OK, it took a few posts, but I think we are getting back to my point apropos the OP: Q of .7 is not a criterion of performance. What it is, if I am not mistaken, is a computationally simplifying assumption that - if you buy-into typical sims - will give you a mathematically tractable result for a constant voltage input* in designing a box.
Among the other things wrong with buying-into that world view is the mistaken notion that you want a flat bass output. You probably shouldn't want a flat output. What you do want depends on a lot of things but probably favours a boost somewhere near the bottom end of what your speakers can muster.
Ben
*fussing over a constant voltage input was important decades ago. Hardly important today unless you have firm convictions (or tube gear) that it is the "right" way to handle signals end-to-end.
Among the other things wrong with buying-into that world view is the mistaken notion that you want a flat bass output. You probably shouldn't want a flat output. What you do want depends on a lot of things but probably favours a boost somewhere near the bottom end of what your speakers can muster.
Ben
*fussing over a constant voltage input was important decades ago. Hardly important today unless you have firm convictions (or tube gear) that it is the "right" way to handle signals end-to-end.
Last edited:
The biggest difference between Just a Guy and Calvin is the assumption that the Builder understands how the driver interacts with the box or does not understand.
I side with Calvin, one MUST first understand what one is designing and secondly one MUST be able to recognise when the answer is close to right or a complete work of fiction.
I must be from 1932 (except my parents were still kids and thousands of miles apart) in that I come from a pre computer world and learned to do everything long hand.
The one bit I was not trained to do was build it wrong and wait for it to fall down.
That was left as a mental exercise: to understand why it fell down and improve on it.
It's the same with speakers.
One should understand what they do and what the adjustments do to that performance.
My first speaker book was by Briggs. It did not help me understand.
I had to wait for "Bullock on Boxes" to get to the "understanding".
Now I was ready for the simulator.
Modeling is great.
It is applied in many forms to all engineering, not just electronics.
But the "basics" need to be known and understood first.
Knowing about Q for a sealed box and how to manipulate that Q for different performance is, for me, the important first step.
Simulators come AFTER understanding.
I side with Calvin, one MUST first understand what one is designing and secondly one MUST be able to recognise when the answer is close to right or a complete work of fiction.
I must be from 1932 (except my parents were still kids and thousands of miles apart) in that I come from a pre computer world and learned to do everything long hand.
The one bit I was not trained to do was build it wrong and wait for it to fall down.
That was left as a mental exercise: to understand why it fell down and improve on it.
It's the same with speakers.
One should understand what they do and what the adjustments do to that performance.
My first speaker book was by Briggs. It did not help me understand.
I had to wait for "Bullock on Boxes" to get to the "understanding".
Now I was ready for the simulator.
Modeling is great.
It is applied in many forms to all engineering, not just electronics.
But the "basics" need to be known and understood first.
Knowing about Q for a sealed box and how to manipulate that Q for different performance is, for me, the important first step.
Simulators come AFTER understanding.
Last edited:
As stated earlier a Q of 0.707 will give you a maximally flat response in amplitude with no peaking. This is the signal out of the sub and with room gain this will often cause a peak in the amplitude response it depends on how low the sub plays and how big the room is.
A Q of 0.5 is close to a Bessel function and will give a maximally flat response in group delay. Group delay is a measure of the delay through a ciruit vrs the frequency entering the circuit. A bessel function still has goup delay but it is all negative rather than peaking at a frequency the turning over and going the other way.
I find a Q of 0.5 works better in most rooms with large subwoofer. However I dont know if this is the smoother roll off of a 0.5 Q or the lack of peaking in group delay or som other factor.
So I would leave it alone if it has a Q of 0.5.
Regards,
Andrew
A Q of 0.5 is close to a Bessel function and will give a maximally flat response in group delay. Group delay is a measure of the delay through a ciruit vrs the frequency entering the circuit. A bessel function still has goup delay but it is all negative rather than peaking at a frequency the turning over and going the other way.
I find a Q of 0.5 works better in most rooms with large subwoofer. However I dont know if this is the smoother roll off of a 0.5 Q or the lack of peaking in group delay or som other factor.
So I would leave it alone if it has a Q of 0.5.
Regards,
Andrew
AndrewT, I completely agree with you even though none of my parents were born in 1932 🙂
I also rather learn the underlying physics first and then apply simulators plus common sense. I have two engineering degrees so the basic physics (and the math) I can mostly handle, but as everything it takes specialized knowledge to get right. I bought Everest's master handbook of acoustics, Toole's Sound Reproduction, and The Loudspeaker Cookbook by Dickason. I'm very interested in book suggestions to learn about box design. I looked up Bullock on Boxes and looks interesting. Do you recommend another book instead?
I've been learning quite a bit from this forum in the last 4 months, but a good engineering book should provide a quantum leap in my learning.
I also rather learn the underlying physics first and then apply simulators plus common sense. I have two engineering degrees so the basic physics (and the math) I can mostly handle, but as everything it takes specialized knowledge to get right. I bought Everest's master handbook of acoustics, Toole's Sound Reproduction, and The Loudspeaker Cookbook by Dickason. I'm very interested in book suggestions to learn about box design. I looked up Bullock on Boxes and looks interesting. Do you recommend another book instead?
I've been learning quite a bit from this forum in the last 4 months, but a good engineering book should provide a quantum leap in my learning.
