Inspired by an argument on the tapped horn thread, I threw together a design for a subwoofer that reaches down to 15hz. This is serious bass folks - and you can build it in an hour.
It's a tapped horn tuned to 18hz. The driver that I used is a Diyma 12, which costs about $122 at www.diymobileaudio.com. The biggest challenge with this design is that the Diyma only works in a very VERY small box! In fact, the box is so small that the woofer doesn't even FIT! So I took a novel approach - the "tapped horn" bolts to baffle, but the horn has a diameter that's SMALLER than the woofer itself! In fact the "tapped horn" is simply sixteen feet of PVC tubing, with a 180 degree bend in the midpoint. You have two choices - you can use four lengths of 3" PVC tubing. Or you can use a single length of 6" tubing. Take your pick. The 3" PVC is much easier to find.
First, here's a diagram of the enclosure:
Here's the diagram above in PDF
Second, here's the modeled response:
One thing you'll notice about this design is that the efficiency is VERY low for a horn. This is due to the fact that the Diyma has an efficiency of just 84db. But there's an upside - this is easily the smallest 15hz subwoofer you're going to find. Remember that horns, like all subs, can be made smaller if you use a driver with a VERY low VAS. And the Diyma has a incredibly small VAS. In fact, if you try to increase the volume of the enclosure it will become "peaky." So it's ideal to keep this tapped horn on the small side, and feed it lots of power.
Here's the AkaBak model I used.
System 'S1'
Def_Const
{
Xwidth = 13.5e-2;
Xthroat = 13.5e-2;
Xmouth = 13.51e-2;
Lhorn = 488e-2;
Ldriver = 26e-2;
}
Def_Driver 'Diyma12'
dD=25.2cm |Piston
fs=25Hz Mms=327g Qms=4.75
Bl=22Tm Re=3.6ohm Le=1.3mH ExpoLe=0.618
Driver 'D1' Node=2=0=100=101 Def='Diyma12'
Duct 'Dr' Node=101=102 WD={Xwidth} HD={Xthroat} Len={Ldriver}
Duct 'Dd' Node=100=104 dD={Ldriver} Len=3cm
Duct 'Df1' Node=103=104 WD={Xwidth} HD={Xmouth} Len={Ldriver/2}
Duct 'Df2' Node=104=105 WD={Xwidth} HD={Xmouth} Len={Ldriver/2}
Waveguide 'W1' Node=102=103 WTh={Xwidth} HTh={Xthroat} WMo={Xwidth} HMo={Xmouth}
Vf=50mm3 Len={Lhorn-Ldriver-Ldriver} Conical
Radiator 'Rad1' Node=105 Def='Df2' WD={Xwidth} HD={Xmouth}
It's a tapped horn tuned to 18hz. The driver that I used is a Diyma 12, which costs about $122 at www.diymobileaudio.com. The biggest challenge with this design is that the Diyma only works in a very VERY small box! In fact, the box is so small that the woofer doesn't even FIT! So I took a novel approach - the "tapped horn" bolts to baffle, but the horn has a diameter that's SMALLER than the woofer itself! In fact the "tapped horn" is simply sixteen feet of PVC tubing, with a 180 degree bend in the midpoint. You have two choices - you can use four lengths of 3" PVC tubing. Or you can use a single length of 6" tubing. Take your pick. The 3" PVC is much easier to find.
First, here's a diagram of the enclosure:
An externally hosted image should be here but it was not working when we last tested it.
Here's the diagram above in PDF
Second, here's the modeled response:
An externally hosted image should be here but it was not working when we last tested it.
One thing you'll notice about this design is that the efficiency is VERY low for a horn. This is due to the fact that the Diyma has an efficiency of just 84db. But there's an upside - this is easily the smallest 15hz subwoofer you're going to find. Remember that horns, like all subs, can be made smaller if you use a driver with a VERY low VAS. And the Diyma has a incredibly small VAS. In fact, if you try to increase the volume of the enclosure it will become "peaky." So it's ideal to keep this tapped horn on the small side, and feed it lots of power.
