There's a lot of questions here but the answers are pretty simple.
ANY sound sources playing the same frequencies will interfere with each other if they are more than 1/4 wave apart. That is fact.
Tones, overtones, pitch, sine vs sawtooth vs square, violin vs clarinet, these are all just differences that do not change the fact that ANY sound sources playing the same frequencies will interfere with each other if they are more than 1/4 wave apart.
The more similar the sound the more they will cancel if they are out of phase.
If they are in phase and playing the same frequencies and they are more than 1/4 wave apart you will see peaks and nulls in response depending on how far the speakers are apart and where the mic (or listener) is in relation to the speakers. That's it, that's all, just altered frequency response, peaks and dips in response based on distance from the sound sources.
This is all fact. It can be seen really easy in a simple crossover simulator. It's simple physics, the rules never change, even if it's a pipe organ or pipe organ speakers playing the sounds.
In THIS case it doesn't have all that much to do with bucks. OP very likely can't construct a horn and has no space to put it because he's got a huge stack of clumped speakers taking up all the space in the pipe room.
Horns are pretty cheap to make if you use OSB, which is fine.
First let me say I am glad that Mark has joined the discussion.
In his last post, he talked about efficiency. Speaking about those people who think the entire description of a loudspeaker is encompassed in the sim curve, a factor not addressed here correctly is loudness.
With real instruments (particularly organs and choirs of women), there are instantaneous peaks far bigger than your RMS calculator would indicate. True, it is a stretch to make speakers loud enough to project RMS levels but all but impossible to reach the instantaneous peaks. Maybe with tin horns you could try to replicate the treble of an organ... but no other way.
(This is true of home music systems too. I believe you need a whole lot of tweeter power to sound realistic, esp. enough drivers. I know that will sound odd to many people who think their tweeter never comes up short... on average.)
It is simple-minded to believe that the couple of parameters we routinely use to characterize sound really do adequately define it or the speakers that play it.
As an old (old, old, old) research guy, one disaster caused by adherence to sims (and to our beloved REW) is that they have removed us from watching the data closely on oscilloscope traces. You'd soon enough notice tweeter power demands by watching mic feeds.
Yes, you can install loudspeakers to replace organ pipes. But they will sound like loudspeakers playing organ music. Nobody who loves organ sound would mistake the two.
Separate point. While nobody has more respect than Danley at this forum, his website seems oddly "commercial" and while there are learned-appearing "white papers", there aren't any stats of performance that I have seen. Can somebody post some, please.
Separate point. I've yet to see somebody remember there are flues and reeds - quite different beasts. A lot of groping in the dark, trying to substitute logic for knowledge of organs. Not satisfactory.
Ben
In his last post, he talked about efficiency. Speaking about those people who think the entire description of a loudspeaker is encompassed in the sim curve, a factor not addressed here correctly is loudness.
With real instruments (particularly organs and choirs of women), there are instantaneous peaks far bigger than your RMS calculator would indicate. True, it is a stretch to make speakers loud enough to project RMS levels but all but impossible to reach the instantaneous peaks. Maybe with tin horns you could try to replicate the treble of an organ... but no other way.
(This is true of home music systems too. I believe you need a whole lot of tweeter power to sound realistic, esp. enough drivers. I know that will sound odd to many people who think their tweeter never comes up short... on average.)
It is simple-minded to believe that the couple of parameters we routinely use to characterize sound really do adequately define it or the speakers that play it.
As an old (old, old, old) research guy, one disaster caused by adherence to sims (and to our beloved REW) is that they have removed us from watching the data closely on oscilloscope traces. You'd soon enough notice tweeter power demands by watching mic feeds.
Yes, you can install loudspeakers to replace organ pipes. But they will sound like loudspeakers playing organ music. Nobody who loves organ sound would mistake the two.