AndrewT, I completely agree with you even though none of my parents were born in 1932 🙂
I also rather learn the underlying physics first and then apply simulators plus common sense. I have two engineering degrees so the basic physics (and the math) I can mostly handle, but as everything it takes specialized knowledge to get right. I bought Everest's master handbook of acoustics, Toole's Sound Reproduction, and The Loudspeaker Cookbook by Dickason. I'm very interested in book suggestions to learn about box design. I looked up Bullock on Boxes and looks interesting. Do you recommend another book instead?
Sorry for the double post. I wanted AndrewT to get noticed I had replied on his post but realized I had not quoted him in my first post.
Just noticed the post on the servo nature of this driver. Do you have it running from the servo amplifier or a standard amplifier. Either way
I wouldt change it it just if you are using the servo function then the acual response will be determined mostly by the servo characteristics rather than the box.
Andrew
I wouldt change it it just if you are using the servo function then the acual response will be determined mostly by the servo characteristics rather than the box.
Andrew
Hi,
short OT, solving the York Minster mystery 😉
Stacking big rocks would probabely have failed as foundation in a swamp.
So they did different.
They laid multiple layers of huge Oak trees, each layer perpendicular to the other.
Now that's what I call thoughtful or simply brilliant.
16000tons of stone resting on wood for 800years!
OT off
jauu
Calvin
short OT, solving the York Minster mystery 😉
Stacking big rocks would probabely have failed as foundation in a swamp.
So they did different.
They laid multiple layers of huge Oak trees, each layer perpendicular to the other.
Now that's what I call thoughtful or simply brilliant.
16000tons of stone resting on wood for 800years!
OT off
jauu
Calvin
OT back on.
How many times did they practice it in the preceding thousand years?
Some roman roads are laid on an organic layer and they still exist.
How many times did they practice it in the preceding thousand years?
Some roman roads are laid on an organic layer and they still exist.
Why are you implying that I'm saying a simulator is a substitute for knowledge? I never said anything like that.
What I DID say is that simulators are faster, easier, give a ton more information and that having only box Q and nothing else as a goal is a dubious pursuit at best.
If you want to pick sides in this, do it using the preceding sentence as my side, not a position that simulators can replace knowledge. I never said that, would never say that. I've spent years studying this audio stuff, saying something like that would be the exact opposite of everything I believe.
That's was not what your previous statements sounded like.
It appeared to be saying that simulators are good, for everything, tons of data over-rules all other needs and requirements.
Now you are making more sense.
Learn how a speaker operates. Learn how the adjustments make an effect, that could be welcome or not. Learn how to use the software. Learn how to check the results of the questions one asks of the software.
THEN, carry out the virtual experiments to determine which adjustments give the required result.
This last paragraph is very different from what you were proposing a few posts back.
It appeared to be saying that simulators are good, for everything, tons of data over-rules all other needs and requirements.
Now you are making more sense.
Learn how a speaker operates. Learn how the adjustments make an effect, that could be welcome or not. Learn how to use the software. Learn how to check the results of the questions one asks of the software.
THEN, carry out the virtual experiments to determine which adjustments give the required result.
This last paragraph is very different from what you were proposing a few posts back.
That makes a lot of sense to me. But the problem comes at the end. Sims are pretty much "black boxes" to everyone (even acoustic experts can't validate the code). So you are essentially buying-into the world-view and assumptions of the person who crafted the sim.
So I always avoid black boxes. Calvin's musings make good sense: if you can't establish designs on the back of an envelope, you are already in over your head. If that's the case, you really need to learn more about speakers and then take out a scrap of paper to do your calculations, as Calvin suggests.
Sims may appear to offer you flexibilities of certain kinds (the parameter inputs) but not of other kinds (like trying to conjure-up the biggest box that meets the spousal approval or why it makes sense not to have equal bumps on ported boxes or, back to OP, how to get the lowest possible Q).
Ben
Last edited:
I don't think so, my story hasn't changed. In my last post I just summarized what I've been saying all along. I have no idea how you misread my intentions so badly. I went back and read it over.
I gave a bunch of features that simulators provide, like excursion data, impedance data, velocity data, .frd and .zma files, potential max spl data etc. You absolutely can design without any of that, but why would you want to? You could run through the formulas to calculate all that without a computer, but why would you want to?
A lot of us can estimate all this stuff fairly accurately without running a sim, and I think that's where this misunderstanding comes from, it seems like at least a couple of people think I'm implying they can't. But if you are going to take the time and go to the expense of building something, why not take the 60 seconds to run a sim first? I always do.
And no one has addressed the high inductance issue. The formula absolutely will not give you the right answer for high inductance drivers. I can sim high inductance drivers pretty well with a minor tweak.
There's nothing wrong with you guys doing the math, I choose not to. I find my simulators have a lot more power, flexibility and versatility than a single formula.
I love chatting with you, Ben. It always makes me smile. You make it sound like simulators are witchcraft. I can assure you that there is math behind the code, and physics behind the math. Sims have proven very accurate and useful several thousand times over (at least) just on this forum alone.
If you want to know what size of box your spouse will tolerate you can just ask. Unfortunately if you ask your spouse what the measured response, impedance and max spl (among other things) of a speaker will be you probably won't get much useful info. That's why simulators are helpful.
OP doesn't need the lowest possible Q, and the lowest possible Q is a terrible goal. All that will give you is a serious lack of low bass, and you already mentioned you prefer a bit of a bass bump at the lower frequency limit so I'm not sure you even understand what Q means.
Anyway, simulators are a fantastic tool, and you could learn a lot if you learned to use one. I can help you learn if you like, I'm confident that it will open up a whole modern world of insight for you.
- Status
- Not open for further replies.