Here's the AkaBak model I used.
System 'S1'
Def_Const
{
Xwidth = 13.5e-2;
Xthroat = 13.5e-2;
Xmouth = 13.51e-2;
Lhorn = 488e-2;
Ldriver = 26e-2;
}
Def_Driver 'Diyma12'
dD=25.2cm |Piston
fs=25Hz Mms=327g Qms=4.75
Bl=22Tm Re=3.6ohm Le=1.3mH ExpoLe=0.618
Driver 'D1' Node=2=0=100=101 Def='Diyma12'
Duct 'Dr' Node=101=102 WD={Xwidth} HD={Xthroat} Len={Ldriver}
Duct 'Dd' Node=100=104 dD={Ldriver} Len=3cm
Duct 'Df1' Node=103=104 WD={Xwidth} HD={Xmouth} Len={Ldriver/2}
Duct 'Df2' Node=104=105 WD={Xwidth} HD={Xmouth} Len={Ldriver/2}
Waveguide 'W1' Node=102=103 WTh={Xwidth} HTh={Xthroat} WMo={Xwidth} HMo={Xmouth}
Vf=50mm3 Len={Lhorn-Ldriver-Ldriver} Conical
Radiator 'Rad1' Node=105 Def='Df2' WD={Xwidth} HD={Xmouth}
Glowbug said:Very cool design.
I was extremely disappointed with the DIYMA 12" that I got to test, the cone flexed way too easily and actually dented due to normal use
But it sounded alright.
was that for the subwoofer shootout? what enclosure did you have it in? i've found that this driver is quite sensitive to enclosure volume increases (beyond optimal), moreso than others. i'm curious because i think it sounds quite good.
It was part of this: http://bigbassman.realmofexcursion.com
I was doing most of the subjective SQ reviews.
It was in a little over 1cF sealed, IIRC.
I was doing most of the subjective SQ reviews.
It was in a little over 1cF sealed, IIRC.
Good work!
it would be interesting if you could run some sims with either a conical or exponential horn to compare to the pipe. Wondering how much gain a horn would provide over the pipe and if it is worth the added effort to build.
In terms of shear bulk Vs bass extension I think that you may have set some kind of record so kudos to you mate! Oi Oi Oi!!! Regards.
it would be interesting if you could run some sims with either a conical or exponential horn to compare to the pipe. Wondering how much gain a horn would provide over the pipe and if it is worth the added effort to build.
In terms of shear bulk Vs bass extension I think that you may have set some kind of record so kudos to you mate! Oi Oi Oi!!! Regards.
Re: Good work!
Oh definitely, I tried that!
Here's what I found:
If you increase the volume on the tapped horn, the response begins to get "peaky." Danley mentioned that the tap smooths out the response. So this seems to be key - it actually works BETTER in a smaller volume.
Increasing the flare raises the F3. For example, the design I posted has a flare of one-to-one. Which means that the throat is the same volume as the mouth. If you increase the ration from 1to1 to 4to1, the overall SPL goes up... but the F3 goes higher too. Personally, I'd rather have a low F3, and it's easier to build with a straight flare too!
Last but not least, you can make it even smaller by reducing the length of the pipe. The response stays pretty flat... But the F3 goes up. The reason why I chose a length of eight feet is that it's JUST tall enough to fit in a house. Any longer and you'd need to raise the roof!
moray james said:
Oh definitely, I tried that!
Here's what I found:
If you increase the volume on the tapped horn, the response begins to get "peaky." Danley mentioned that the tap smooths out the response. So this seems to be key - it actually works BETTER in a smaller volume.
Increasing the flare raises the F3. For example, the design I posted has a flare of one-to-one. Which means that the throat is the same volume as the mouth. If you increase the ration from 1to1 to 4to1, the overall SPL goes up... but the F3 goes higher too. Personally, I'd rather have a low F3, and it's easier to build with a straight flare too!