Separate point. While nobody has more respect than Danley at this forum, his website seems oddly "commercial" and while there are learned-appearing "white papers", there aren't any stats of performance that I have seen. Can somebody post some, please.
Separate point. I've yet to see somebody remember there are flues and reeds - quite different beasts. A lot of groping in the dark, trying to substitute logic for knowledge of organs. Not satisfactory.
Ben
Last edited:
Clearly you've never even seen a simulator before. Hornresp shows efficiency and loudness (spl at xmax or any other level you want to see).
Based on your previous criticisms of simulators I strongly suspected you had never seen one, I think this proves it.
You are using words but I don't think you know what they mean.
By definition, RMS is .707x peak.
There is no such thing as an RMS calculator that I am aware of, and if there was all it would do is calculate .707x peak.
There's a lot more than a couple of parameters that go into a sim's input and output.
A sim will not tell you how something will sound, or if you will like that sound, but it can quantify just about everything else about it's acoustic performance.
How do you know? Have you ever heard a good organ speaker system?
You are by far the biggest critic of tapped horns on this forum. You admitted that you have never heard a tapped horn before in your life. Yet you continue to yammer on about how resonances are evil and tapped horns sound terrible despite the fact that your own beloved K-horn, the cornerstone of your system is MORE resonant than most tapped horns.
You've clearly never heard a well designed ported box in your life, or you would realize that you wouldn't be able to tell your beloved leaky sealed design from a well designed ported box if they were designed and eq'ed to have the same frequency response.
I don't think you have much if any grounds to speculate about how a well designed church system might sound.
There are performance stats on every product page and every product pdf.
Hey Ben, by the way, can you prove that adding a leak and a bit of stuffing to OP's box will do anything other than change the q a bit?
This is about the simplest technical question you could hope for, it should be pretty easy to answer, I've asked probably over a dozen times now and you ignore it like the plague. How about a straight answer for once?
What do flues and reeds have to do with anything? Are they ruled by a separate set of rules of physics? Are you suggesting it's impossible to produce their sound?
What is not satisfactory is the fact that you've been in a technical discussion for days now and haven't made a single technical comment. Just speculation, conjecture, and avoiding technical questions at all costs.
If you understood what the sims were telling you, you wouldn't have accidents like this. The sims tell you how the product will perform.
What you fail to realize is that the media content is NOT a part of the sim. You can look at that too, but it's something completely different. If you have very dynamic content it's going to produce a very dynamic result in the driver. If you blow the driver that's YOUR fault, not the sim's fault.
Last edited:
@just a guy.
You might want to tone it down a bit.
Your young enough to be my kid never mind Ben's kid.
If you go to high up on your soap box the way down can be a bit rough.
You are and intelligent fellow. I have seen some very insightful posts from you. But what you are typing in this thread shows your age and lack of experience.
As in you put your head in the sand when it comes to your beloved facts.
There are many more things unknown in loudspeaker design than there are known. In fact the more I work at this profession the more I have to admit that I have questions about things that I thought I understood. It comes with the territory. With greater knowledge and hopefully wisdom you learn that few things are as cut and dry simple as you purport them to be.
I respect mathematics. But I know from experience that of all the simulation programs I have used the words simulation sum it up. Simulation is synonymous with best informed guess. It literally depends on what you are trying to simulate.
None of the programs that you are referencing have the ability to model the enclosure resonances. And yet they exist. I regularly measure them. In fact you might be rather amazed by putting a microphone in front of a port and an oscilloscope or an RTA on the other end of the mic. Play a wideband pink noise signal and tell me what you measure.
Because they will inform you in real time that your concept of what happens in a ported enclosure is completely incorrect. You are missing the practical application of the knowledge you have.
Even more interesting set up a sine wave generator and an oscilloscope and a microphone and watch what you beloved vented enclosure will really produce.
Don't act like a big shot telling people that they do not know what they are talking about.
Some of use have paid our dues in the test and measurement department. And long before you were filling diapers and crying for mommy!
