What was so hard about building your tapped horn? Mine didn't require any mitering and almost all of the angles were at 90 degrees. The one angle that wasnt at 90 was within 5 degrees or so of 90 degrees. The most time consuming part was drawing out the design on both sides of the side pannels to make it easier to line up the screws etc when assembling it.
I'm a big believer in using multiple subs, so the goals of my tapped horn were:
#1 - smallest size humanly possible
#2 - since i'm using a bunch, total spl weren't big requirements
#3 - all of the tapped horns on the big thread require a fair amount of wood working, so I *tried* to design one that didn't
The end result is this thread. Without a doubt, the achilles heel of this design is that it's tall and narrow, which makes it exceptionally difficult to work with. For example, I had to saw the enclosure into quarters just so that I could seal the internal partition, then re-assemble the whole thing.
Again, I'm not trashing the tapped horn, just saying it's not ideal for a home sub IF you agree that multiple subs is de rigeur.
#1 - smallest size humanly possible
#2 - since i'm using a bunch, total spl weren't big requirements
#3 - all of the tapped horns on the big thread require a fair amount of wood working, so I *tried* to design one that didn't
The end result is this thread. Without a doubt, the achilles heel of this design is that it's tall and narrow, which makes it exceptionally difficult to work with. For example, I had to saw the enclosure into quarters just so that I could seal the internal partition, then re-assemble the whole thing.
Again, I'm not trashing the tapped horn, just saying it's not ideal for a home sub IF you agree that multiple subs is de rigeur.
Were it only based upon what I agreed.....
In my case, the architectural review committee has clearly indicated that there are to be a minimum of speakers in the room, and that they are to be as small as possible.
I have approval for 6 (and only 6) enclosures - FR, C, FL, RR, RL, S - and even this is pushing it. Due to the layout of the room, I have limited floor space to put them in. Turns out that in my case - your Achilles' heel is my godsend. A slim tower that makes bass is a much easier enclosure to integrate into my room.
Add to that my desire to experiment and play, throw in my woodworking abilities, and a tapped horn is definitely a viable alternative for me.
In my case, the architectural review committee has clearly indicated that there are to be a minimum of speakers in the room, and that they are to be as small as possible.
I have approval for 6 (and only 6) enclosures - FR, C, FL, RR, RL, S - and even this is pushing it. Due to the layout of the room, I have limited floor space to put them in. Turns out that in my case - your Achilles' heel is my godsend. A slim tower that makes bass is a much easier enclosure to integrate into my room.
Add to that my desire to experiment and play, throw in my woodworking abilities, and a tapped horn is definitely a viable alternative for me.
Patrick Bateman said:
Dont judge efficiency based on listenings.
The ear has a lot more complicated way to caculate loudness than a simple spl meter. Time and frequencycontent is very important to the ear.
http://www.tcelectronic.com/media/skovenborg_2004_loudness_m.pdf
Oh I agree completely. The fact that the tapped horn is playing noticeably deeper AND we're less sensitive to deep bass might imply the tapped horn is in fact louder.
But the bottom line is that the bandpass is half the size, easier to build, and easier to hide.
I'd love to see someone try a tapped horn with a five inch or a six inch driver. You can get a STUPID amount of low bass out of a tapped horn. Even the eight inch woofer I used was overkill to some degree.
If I could come up with a tapped horn that was one half the size, I'd be all over it. The one in this thread is four cubic feet, and that's quite a lot to hide when you need three or four.
But the bottom line is that the bandpass is half the size, easier to build, and easier to hide.
I'd love to see someone try a tapped horn with a five inch or a six inch driver. You can get a STUPID amount of low bass out of a tapped horn. Even the eight inch woofer I used was overkill to some degree.
If I could come up with a tapped horn that was one half the size, I'd be all over it. The one in this thread is four cubic feet, and that's quite a lot to hide when you need three or four.
Patrick Bateman said:
I've been measuring and simulating some smaller woofers too. I have come up with several ideas that use inexpensive 6.5 inch drivers. Depending on the driver, you can come up with a tapped horn cabinet that is less than one cubic feet in volume and will play to 45 Hz with authority. As with most of the small drivers, they are limited in the amount of power they can take due to excursion limits.
The Tang Band 6.5" Neo sub driver is usable in a tapped horn as well, and has a lot of excursion. Unfortunately, it is not exactly cheap.
IIRC Volvotreter (sp?) has done some work with them, and has posted several designs and favorable comments on his website.