Last but not least, you can make it even smaller by reducing the length of the pipe. The response stays pretty flat... But the F3 goes up. The reason why I chose a length of eight feet is that it's JUST tall enough to fit in a house. Any longer and you'd need to raise the roof!
Glowbug said:
yea. the first enclosure i listened to it was just around there and it was overly bottom heavy. this sub really needs a small box. i wasnt at all impressed with those reviews tho, some numbers and worthless graphs with hardly any substantial comments. pretty much only useful for those interested in which sub gets louder in the arbitrarily chosen enclosures (close for most woofers but shortchanged woofers like the diyma imo, another huge flaw with the test imo). then again, i'm just too used to klippel results followed by extensive reviews explaining the listening sessions.
454Casull said:
Take your pic:
Use one length of 6" tube that's a total of sixteen feet long. The 180 degree bend will burn up about a foot, so use TWO pieces of 6" x 78" pipe.
::OR::
Use FOUR lengths of 3" tube that's a total of sixteen feet long. The 180 degree bend will burn up a bit less than a foot, so use EIGHT pieces of 3" x 78" pipe (approximately.) Four piece of 3" PVC tubing have an IDENTICAL volume as one piece of 6" tubing.
If the total length is off by half a foot or so, it's no big deal.
The general idea is cheap, easy and fast.
FWIW here's how AJ-horn sees that pipe with DIYMA12 - AJ should be able also to move the driver down as with Bose wave cannon
Diyma12
attempt to see offset - GM ~said AJ won't do it - of course with 12" this won't work but could be fun for other cases if AJ really models this type (?)
"if" driver could be offset 1/3 the way in pipe
http://img205.imageshack.us/img205/7231/diyma12offsetco4.gif
Diyma12
An externally hosted image should be here but it was not working when we last tested it.
attempt to see offset - GM ~said AJ won't do it - of course with 12" this won't work but could be fun for other cases if AJ really models this type (?)
"if" driver could be offset 1/3 the way in pipe
http://img205.imageshack.us/img205/7231/diyma12offsetco4.gif
and for tradeoffs here's a guess at Fitzmaurice's Table Tuba with MCM 8" vs DIYMA in "PB" long pipe - TT has about twice the air volume of the pipe and is front-load.
(graph in post above had dl set to 6" so sim had some fill of dips)
http://img228.imageshack.us/img228/1340/diyma12pvttmcmxq3.gif
(graph in post above had dl set to 6" so sim had some fill of dips)
http://img228.imageshack.us/img228/1340/diyma12pvttmcmxq3.gif
I'm not sure how this qualifies as a tapped horn. I thought the definition of a tapped horn was a horn where the front AND backside of the driver tap into the horn at different locations. I don't see anything connected to the backside of the driver.
What I see is (in simplest form) a six inch diameter pipe about 16 feet long connected to the front of the driver. Isn't that a regular end loaded tl?
Sorry if I'm missing something here, I must not be understanding something.
What I see is (in simplest form) a six inch diameter pipe about 16 feet long connected to the front of the driver. Isn't that a regular end loaded tl?
Sorry if I'm missing something here, I must not be understanding something.
some of Karlson's stuff "kinda" had a tap but not TH
box from 1955 patent
box from 1955 patent
An externally hosted image should be here but it was not working when we last tested it.
just a guy said:
I see a handful of differences between a tapped horn and a transmission line.
1. In a tapped horn the designer is placing the mouth of the horn in a specific location to smooth out the overall response. In a transmission line, designers rarely consider how the response is affected by the location of the mouth. You'll notice that TLs often have the mouth in random location, such as the front of the enclosure, the back, the floor, the top, etc... In a TH, the mouth location is critical.
2. Most TLs uses a reverse taper - the mouth is smaller than the throat. At this point, every TH I've seen use the same taper as a conical horn - the mouth is bigger than the throat. My design DOES violate this rule, as I found that a straight pipe plays lower (and is easier to build.)