I have designed Anechoic chambers, calibrated measurement systems, designed drivers from tweeters to AMT's to planars to midranges to woofers and subwoofers. And a great many enclosures. When you have that under your belt we will talk about who knows what.
Until then act a little more respectfully. I'm all for being questioned. And all for even being corrected. But please watch how much of your own Koolaid you are drinking.
You might want to tone it down a bit.
Your young enough to be my kid never mind Ben's kid.
If you go to high up on your soap box the way down can be a bit rough.
You are and intelligent fellow. I have seen some very insightful posts from you. But what you are typing in this thread shows your age and lack of experience.
As in you put your head in the sand when it comes to your beloved facts.
There are many more things unknown in loudspeaker design than there are known. In fact the more I work at this profession the more I have to admit that I have questions about things that I thought I understood. It comes with the territory. With greater knowledge and hopefully wisdom you learn that few things are as cut and dry simple as you purport them to be.
I respect mathematics. But I know from experience that of all the simulation programs I have used the words simulation sum it up. Simulation is synonymous with best informed guess. It literally depends on what you are trying to simulate.
None of the programs that you are referencing have the ability to model the enclosure resonances. And yet they exist. I regularly measure them. In fact you might be rather amazed by putting a microphone in front of a port and an oscilloscope or an RTA on the other end of the mic. Play a wideband pink noise signal and tell me what you measure.
Because they will inform you in real time that your concept of what happens in a ported enclosure is completely incorrect. You are missing the practical application of the knowledge you have.
Even more interesting set up a sine wave generator and an oscilloscope and a microphone and watch what you beloved vented enclosure will really produce.
Don't act like a big shot telling people that they do not know what they are talking about.
Some of use have paid our dues in the test and measurement department. And long before you were filling diapers and crying for mommy!
I have designed Anechoic chambers, calibrated measurement systems, designed drivers from tweeters to AMT's to planars to midranges to woofers and subwoofers. And a great many enclosures. When you have that under your belt we will talk about who knows what.
Until then act a little more respectfully. I'm all for being questioned. And all for even being corrected. But please watch how much of your own Koolaid you are drinking.
I lika dah piktures!
So I did a sanity check on your simulation ( just a guy )
You can't miss the resonances that are present.
I agree with DrDyna that they will not be as awful as simulated.
But note that a number of factors are missing in this simulation.
The driver is simulated as a perfect plane piston.
No breakup modes or any other driver distortions are in this simulation.
There will be in real life as you are pumping in 100 watts or so into this sim.
There is no modeling of driver suspension nonlinearities either.
In short you have a simulation of a perfect driver in an enclosure.
A perfect driver does not exist.
And I have tried hard to make them.
Still have absolute faith in your simulator?
So I did a sanity check on your simulation ( just a guy )
You can't miss the resonances that are present.
I agree with DrDyna that they will not be as awful as simulated.
But note that a number of factors are missing in this simulation.
The driver is simulated as a perfect plane piston.
No breakup modes or any other driver distortions are in this simulation.
There will be in real life as you are pumping in 100 watts or so into this sim.
There is no modeling of driver suspension nonlinearities either.
In short you have a simulation of a perfect driver in an enclosure.
A perfect driver does not exist.
And I have tried hard to make them.
Still have absolute faith in your simulator?
I said that a long time before he did. Several times.
So what? It's a ported box. The sim will match the measurement if you accurately sim what you build.
So what? Cone breakup is at 1.2 khz on this driver. Here's the data sheet - http://www.parts-express.com/pedocs/specs/295-130-dayton-audio-st385-8-specifications.pdf
If you don't like the breakup notch it out, but at 1.2 khz I doubt it matters much.
So what? We are all well aware of distortion at high power, power compression and port compression among a whole host of other things. These are so commonplace and well known these things usually are not even mentioned because even beginners know all about these issues, expect them, and even know approximately when they are going to kick in and how bad they will be.