The 6.5 I am playing with is the Dayton PA165 driver, which is very surprising in a TH - flat from 50 to 160 Hz, in about a cubic foot or so. In other words - tough to even fit the driver into the box once it is folded up.
Once I come up with a folding pattern to build one, I'll knock together a test enclosure, though at the rate my fun projects are progressing don't hold your breath waiting.
The 6.5 I am playing with is the Dayton PA165 driver, which is very surprising in a TH - flat from 50 to 160 Hz, in about a cubic foot or so. In other words - tough to even fit the driver into the box once it is folded up.
Once I come up with a folding pattern to build one, I'll knock together a test enclosure, though at the rate my fun projects are progressing don't hold your breath waiting.
I'm not even sure that I know what a "tapped horn" is but here is what I do know:
I've studied LF sound sources my whole life and there are no free lunches. Nothing has ever turned out to be an effective improvement in LF design except the Acoustic Lever and its just plain impractical.
Horns at LF are a joke. There is no such thing as a LF horn, unless its the size of a house. Smaller than that and its just a port and ports have been studied to death; series, multiple, transmission lines, its all a bunch of hand waving because when the wavelengths are ten feet long details simply don't matter.
A horn - at best - has an impedance transformation of the square root of the inlet and outlet areas. A lever goes as the square of these areas. Thus for a 2:1 output to input area ratio the lever is twice as effective as a horn. So much for horns.
I'm sorry, but I'll stick to my belief that subs are simple sources and closed boxs work great. Or if you want to get fancy, use a bandpass to lower the HF distortion and increase the efficiency (a little), but anything beyond that is simple "blowing in the wind".
I've studied LF sound sources my whole life and there are no free lunches. Nothing has ever turned out to be an effective improvement in LF design except the Acoustic Lever and its just plain impractical.
Horns at LF are a joke. There is no such thing as a LF horn, unless its the size of a house. Smaller than that and its just a port and ports have been studied to death; series, multiple, transmission lines, its all a bunch of hand waving because when the wavelengths are ten feet long details simply don't matter.
A horn - at best - has an impedance transformation of the square root of the inlet and outlet areas. A lever goes as the square of these areas. Thus for a 2:1 output to input area ratio the lever is twice as effective as a horn. So much for horns.
I'm sorry, but I'll stick to my belief that subs are simple sources and closed boxs work great. Or if you want to get fancy, use a bandpass to lower the HF distortion and increase the efficiency (a little), but anything beyond that is simple "blowing in the wind".
gedlee
After your post: I have been reading the material on your web site.
RE No LF horns: For what frequency ranges do you consider horns as viable?
I found your use of open cell foam in upper frequency horns quite interesting, and wondered about the lowest usable frequency.
Also enlightening was the response letter on the recent Horn Theory articles
Thanks
After your post: I have been reading the material on your web site.
RE No LF horns: For what frequency ranges do you consider horns as viable?
I found your use of open cell foam in upper frequency horns quite interesting, and wondered about the lowest usable frequency.
Also enlightening was the response letter on the recent Horn Theory articles
Thanks
I recently did something that I should have done a long time ago. I used the OS contour in Websters Equation in order to determine the accuracy of the later. The OS in my approach is exact, but Webster is never exact - but how bad is it? Turns out its REALLY bad. The traditional idea that the rate of change of the area has to be small for Webster to be accurate is optimistic. It turns out that the rate of change has to be small compared to a wavelength!! This means that ALL predictions of Websters Equation are completely wrong at LFs.
I return to what I said - basicly all horns behave as simple ports at LFs. Ports are not that effective at increasing LF sound as they really can have gain or amplify. The Lever can and does actually amplify. But short of using a lever one may as well just use a single port or no port as nothing else does much good.
How low does a horn work as a horn? Based on what I am seeing in my work, about twice cutoff. Below that the air mass is simply a lump of moving air that has no notion what its boundaries look like.
It also turns out that the parameter theta / a , where "a" is the throat radius, plays a virtually identical role to "m" for an exponential horn - it sets the ficticious cutoff frequency. The OS has a large peak at cutoff while the exponential go to zero. Both solutions are imaginary below cutoff, which is erroneous.
I return to what I said - basicly all horns behave as simple ports at LFs. Ports are not that effective at increasing LF sound as they really can have gain or amplify. The Lever can and does actually amplify. But short of using a lever one may as well just use a single port or no port as nothing else does much good.