3. I agree that there are tapped horns where the woofer is placed midway down the horn; however it's IMPOSSIBLE with the Diyma. With the Diyma the VAS is SO SMALL, you can't even FIT it into tapped horn. I tried throwing it into a TH intended for a LAB12, and the response was far too peaky. This is because the LAB12 has a higher VAS.
Based on my modeling, the best response was in a TH whose cross section couldn't even contain the basket. (The basket is 7.5" deep.)
So that's why I had to use such an unconventional enclosure - it was the only practical way to do it.
kappa546 said:
If we had a Klippel analyzer laying around in one of our basements, we'd have used it
As for the only being interested in what woofer gets louder, well, that's a function of the forums we did the tests for. Most of the users who looked at those results were of high school or college age and only cared about how hard things "bump". *shrugs*
(I can't get taut bass from Klipschorn - partly flexing-resonant walls, etc but even harpsichord and acoustic guitar sound weird giving impression of "undertones" - plus is spatially disconnected - muddy on bottom, schrill in middle and a bit harsh & forwards on top)
bowed bass, cello, some drumkit should be good tests of subwoofer and its integration - in a small room one Karlson w. 416 re-tuned ~exemplar and boosted was pretty good for me w. theatre organ.
what can be done for moderate-cheap cost to make a cohesive and articulate system with reasonably low disptortion at 120+ dB peaks/1M?
bowed bass, cello, some drumkit should be good tests of subwoofer and its integration - in a small room one Karlson w. 416 re-tuned ~exemplar and boosted was pretty good for me w. theatre organ.
what can be done for moderate-cheap cost to make a cohesive and articulate system with reasonably low disptortion at 120+ dB peaks/1M?
Patrick Bateman said:
Greets!
1, 2. True, but based on my understanding of what constitutes a TL and a tapped horn tells me you did an end loaded TL, not a tapped horn or even a tapped pipe. FYI, the sim I did in MathCad that looked virtually identical to yours on the other thread is so close because I used MJK's earlier TL WS that for programming simplicity plots the vent and driver output at the same point in space.
3. Actually, if I understand what's going on in an offset tapped horn I believe this driver might be better suited than most, but I haven't seen TD post certain aspects of its design that would explain why, so I'll leave it to ya'll's inventiveness.
GM
Interesting looking little device Patrick.
A couple of comments about it, The baffel on the front is going to need to have a deeeep cut out or a couple of laminated layers to keep to give the surround enough room to xceed Xmax if need be. That will be an added chamber that you may want to add to the aKaBak Model.
In your model does it account for the distance of the rear of the driver in free air from the tube exits.
The pipes are probably going to whip around like crazzzzy without much bracing.
I thought I braced my Tapped horn proto type well, its still got some buzzing and creaking, and I don't think my compression ration is as high at the Throat as the PVC special.
Will there be much turbulance from all that preasure in such a small pipe? will it effect over all sound quality?
Build it and you will know for sure.
Did you model it in a TH intended for a lab? I posted an aKaBak File with lab12 on the driver discriptor but but the TS parameters were changed to the KOVE. I found the Lab12 to be not usefull for the TH I was building and ended up using a Parts Express Buy out Kove Audio subwoofer.
Antone-
A couple of comments about it, The baffel on the front is going to need to have a deeeep cut out or a couple of laminated layers to keep to give the surround enough room to xceed Xmax if need be. That will be an added chamber that you may want to add to the aKaBak Model.
In your model does it account for the distance of the rear of the driver in free air from the tube exits.
The pipes are probably going to whip around like crazzzzy without much bracing.
I thought I braced my Tapped horn proto type well, its still got some buzzing and creaking, and I don't think my compression ration is as high at the Throat as the PVC special.
Will there be much turbulance from all that preasure in such a small pipe? will it effect over all sound quality?
Build it and you will know for sure.
Did you model it in a TH intended for a lab? I posted an aKaBak File with lab12 on the driver discriptor but but the TS parameters were changed to the KOVE. I found the Lab12 to be not usefull for the TH I was building and ended up using a Parts Express Buy out Kove Audio subwoofer.
Antone-
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.