So what? That's what klippel is for. I assume by now you've learned how to read the graphs. Dayton driver are very conservatively rated. This driver will be fine at least until it's 8 mm xmax.
Absolutely I have absolute faith in my simulator, because like most people I know it's shortcomings and the things it fails to address. You do to. So do most people. My simulator has been proven thousands of times over to be accurate within it's limitations.
You are desperately grasping at straws here.
I'll mention this again since you apparently missed it. When measured in real life those resonances will be about +/-1 db ripple and only above 140 hz.
A small amount of stuffing will make that curve smooth as a baby's butt.
TB46 simulated this in Akabak as mltl to determine proper port placement to further reduce the ripple.
Those resonances are not a problem. If that's what you are stuck on, forgetting your lunch and having to buy lunch should be a much bigger concern, that's a problem orders of magnitude more serious than those resonances.
I agree with just a guy on this one. Resonances in a properly designed ported box are small enough.
@ Bach On
This is going too long. End this agony, please. Majority of the members (including me) recommended venting your existing box (and adding another 15"), that may be the cheapest way to solve your problem. Put some tubes in that box, please. Pretty please? Oh, and tell us how it sounds, please.
Small enough for what? If you can't hear the box resonance there's no point in ported boxes . . . hearing the resonance is what we build them for.
You can't hear the pipe (port) resonance if it's high enough above crossover (which it is in almost all subwoofers). But . . . you don't want it in the passband if you can avoid it, which it is (or would be) with the described organ speaker, and you definitely don't want it so low that it's excited by the second or third harmonic of the fundamental. A 120 Hz pipe resonance from the port on an organ woofer is simply stupid . . . evidence of a complete lack of understanding of simple acoustics . . .
Case in point . . . my own home subwoofer. Four 12” drivers (13mm x-max) and two 6” diameter by 12” long ports in about a 50 cu ft box. The box resonance is about 15 Hz., which gives essentially flat response to that frequency at 115dB 1 meter.
Close the ports and, according to the must-be-correct sim, the output at 20Hz drops by 6 dB and at 16Hz it drops by almost 9 dB (plus going well over x-max . . . if I lower the amplifier power to stay below x-max the overall acoustic output is down 12 dB). That’s the difference the box resonance makes. You can definitely hear it. That's why we use ported boxes for subwoofers.
The port resonance, on the other hand, because of the large diameter and short length, is up about 750 Hz. . . . far enough above the 50 Hz. crossover that it doesn’t even remotely matter. Even without the crossover it would have negligible effect on the sound of a 8’ (top of the 32’stop) organ pipe. That’s why only an idiot would use long skinny ports instead of short stout ones . . .
Close the ports and, according to the must-be-correct sim, the output at 20Hz drops by 6 dB and at 16Hz it drops by almost 9 dB (plus going well over x-max . . . if I lower the amplifier power to stay below x-max the overall acoustic output is down 12 dB). That’s the difference the box resonance makes. You can definitely hear it. That's why we use ported boxes for subwoofers.
The port resonance, on the other hand, because of the large diameter and short length, is up about 750 Hz. . . . far enough above the 50 Hz. crossover that it doesn’t even remotely matter. Even without the crossover it would have negligible effect on the sound of a 8’ (top of the 32’stop) organ pipe. That’s why only an idiot would use long skinny ports instead of short stout ones . . .
Last edited:
Parasitic resonances (e.g. 120 Hz) should be small, not the fundamental Helmholtz box resonance Fb.
There are techniques to make the unwanted parasitic resonance small. For example, to couple the wall of the port to a small Helmholtz lossy resonator. I have tried that, works like a charm. Also, crossover frequency can be 80 Hz, not 177 Hz. With 24 dB slope active crossover, there will be additional attenuation at 120 Hz.
But first, we should have to establish if the existing loudspeaker box would be loud enough with the port mounted in it. It is almost free to try it (Bach On, please...). Than, and only than, we can discuss how to kill the unwanted resonances - with passive radiators instead of tube port, or something else.
Last edited:
Right. Quite a blunder for a newbie to confuse the instantaneous empirical peaks in music for the peak-to-peak parameter of a sine wave.
Odd that amps - and organs too - have a headroom dimension. That dimension characterizes how loud they can go briefly versus how loud they can go continuously. With speakers the only thing similar is a kind of burn-out power handling parameter. Or that often irrelevant coil-heating resistance change.
We ordinarily assume that when you run out of cone travel, you done-for. That is, there is no headroom. A more dynamic characterization would be more like derating charts for transistors where, in the case of speakers, you are losing fidelity at increasing loudness. But, up to the point where the cardboard bursts into flames or the spider splits, it is always a trade-off.
It is the latter kind of derating that would be meaningful for a speaker intended in this thread.
Ben
Last edited:
JAG suggested a different woofer. He seems to feel a driver with a longer cone excursion would be better. I've looked at it and it looks reasonable to me.
But not nearly everyone recommended a second driver in the box.
My thinking was that the box is oversized for the existing 15 driver - with or without a port. It's around 13 cu. ft. - even taking in consideration the rear of the driver and the internal support structure. I believe the area of the port would also need to be deducted from the internal cubic footage.
Most feel I'll need more ported boxes to produce sound sufficient for the space that will sound balanced with the rest of the frequencies being produced.
I'll get the port done ASAP. But as I've said, a different driver (or more drivers) will create a need for a different length in the port. And besides, Uncle Sam got all my extra cash on April 15th to spend on such experiments.
I request patience.
Bach On
Perhaps you are aware that in the case of almost all ported 2 way designs the port's pipe resonances are right smack in the middle of the woofer's passband - in fact right in the middle of the midrange, right where our hearing is most sensitive. And it's not a problem if the response curve is smooth.
You don't hear pipe resonances if you can't seen them in a graph. It's as simple as that.
Long skinny ports won't give the same tuning as a short stout port, so I guess the idiot with the long skinny port will have the desired 16 hz tuning while the smart guy with the short stout port will be tuned a couple of octaves higher but at least he won't have to worry about pipe resonance problems that don't exist.
He was talking about simulators, therefore he was talking about sine waves only.
I would definitely also suggest a different woofer, if it's in the budget. I also strongly stand by the vented recommendation. If multiple woofers are also in the budget, 2 or 3 would be better. Also to note, according to Hornresp, adding in more drivers also helps tame the predicted (if nearly absent in reality) port harmonics.
The primary reason I'm attached to this recommendation and am willing to defend it, with measurements if need be, is that this is almost identical to the subwoofers I'm currently using. My enclosures are about 13 cubic feet, with 1 15" Dayton Ultimax woofer, mounted in almost exactly the same orientation that yours is. The only notable difference between the subwoofer you have and the ones I have and enjoy on a daily basis is mine are painted black, and my driver is much, much more capable. Oh, and they have a slot port that's tuned to 14 hz.
Sure, some harmonic aberrations will appear in the box, but the truth is, if you want to stay away from any type of tuned enclosure, you'll have a much harder time getting down below 20 hz with a reasonable sized cabinet and driver, and once you realize the sims won't be anything like reality when it comes to port resonance, you'd quickly find yourself throwing 5 times as much money at it in order to combat a lion sized problem that's really just a kitten under the magnifying glass of software which isn't armed with enough information to make a 100% accurate prediction, I.E wood is wood, not the "immovable object" that simulators like to pretend cabinets are made of.
Here's a picture of one of my subs. Note that it's nearly identical to what you have, except it has a decent driver and a slot port, and I can tell you from actual, real experience that it does not have the port resonance indicated in the sim, but rather an insignificant bump that is far enough out of it's passband to be completely ignored.
Port it.
Attachments
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.