How low does a horn work as a horn? Based on what I am seeing in my work, about twice cutoff. Below that the air mass is simply a lump of moving air that has no notion what its boundaries look like.
It also turns out that the parameter theta / a , where "a" is the throat radius, plays a virtually identical role to "m" for an exponential horn - it sets the ficticious cutoff frequency. The OS has a large peak at cutoff while the exponential go to zero. Both solutions are imaginary below cutoff, which is erroneous.
The tapped horn is certainly efficient, but it's not because it's a horn. It's efficient because Danley has come up with a clever method to delay the rear wave so that it's mostly in phase with the front. But it's still not a horn - it's closer to a transmission line really. I agree that the front and rear paths are operating as glorified ports, not as true horns.
So how does it differ from a transmission line since these have not ever been shown to be all that effective.
Unless the delay if frequency dependent, it will comb filter, gaining and cutting the output periodically. Over the lowest narrow band where it is gaining, its virtually identical to a dual bandpass design, which can achieve some impressive efficiencies, but only over some pretty narrow bandwidths. If you stager these plus and minuses - so that they interleave - then you have the Bose wave radio or the 1/3 - 2/3 tuning of the prior art.
Really in LF sources its pretty much all been done, but people keep making small changes that have no real effect but then they claim a "revolution". I'm not saying that the TH is like this, I really don't know, but I would be surprised if it is anything more than a new way of doing the same old stuff.
Unless the delay if frequency dependent, it will comb filter, gaining and cutting the output periodically. Over the lowest narrow band where it is gaining, its virtually identical to a dual bandpass design, which can achieve some impressive efficiencies, but only over some pretty narrow bandwidths. If you stager these plus and minuses - so that they interleave - then you have the Bose wave radio or the 1/3 - 2/3 tuning of the prior art.
Really in LF sources its pretty much all been done, but people keep making small changes that have no real effect but then they claim a "revolution". I'm not saying that the TH is like this, I really don't know, but I would be surprised if it is anything more than a new way of doing the same old stuff.
It's quite similar to the Bose Wave cannon. I'd say the most dramatic improvement is the ability to use computer software to model the location of the mouth in two dimensions, and factor in the delay also.
When you look at the simulations, the location of the ports makes quite a dramatic difference in the response.
All of the bandpass programs I've seen until 2006 or so didn't include these factors. Martin King's Mathcad models from quarter-wave.com were the first. Ironically, he doesn't even claim to model bandpass enclosures, but you can model them using his "front loaded horn" simulation.
McBeans horn response is a lot easier to use, the wizards are just genius.
When you look at the simulations, the location of the ports makes quite a dramatic difference in the response.
All of the bandpass programs I've seen until 2006 or so didn't include these factors. Martin King's Mathcad models from quarter-wave.com were the first. Ironically, he doesn't even claim to model bandpass enclosures, but you can model them using his "front loaded horn" simulation.
McBeans horn response is a lot easier to use, the wizards are just genius.
gedlee said:
I hate to say it, but there's a big fat fundamental flaw to all the tapped horn simulations that everyone's been doing here.
When I built a bandpass and a tapped horn, I was surprised to find that the bandpass seemed to be louder. The simulations indicated the TH would be dramatically louder.
Well here's the prob. Everyone is doing their simulations as if the tapped horn is perfectly corner loaded. To see if my hunch was correct, I used horn response to model a bandpass subwoofer, then modeled the *exact* same bandpass subwoofer in bandpass boxmodel (which has proven to be quite accurate.)
Sure enough, horn response predicted an SPL level that was DRAMATICALLY higher than the other program.
Once you tell horn response to do it's simulation in "free space", the results are a lot closer to reality.
Just to do my due dilligence, I modeled a tapped horn with the same driver, this time in free space. It confirmed exactly what I'm hearing. The tapped horn wasn't more efficient than a bandpass, but it played about half an octave deeper, in a box that was marginally larger.
Again, I'm not trashing tapped horns, they're a clever invention. But if people think they're going to get an extra 12DB of output with the same driver, they are DREAMING.
Here's a schematic of a front loaded horn modeled in horn response, which is actually a bandpass subwoofer. Though a bandpass sub is not a horn, horn response will happily model it. As Geddes noted, most sub horns aren't horns either, they're just deviations on dual and single reflex bandpass boxes. In this schematic you will note a front chamber, a rear chamber, and a port. It's a single reflex bandpass.
Attachments
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.