of it will produce too much bottom end, an increased double hump group delay, and additional phase rotation. 😆
Camplo, A and B here are the same, and will be the same at the listening position.
The Blue being Direct energy, Yellow is the listening point.
Thinking about omnidirectional frequencies. If the sound expands omnidirectionally from from all points of the radiation source.
The larger Source will provide more Direct energy to the listener. It will also provide more sound to the Boundaries sooo....
So what I theorize here is that.....The more Radiation area you have... You are creating more paths, directly to the radiator and the listening point, in so many words. There will be more energy from the speaker...that reaches you sooner, via more direct paths. That should be enough to increase Accuracy in between the speaker and the ear, in a room, but to what degree.
What that means to perception. I'd think it would relate to what Paul said
What I really am trying to figure out is how this relates to the physical sensation. Maybe the same idea but applied to the whole body? The larger radiation source will provide a wide spread of the energy, maybe communication to wider portion of skin
What that means to perception. I'd think it would relate to what Paul said
I would predict that we have it already in our natural ability to judge source size, to some degree. When things are made "larger" they can be easier to "read/see" even if all other things are larger in the same proportions
What I really am trying to figure out is how this relates to the physical sensation. Maybe the same idea but applied to the whole body? The larger radiation source will provide a wide spread of the energy, maybe communication to wider portion of skin
Last edited: